Problem Gambling services in NZ: from experience to ... · Problem Gambling Foundation of New Zealand (including Asian Services) Oasis Centres, New Zealand Salvation Army Gambling

Confidential to the Ministry of Health

Final Report Problem Gambling Services

In New Zealand: From Experience to Effectiveness

November 2008 (Draft report was submitted on 21st December 2007)

Report Submitted by: Associate Professor Samson Tse Principal Investigator

Centre for Gambling Studies Social & Community Health School of Population Health University of Auckland Email: [email protected] Phone: (64 9) 3737-599 ext 86538

Ms Megan Putterill Business Manager – Health Auckland UniServices Ltd Email: [email protected]

Report Prepared for: Ms Barbara Phillips Problem Gambling

Mental Health Directorate Ministry of Health Wellington

XGAMBLE Report

“XGAMBLE” The logo of “XGAMBLE” was used to promote the present project in the community. The “X” represents most of the research participants’ desire to reduce or stop gambling; the four population groups (New Zealand Europeans, Maori, Pacific peoples and Asians involved in this project and finally the sign also represents the weaving of qualitative and quantitative data collected during the course of the study.

Conflict of interest statement All members of the research team involved in the present project and production of the Final Report are employed by universities or organisations that they are affiliated with.

They declare no conflict of interests to this research project.

Disclaimer This Report summarises qualitative and quantitative data collected between 2006 and 2007. The report reflects the view and experiences of the interviewed people who gamble, people with problem gambling, individuals affected by problem gambling and professionals working in the field of problem gambling. Members of the research team have taken all care to accurately capture and interpret the views of participants while maintaining their privacy and confidentiality. We also would like to highlight the fact that the phone intervention used in Phase Two of this research is not Gambling Helpline New Zealand’s model of care but has been developed by gambling and telephone experts based on Phase One findings.

Reports from Auckland UniServices Limited should only be used for the purposes for which they were commissioned. If it is proposed to use a report prepared by Auckland UniServices Limited for a different purpose or in a different context from that intended at the time of commissioning the work, then UniServices should be consulted to verify whether the report is being correctly interpreted. In particular it is requested that, where quoted, conclusions given in UniServices reports should be stated in full.

This report should be referenced as follows: Tse, S., Campbell, L., Rossen, F., Jackson, A., Shepherd, R., Dyall, L., Perese, L. &. Jull,

A. (2008). Problem gambling services in New Zealand: From experience to effectiveness. Centre for Gambling Studies, School of Population Health. Prepared for Ministry of Health, Problem Gambling. Auckland UniServices Limited, University of Auckland, 214pp.

2

XGAMBLE Report

FOREWORD

Gambling is fast becoming a major public health issue in New Zealand, and around the world. To deal with this, research is needed to a guide policy and treatment and to assist cultural and community responsiveness.

The Centre for Gambling Studies in the School of Population Health, University of Auckland, was established to respond to the recognised need for quality and independent research to assist in minimising harm from gambling. The Centre has conducted research in a broad range of areas regarding gambling including public health, health promotion, host responsibility, social and economic impacts and workforce development.

This research is the first systematic study of intervention effectiveness undertaken in New Zealand. Its purpose was to gain a better understanding of the range of treatment approaches utilised in New Zealand and whom they are effective for. While there are, inevitably, some limitations on what conclusions can be drawn, it is important to note the study suggested it is possible to achieve constructive change in attitudes and beliefs with concomitant reductions in harmful gambling behaviours.

The specific recommendations are based on the research team’s experiences and reflections and the considerable learning gained from conducting this project. Valuable feedback was gathered while the Project Team met and worked with key stakeholders throughout the project.

The challenge for the future is to determine whether the improvements achieved can be sustained over time and whether a realistic and effective research design can be employed to ascertain the cause-effect relationships between interventions and outcome measures.

In the meantime, this research has provided a basis for development of appropriate interventions to respond to the emerging challenges of gambling in the unique context and settings of Aotearoa, New Zealand.

Professor Alistair Woodward, Head of School of Population Health, Faculty of medical and Health Sciences, University of Auckland

3

XGAMBLE Report

CONTRIBUTORS Associate Professor Samson Tse (Principal Investigator, Centre for Gambling Studies, University of Auckland; key responsibilities in this project were to: oversee the design and implementation of the project, monitor the progresses and be responsible for the Asian stream of the present study)

Ms Lisa Campbell (Project Manager, Centre for Gambling Studies, University of Auckland; key responsibilities in this project were to: implement the project and ensure smooth running of the studies)

Dr Fiona Rossen (Co-Director, Centre for Gambling Studies, University of Auckland; key responsibilities in this project were to assist in data analysis, synthesising and editing the final reports)

Professor Alun Jackson (School of Social Work, University of Melbourne; key responsibilities in this project were to: provide inputs to the design of the project, in particular Phase One and be responsible for the data analysis of the Clinical Task Analysis)

Dr Robin Shepherd (Researcher, Centre for Gambling Studies, University of Auckland; key responsibilities in this project were to be responsible for the European stream of the present study, in particular, the data collection, analyses and writing of relevant sections of the midway and final reports)

Dr Lorna Dyall (Maori Health Studies, University of Auckland; key responsibilities in this project were to be responsible for the Maori stream of the present study, in particular, liaising with the Maori community, the data collection, analyses and writing of relevant sections of the midway and final reports)

Ms Lana Perese (Pacific Researcher, PhD Candidate; key responsibilities in this project were to be responsible for the Pacific Island stream of the present study, in particular, liaising with the Pacific Island communities, the data collection, analyses and writing of relevant sections of the midway and final reports)

Dr Andrew Jull (Clinical Trial Research Unit, University of Auckland; key responsibilities in this project were to: advise on the design of the clinical trial in Phase Two and the data analyses, and serve as the liaison person between the research team and Clinical Trial Research Unit)

4

XGAMBLE Report

ACKNOWLEDGEMENTS On behalf of the Project Team, I would like to offer our sincere gratitude to the study participants for their immense courage and generosity in giving their time to share their experience, wisdom and show their support to the project. This study would not have been completed without their support. Our wish is that their voices are being captured in this work, which in turn helps provide better support and therapeutic interventions for individuals and families affected by problem gambling. Lastly I extend our appreciation to the reviewer(s) for their comments on the early draft of the report and Mr Dean Adam from the Ministry of Health in collating the comments.

We wish to express our sincere appreciation to the following individuals and organisations for their support and work in this project at various stages of the project planning and implementation:

Phase One and Two, Ethnic Researchers Robin Shepherd, Shona Jones, Luisa Ape-Esera, and Yanbing Li - thanks for engaging with service providers to access participants, conducting the Phase One Interviews, and assisting with the data analysis and write up of these results.

Robin Shepherd, Judy-Anne Wanakore, Seini Taufa, Luisa Ape-Esera, Lorna Dyall, Ruth Herd, Robyn Stewart, Hagyun Kim, Yanbing Li, Louise Manaena, Alice McMillan, Eugene Song, and Iris Zhang all gave their valuable time to go out into their local and ethnic communities to recruit the Phase Two participants.

Thanks to Josephine Jackson (Tupu) and Pam McQuoid (Lifeline) in providing their knowledge and experience in training the call centre staff.

Technical Advisory Group Thanks to all that have contributed to this group: Phil Townsend, John Stansfield, Cynthia Orme, Kawshi De Silva, Lynette Hutson, Paul Clifford, Brent Diack, Bruce Levi, Josephine Gray, John Wong, Wenli Zhang, Lorna Dyall, Monica Stockdale, Puti Lancaster, Lana Perese, Andrew Jull, Robert Brown, Vicki Berkahn.

Members of the manual writing Group Josephine Jackson, Laurie Siegel-Woodward, Phil Townshend, David Coom, Puti Lancaster, Pam McQuiod, Samson Tse all contributed to the development of the manual.

Service providers Problem Gambling Foundation of New Zealand (including Asian Services) Oasis Centres, New Zealand Salvation Army Gambling Helpline New Zealand Te Rangihaeata Oranga (Maori Problem Gambling Services) Pacific Mental Health, Alcohol, Drugs and Gambling Services – Tupu Team, Te Kahui Hauora o Ngati Koate Trust Pacificare Trust

5

XGAMBLE Report

Without the support and input from service providers this project would not have progressed.

Gambling Industry Jonathan Gee (Clubs New Zealand) kindly assisted in distributing the project promotional material and he did not exert any influence on either the design or the implementation of the project.

Community and cultural groups NIU Development – especially Pefi Kingi and Mua’autofia Clarke, Citizen Advices Bureaus, Budgeting Centres, Age Concern, The Aucklander, Fairfax Media (Suburban Newspapers), NZ Herald, Hastings Leader/Napier Courier, The Nelson Mail, Christchurch Star, Korea Review, Older & Bolder, Radio 1ZB, More FM, Flava FM, Canterbury Television

Relevant government departments and offices National and Regional Probation Centres (particularly the Auckland Office)

Auckland UniServices Kate O’Connor, Manu Keung, Shireen Nanayakkara-McDonald, Billie Harbidge, Megan Putterill Clinical Trial Research Unit – Yannan Jiang, Stephen Vanderhoorn, Sheila Fisher, Amanda Milne, Rina Prasad, John Faatui, Michelle Jenkins, Barry Gray, Clark Mills, Michael Ng and Terry Holloway Survey Research Unit – Maria Butler, Teik-Aun Cheah, Kim Meredith

Ministry of Health: Problem Gambling, Mental Health Section and Public Health Intelligence Shayne Nahu, Jason Landon, Paul MacLennan, Anne Beach, Barry Borman, Barbara Philips, Dean Adam

Faculty of Medical & Health Sciences, University of Auckland Peter Adams (Director, Centre for Gambling Studies), Emma Timewell (Communications Adviser). Val Grey (Graphics Designer)

6

XGAMBLE Report

TABLE OF CONTENTS

FOREWORD ........................................................................................................3

CONTRIBUTORS.................................................................................................4

ACKNOWLEDGEMENTS ....................................................................................5

TABLE OF CONTENTS .......................................................................................7

LIST OF TABLES...............................................................................................10

LIST OF FIGURES .............................................................................................12

LIST OF FIGURES .............................................................................................12

EXECUTIVE SUMMARY ....................................................................................13

CHAPTER 1 INTRODUCTION .......................................................................24

1.1 Brief background of the study ............................................................................................. 24

1.2 Literature Review and Context............................................................................................. 26 1.2.1 Interventions and Evidence Available ........................................................................ 26 1.2.2 Telephone and Face-to-Face Counselling ................................................................. 33 1.2.3 Methodological issues related to interventions studies for problem gambling ........... 34 1.2.4 Conclusion.................................................................................................................. 37

CHAPTER 2 PHASE ONE STUDY.................................................................38

2.1 Research aims ................................................................................................................ 38

2.2 Methodology ................................................................................................................... 38 2.2.1 Recruitment process................................................................................................... 39 2.2.2 Participants: Sample size and distribution.................................................................. 41 2.2.3 Data collection instruments/semi-structured interview guidelines.............................. 41 2.2.4 Data analysis .............................................................................................................. 44 2.2.5 Research rigour .......................................................................................................... 46

2.3 Results............................................................................................................................. 47 2.3.1 European experiences of utilising problem gambling services .................................. 49 2.3.2 Maori experiences of utilising problem gambling services......................................... 55 2.3.3 Pacific (Samoan, Tongan, Niuean) experiences of utilising problem gambling

services....................................................................................................................... 61 2.3.4 Asian (Chinese and Korean) experiences of utilising problem gambling

services....................................................................................................................... 66

7

XGAMBLE Report

2.3.5 European practitioners’ experiences of providing problem gambling services .......... 75 2.3.6 Maori practitioners’ experiences of providing problem gambling services................. 81 2.3.7 Pacific Island (Niue, Samoan, Tongan) practitioners’ experiences of providing

problem gambling services......................................................................................... 86 2.3.8 Asian (Chinese and Korean) practitioners’ experiences of providing problem

gambling services....................................................................................................... 91 2.3.9 Results from the Clinical Tasks Analysis (Problem Gambling) (CTA) (PG)............... 96

2.4 Phase One Discussion................................................................................................. 104 2.4.1 Client experiences of problem gambling services.................................................... 104 2.4.2 Practitioners experiences of providing problem gambling services ......................... 108 2.4.3 Summary: Clients and practitioners experience of problem gambling services ...... 110 2.4.4 CTA discussion on comparison between New Zealand and Australian

practitioners .............................................................................................................. 111 2.4.4 CTA discussion on comparison between New Zealand and Australian

practitioners .............................................................................................................. 112

CHAPTER 3 PHASE TWO STUDY..............................................................118

3.1 Research aims .............................................................................................................. 118

3.2 Methodology ................................................................................................................. 118 3.2.1 Inclusion/exclusion criteria........................................................................................ 118 3.2.2 Production of Manual ............................................................................................... 119 3.2.3 Schedule of intervention and measures ................................................................... 121 3.2.4 Study Design ............................................................................................................ 121 3.2.5 Recruitment strategies.............................................................................................. 125 3.2.6 Project infrastructure ................................................................................................126 3.2.7 Data collection/ measures ........................................................................................ 128 3.2.8 Follow up and retention ............................................................................................ 132 3.2.9 Data analysis and statistical methods ...................................................................... 132

3.3 Results........................................................................................................................... 134 3.3.1 Evaluation of intervention effects ............................................................................. 134 3.3.2 Participants’ responses to open-ended questions in Form A and B ........................ 163

3.4 Discussion..................................................................................................................... 173 3.4.1 Intervention effectiveness and measures ................................................................ 173 3.4.2 Effectiveness of recruitment strategies .................................................................... 180 3.4.3 Effectiveness of follow up methods .......................................................................... 183 3.4.4 Learning from participants’ experiences in participating in the trial.......................... 184 3.4.5 Reflections on working with multiple stakeholders................................................... 185

CHAPTER 4 SUMMARY AND CONCLUSIONS ..........................................191

4.1 Achievements ............................................................................................................... 191

4.2 Strengths of the study ................................................................................................. 194

4.3 Limitations of the study ............................................................................................... 196

4.3 Implications................................................................................................................... 199

4.4 Recommendations for future studies........................................................................ 200

8

XGAMBLE Report

4.4.1 Contextual issues ..................................................................................................... 200 4.4.2 Specific methodological considerations ................................................................... 201 4.4.3 Priorities for future studies........................................................................................ 203

4.5 Dissemination plan....................................................................................................... 203

CHAPTER 5 REFERENCES ........................................................................205

9

XGAMBLE Report

LIST OF TABLES

Table 1: Research rigour in Phase One study ....................................................46 Table 2: Demographic and background information for service users who completed Phase One semi-structured interviews..............................................48 Table 3: Demographic and background details for European service users who completed Phase One semi-structure interviews................................................49 Table 4: Demographic and background details for Maori service users who completed Phase One semi-structure interviews................................................55 Table 5: Demographic and background details for Pacific service users who completed Phase One semi-structure interviews................................................61 Table 6: Demographic and background details for Asian service users who completed Phase One semi-structure interviews................................................66 Table 7: Frequency of performing counselling tasks and importance attached to those tasks as per the CTA (PG) (n= 27)............................................................96 Table 8: Means, Reliabilities, Standard Error, Corrected Item-Total Correlations of ‘How Often’ Subscales..................................................................................114 Table 9: Variance and Eigen Values of Frequency Subscales for NZ and Australia............................................................................................................115 Table 10: Correlation of Subscales Relating to ‘HOW OFTEN’ in New Zealand ..........................................................................................................................116 Table 11: Correlation Statistics – New Zealand and Australian FREQUENCY Data ..................................................................................................................116 Table 12: Minimum documentation to be completed by practitioners subsequent to the third counselling session.........................................................................131 Table 13: Distribution of participants’ age and DSM-IV scores (n=96)..............135 Table 14: Distribution of participants scoring eight or more on the DSM-IV according to ethnicity ........................................................................................135 Table 15: Distribution of participants’ location according to intervention modality (n=96) ...............................................................................................................135 Table 16: Distribution of participants according to preferred ethnicity (n= 96) ..136 Table 17: Distribution of participants current work situation according to intervention modality (n= 96).............................................................................136 Table 18: Distribution of participants’ employment status by preferred ethnicity (n=96) ...............................................................................................................137 Table 19: Distribution of participants’ highest educational qualifications according to intervention modality (n=96)..........................................................................137 Table 20: Distribution of participants’ marital status according to intervention modality (n=96) .................................................................................................138 Table 21: Source of participants’ initial knowledge of the study by treatment modality (n=96) .................................................................................................139 Table 22: Distribution of participants’ initial knowledge of the study according to preferred ethnic group (n=96) ...........................................................................139

10

XGAMBLE Report

Table 23: Distribution of participants receiving treatment for alcohol and/or drug and mental health issues ..................................................................................140 Table 24: Distribution of participants receiving treatment for mental health issues across five groups of preferred ethnicity (n=96)................................................140 Table 25: Distribution of participants receiving problem gambling counselling (n=96) ...............................................................................................................141 Table 26: Distribution of participants receiving problem gambling counselling across five groups of preferred ethnicity (n=96)................................................141 Table 27: Distribution of participants’ annual household income according to intervention modality (n=96)..............................................................................141 Table 28: Distribution of participants’ annual household income across five groups of preferred ethnicity (n=96)..................................................................142 Table 29: Profile of self-excluded or banned participants .................................142 Table 30: Participants’ primary problem gambling mode by intervention modality (n=96) ...............................................................................................................143 Table 31: Distribution of participants primary problem gambling mode across five groups of preferred ethnicity (n=96)..................................................................143 Table 32: Distribution of money and proportion of income spent gambling and gender according to gambling mode (n=96) .....................................................144 Table 33: Descriptive summary of primary and secondary measures - baseline (n=92) ...............................................................................................................147 Table 34: Descriptive summary of primary and secondary measures - post intervention (n=27)............................................................................................149 Table 35: Descriptive summary of primary and secondary measures - follow up (n=12) ...............................................................................................................150 Table 36: Sense of control of gambling at baseline and post-intervention by intervention modality .........................................................................................151 Table 37: Anticipated effectiveness of intervention at baseline measure by intervention modality .........................................................................................152 Table 38: Experienced effectiveness of intervention at post-intervention measure by intervention modality ....................................................................................152 Table 39: Level of satisfaction with provided intervention service by intervention modality ............................................................................................................153 Table 40: Accessibility to services by intervention modality..............................153 Table 41: Number of intervention sessions attended ........................................154 Table 42: Number of intervention sessions attended according to anticipated effectiveness of intervention .............................................................................154 Table 43: Correlations among primary and secondary outcome measures ......154 Table 44: Problems (and outcomes perceived by practitioners) and intervention techniques used................................................................................................155 Table 45: Period effects between baseline and post-intervention visits by outcome measure .............................................................................................159 Table 46: Why people seek help for problem gambling (Hodgins et al., 2002) .175 Table 47: Summary of the activities and outcomes of the project .....................191 Table 48: Compare those who adhered to post-intervention visits and those who left the study .....................................................................................................197

11

XGAMBLE Report

LIST OF FIGURES

Figure 1: Problem areas taking up most of the practitioners’ time - European....78 Figure 2: Problem areas taking up most of the practitioners’ time - Maori ..........83 Figure 3: Problem areas taking up most of the practitioners’ time - Pacific.........88 Figure 4: Problem areas taking up most of the practitioners’ time - Asian ..........93 Figure 5: Diagrammatic representation of key factors that contribute to the outcomes of problem gambling interventions....................................................111 Figure 6: Flowchart depicting Phase Two study design ....................................122 Figure 7: Flowchart recording number of participants recruited in Phase Two study .................................................................................................................123 Figure 8: Total money spent on gambling at baseline, post-intervention, and follow up by intervention modality .....................................................................156 Figure 9: Total time (hour) spent on gambling at baseline, post-intervention, and follow up by intervention modality .....................................................................157 Figure 10: Proportion of total money spent on gambling at baseline, post-intervention, and follow up by intervention modality..........................................157 Figure 11: Gambling Attitudes and Beliefs (GABS) score at baseline, post-intervention, and follow up by intervention modality..........................................158 Figure 12: Gambling Readiness To Change (GRTC) score at baseline, post-intervention, and follow up by intervention modality..........................................158 Figure 13: Total time (hour) spent on gambling, model-adjusted means and their 95% confidence intervals from the regression for each group at each visit by intervention modality .........................................................................................160 Figure 14: GAB scores, model-adjusted means and their 95% confidence

..........................................................................................................................160 intervals from the regression for each group at each visit by intervention modality

12

XGAMBLE Report

EXECUTIVE SUMMARY

Background This research was commissioned by the Ministry of Health in March 2006 to provide a better understanding of the range of treatment approaches that are utilised in New Zealand and whom they are effective for.

In New Zealand, problem gambling has been declared a significant public health issue, a stance which is reflected in the Gambling Act. As such, the Ministry of Health has developed an overall gambling strategy for public health, prevention, treatment and research. The Ministry of Health funds a range of problem gambling services that cover the continuum of gambling related harm across the population. The Ministry has identified the need for a range of intervention services to be available in secondary and tertiary prevention areas and to be appropriate for all population groups. These services aim to work with people affected by gambling problems (including problem gamblers, family/whanau and significant others) to identify and manage problems, thus reducing gambling related harm. Currently in New Zealand, the Ministry purchases both face-to-face counselling services and a national telephone helpline. Intervention services provide screening and early interventions in primary care settings, assessment, brief interventions, full psycho-therapeutic interventions, follow up, and maintenance/support.

‘Objective Four’ of the Ministry of Health’s six year strategic plan, states it will “maintain and develop accessible, responsive and effective interventions”. Actions to achieve this objective include implementing a range of responsive services, identifying and validating screening tools, developing outcome measurement tools and brief interventions, developing culturally responsive services for Maori, Pacific, Asian and young people, and to develop a knowledge base around effective interventions.

The research detailed in this report, is the first systematic study of intervention effectiveness undertaken in New Zealand. It will provide the beginnings of the evidence and knowledge base for maintaining and developing successful and appropriate interventions within the emerging field of problem gambling in the unique cultural context and settings of Aotearoa, New Zealand.

Aims and Methods The research detailed in this report was multi-method in nature and entailed the two research phases described below. Throughout the course of this research, a specialist Technical Advisory Group (TAG) was convened and consulted. The TAG provided vital expertise and guidance for the project and assisted in areas such as intervention services input, knowledge, recruitment of practitioners and clientele, and data interpretation.

Phase One Study Phase One of this research sought to identify and analyse a range of problem gambling intervention services and approaches utilised in New Zealand across the four main population groups: Pakeha, Maori, Pacific and Asian. The specific aims were:

13

XGAMBLE Report

1. Identify different problem gambling intervention services as representative of the range of services in New Zealand (including services located outside Auckland);

2. Analyse the range of tasks and therapeutic techniques undertaken by practitioners in problem gambling services;

3. Identify important features of these problem gambling intervention services, as perceived by clients (people who gambling, and their family/whanau); and,

4. Explore differences in terms of perceived effectiveness, between clients who attend one session only, and those who attend three or more sessions.

This phase provided stand alone qualitative information on problem gambling intervention services in New Zealand, as well as informing and enhancing the methodology of Phase Two and developing an intervention manual for practitioners to ensure consistency of approach within the clinical randomised study.

In order to provide a comprehensive analysis and overview of problem gambling treatment services, a number of data collection methods were employed:

1. Administration of a standardised questionnaire to problem gambling service practitioners;

2. Interviews with problem gambling service practitioners; and, 3. Interviews with problem gambling service clientele (both family/whanau and

individuals with problem gambling who had used the services in the past).

Problem gambling service practitioners and clientele were recruited through the problem gambling intervention services involved in this study.

Semi structured interviews were conducted with both the service users and practitioners. Interviews involving Maori, Pacific, Asian and New Zealand European service users and practitioners were conducted with ethnic specific interviewers to ensure cultural safety.

Semi-structured interview schedules were developed based upon the reviewed literature and the interview group schedules which have been previously developed, piloted and employed by Jackson, Thomas and Blaszczynski (2003) in Victoria, Australia.

Individual interviews focused on items such as: • Why do people gamble? • Why seek help in the first place? • What defines a successful intervention? • What helps/hinders successful interventions?

Practitioner interviews focused on similar items such as: • What is your organisations purpose? • What therapeutic approaches/techniques do you utilise? • What do you consider is meant by a ‘successful’ intervention(s)? • How do you currently measure success in your service?

14

XGAMBLE Report

• What do you consider to be the key elements of ‘successful’ methods of intervention for people with gambling problems?

Qualitative data in the form of detailed information provided by practitioners and clients were formed into categories and analysed thematically.

The CTA (PG) questionnaire gave practitioners the opportunity to document the practice and theories they use when dealing with a problem gambler, a member of the problem gambler’s family and the community at large. It aims to provide a broad overview of the complexity of the counsellor’s role, specify the range of tasks they perform, and document the relationship between the frequency of task performance and the counsellor’s beliefs about the importance of the tasks performed.

Descriptive analysis was used to create a profile of the types of activities performed by practitioners and the frequency and level of importance allocated to each of those activities/tasks identified. The analysis focused on identifying trends and the distribution of the collected data set (e.g. the clustering of the items). The New Zealand CTA results were also compared to the earlier study conducted in Australia. As there were only 27 practitioners involved in this survey, no further analytical or inferential analyses were performed.

Phase Two Study The pilot clinical trial phase of this research aimed to investigate the effectiveness of an intervention model delivered across two different intervention modalities: face-to-face counselling and interventions over the phone. Standard measurement instruments and the intervention manual (based on the actual practices in New Zealand identified from Phase One findings) were utilised.

As this was the first time that an investigation of the effectiveness of problem gambling interventions in New Zealand was conducted, it was proposed that Phase Two of this project be kept at a small scale (pilot/vanguard study), thus enabling it to inform the conduction of a full-scale trial. This research therefore aimed to obtain only general overall trends (due to the limited sample sizes) and exploratory information across ethnic subgroups. This research generated hypotheses about differences across ethnic groups, but was not able to draw definite conclusions for the observed differences.

The specific aims of this research were: 1. To investigate the effectiveness of these interventions in controlling gambling

behaviours and reducing harm caused by problem gambling. 2. To

investigate the effectiveness of these interventions, with regard to the magnitude of effect.

3. To provide preliminary information on the relative effectiveness of the interventions for ethnic groups (NZ European, Maori, Pacific and Asian).

4. To develop and recommend processes for future studies to extend research in the

15

XGAMBLE Report

field of problem gambling.

The overall design of Phase Two is known as a ‘pragmatic trial’ which aims to capture and compare the delivery of intervention services in the field. In comparison, an ‘explanatory trial’ tends to have very tight, laboratory-type control of various treatment conditions and its treatment effects are often diluted in the real world.

Schedule of intervention and measures • Requested informed consent, after introducing the research and ascertaining that the

potential participant had met the “participants eligibility”; • Collected demographic and baseline data; • Randomly allocated the participant to one of the two groups (face-to-face or

telephone modality) and an appropriate practitioner; • The practitioner contacted the participant and commenced the intervention (lasting up

to 6 sessions); • Following termination of the intervention, the Project Team were notified by the

practitioner; and, • Outcome assessments were undertaken at the termination of interventions and at the

subsequent follow up, six weeks after termination.

Face-to-face and Telephone services Intervention services were provided according to the Interventions Manual, which outlined the assessments, major intervention approaches and activities/tasks to be performed by practitioners. These were identical for both face-to-face and telephone interventions.

The purpose of the manual was to ensure consistency of approach in providing problem gambling interventions within the randomised trial, rather than addressing competence issues. Practitioners would follow their own agency policy including procedures dealing with emergency situations, cultural safety, supervision and accountability. The manual was to ensure there was a consistent, basic therapeutic outline/framework to the problem gambling interventions.

There was no restriction on the frequency or duration of counselling sessions - that is, this trial evaluated the delivery of intervention services in a naturalistic setting. Similarly, if they were deemed to be appropriate, the services could offer any culturally-specific interventions. The intensive part of the intervention lasted up to six sessions. Practitioners were asked to complete a form documenting the major elements and/or activities of the services provided for each client involved in this study. Six weeks after the intervention concluded, participant follow ups were completed. Outcome measures at the end of the intervention and six-week follow ups were collected by the call centre researchers.

Please note that the telephone counselling employed in this study does not represent Gambling Helpline’s model of care. In this study, telephone counselling involves therapeutic work between the same practitioner and his/her client over a period of time and series of scheduled sessions.

16

XGAMBLE Report

Assessments (as performed by research assistants, not practitioners involved in this study) • Demographic information: age, sex, ethnicity, income level, job type, preferred

gambling activities; • Health and wellbeing; • Money (proportional to income) spent gambling; • Time spent gambling; • Gambling Attitudes and Beliefs Survey (GABS) (Breen & Zuckerman, 1999); • Gambling Readiness To Change scale (GRTC) (Neighbors, Lostutter, Larimer &

Takushi, 2002); and, • Overall ratings on acceptability, satisfaction and accessibility of intervention as well

as subjective ratings of the interventions’ success and support (Riley & Jean-Mary, 2004).

Data analysis and statistical methods Data was analysed following a pre-specified analysis plan.

Demographic information (age, gender, ethnicity, region of residence, occupation, education, income level, Alcohol and Drug issues, mental health concerns, etc.), along with relevant assessments of their gambling behaviours (DSM-IV score, gambling activity, self-excluded or banned from entering gambling venue) were summarised for participants in each of the intervention arms. Summaries of continuous baseline variables with normal distribution were presented as means and standard deviations, or medians and inter-quartiles for skewed data, while categorical variables were described as frequencies and percentages.

Summary statistics were provided for each group at each visit on the primary endpoints measured from all eligible participants. Analysis of the primary endpoints was carried out using appropriate repeated measurement analyses, assuming that non-observed data were missing at random. If the data were essentially normally distributed, the Mixed model was used to compare the treatment effects between the two intervention groups at baseline, post-intervention and end of 6-week follow up, while adjusting for their ethnicities and other stratification factors (for randomisation) in the regression model. Other potential covariates such as whether or not they had received prior counselling and their health histories were adjusted in the regression model, if they were statistically significant. Preliminary estimates of the intervention effect, together with standard errors, were calculated in order to establish what clinically relevant differences would be desirable to detect in a full-scale (nation wide) trial.

Similar repeated measurement analyses were carried out for the secondary endpoints using the collected information. Descriptive information was provided for additional variables of interest (sense of control; post-intervention measurements). Results were compared between treatment groups using standard statistical techniques.

Sensitivity analysis was also carried out on the two primary endpoints.

17

XGAMBLE Report

Key Findings

Literature Review The aims of the brief literature review were twofold. Firstly, it highlighted the most commonly used interventions for individuals with problem gambling and the evidences available. Instead of reproducing what has previously been covered in other comprehensive reviews, this brief review focused on how the design of the present study was informed by the existing literature. Secondly, the review summarised the key methodological issues related to intervention outcome studies in the gambling field.

The review confirmed that there are currently a wide range of interventions available worldwide for individuals with problem gambling. These range from public health and health promotion programmes aimed at early detection and screening, to acute and crisis interventions, relapse prevention and rehabilitation programmes. The focus of the review was on therapeutic and counselling interventions for individuals or small groups (usually the family members/whanau). Types of interventions include cognitive and cognitive-behavioural therapy, motivational interviewing, self help programmes, pharmacological treatments, natural recovery and combinations of all the above. Recent research indicates that a combination of interventions work best, especially those incorporating elements of cognitive-behavioural strategies.

The literature revealed that future research designs need to produce systematic, credible data on intervention effectiveness and should include both qualitative research and quantitative (including the use of standard measurement instruments for reliability and validity purposes) to provide more information and richer results. It also became clear that New Zealand is a unique setting (e.g. pokie machines are not only placed in big gambling venues but also in pubs and clubs, restaurants) with diverse cultures that require different approaches. It is important that research methodologies be modified to suit the cultural context. It is envisioned that the results from this proposed research will go someway to providing more comprehensive information of gambling interventions in New Zealand.

Phase One Study In total, 58 service users (39 problem gambling clients and 19 family members/Whanau; including 14 European, 19 Maori, 11 Pacific and 14 Asian) and 27 practitioners (including 10 European, 7 Maori, 6 Pacific and 4 Asian) participated in the Phase One interviews.

Participants’ understanding of “When gambling becomes problematic?” centred on a number of themes. Examples cited by all population groups include: losing control, losing more than one can afford, some personality changes, and chasing losses. Maori and Pacific also identified “negative impacts on family, dysfunctional family” as additional signs that gambling has become problematic.

18

XGAMBLE Report

People identified a number of reasons why they sought help. These included the following: • Lost too much; • Pressure from family to seek help; and, • Being sick of lying to family and friends.

Pacific and Maori participants’ also mentioned that they sought help after seeing the problem gambling services in the media. Some Asian participants said they sought help because they were directed to under police order as part of self-exclusion procedures.

For those who sought help, the majority found the assistance received from problem gambling services very helpful. The following outcomes were identified: • Control or cessation of gambling; • Gaining helpful and specific information about gambling and problem gambling –

e.g. Why did I get into such trouble? Why did I lose control? What are the triggers? How can I help myself?;

• Dealing with other relevant issues such as having a better marriage, improved inter-personal relationships, dealing with unresolved bereavement or past abuse, building up confidence, setting life goals, and finding meaning in life;

• Obtaining a better understanding of why family members are affected by gambling - removes the blaming, shamefulness, and secrecy;

• Feeling that they are being listened to, understood and not being judged; • Is referred on for specialised services e.g. financial management; and, • Feel comfortable and respected; culturally appropriate approaches or methods are

being used. Family members also identified the following: • Having someone to turn to for the first time (outside the family), without relying

on friends and relatives who have a limited understanding of what problem gambling is; and,

• They understand better how to support and prevent their family members relapse.

Most of the participants (the past users of problem gambling services) felt their needs were being met; they appreciated the fact that the services were free, confidential, sensitive to their cultural needs, had flexible working hours, and that they could contact the services by phone or see practitioners face-to-face.

Practitioners identified the following themes for what contributes to successful interventions: • Client/practitioner relationship; • Support networks; • Client centeredness; • Client empowerment; and, • Client understanding of themselves and their triggers.

Pacific and Maori practitioners also mentioned the importance of family involvement and having compassionate, directive practitioners who incorporate culture into an intervention plan.

19

XGAMBLE Report

Phase Two Study In total, 103 people registered for the Phase Two study. Ninety six of these were eligible and were subsequently randomised to a treatment group (47 face-to-face and 49 telephone). The retention of participants throughout the intervention was a challenge to this study:

- 92 participants completed baseline data (46 for both face-to-face and telephone); - 27 participants completed post-intervention data (14 for face-to-face and 13 for

telephone); and, - 12 participants completed six-week follow up data (2 for face-to-face and 10 for

telephone).

Pertinent demographic data relating to participants include: - Mean age of participants was 43 years, with only one participant aged under 20; - Mean DSM-IV score was 7.4 (indicative of pathological gambling). The mean

DSM-IV score was higher for non-European ethnicities; - 67% of participants were female; - 45% of participants were non-European; - A broad range of education levels and employment status were recorded; and, - 54% of participants were not married or in a partner relationship.

Main findings from this phase include: - The observed primary data for both telephone and face-to-face intervention modes

had a significant reduction in the amount of dollars spent and time spent gambling between the baseline data and the post intervention groups;

o Cautions: � Without a control group, the changes are not attributable to

positive intervention effect – a full scale trial with control group is required.

� Rather big drop out rate (baseline n=92; post-intervention n=27; or a retention rate of 29.35%; the six-week follow up analysis was not performed due to the high drop out rate).

- No significant differences were observed between the telephone and face-to-face intervention modes, but observed data showed useful response trends among participants randomly allocated to telephone vs face-to-face contacts;

- Gambling Attitude and Beliefs Survey (GABS) scores supported the hypothesis that changes in attitude to gambling are associated with changes in behaviour;

- No statistically significant relationship was observed between Gambling Readiness To Change (GRTC) scores and measures of time/money spent gambling; and,

- Qualitative data indicated that reasons for successful intervention included 1) readiness to change, 2) information, knowledge, and skills imparted, 3) warmth and acceptance as perceived by client, and 4) social support available to participants.

20

XGAMBLE Report

Implications and Recommendations It is very important to emphasise again that the findings and the implications from this project have to be interpreted within the constraints of the present project. It is flawed or even dangerous to make any substantial changes to services based on the present study. What would be useful though is to begin exploring what the findings mean to the following areas: namely service delivery and research workforce development in the problem gambling field.

Service delivery • There are strong parallels between the Phase Two study and establishing a ‘new

problem gambling service’. It took the research team at least one year to recruit 111 people onto the register for the clinical trial. It is evident to the research team, that newly established services, both general and culturally specific, do take time (perhaps between one and half and two years) to establish their name and credibility in the community.

• It was evident in both the Phase One and Two studies that the quality of the therapeutic relationship between the practitioners and clients is critical. In particular, it is very important for practitioners to be empathetic, non-judgemental and respectful. It is paramount, both at the level of service engagement and at the level of encounters between practitioners and clients, that this kind of warmth and engaging atmosphere is conveyed clearly to the service users, regardless of the seriousness of the problems or of the clients’ socio-cultural, linguistic background.

• The client-service engagement issues (e.g. time taken, and in some cases the extra effort required – such as ethnic researchers accompanying prospective participants when enrolling in the study) revealed in this project, suggest that more resources are needed to ensure that new clients are comfortable and confident about the problem gambling services.

• One clear message from this project was the value of giving clients options or different modes of problem gambling services. The observed data and the trend of findings provide emerging evidence to support the use of both ‘face-to-face’ and ‘telephone’ based interventions for problem gambling. It is also important to note that this study only recruited people who were affected by pokie machines in casinos, pubs/bars and clubs. It is highly likely that different populations (age and gender groups, people affected by different forms of gambling activities) and interacting with individuals’ stage of gambling problems, might benefit from the option of alternative forms of services. For example, texting on mobile phones, accessing web-based counselling, group (vs. individual), attending couple counselling, working through a self-paced workbook, attending a residential programme (versus community), and accessing culturally specific and language based (versus generic) treatment facilities.

• When comparing the responses from the New Zealand practitioners with their Australian counterparts on the Counsellors Tasks Analysis, it was apparent that the New Zealand practitioners tended to adopt an ‘integrated approach’ in their work; that is, they tend to combine counselling together with health promotion and community work. This seems to be a distinctive feature of the New Zealand problem gambling services.

21

XGAMBLE Report

Research Workforce • The majority of the agencies and individual practitioners were immensely supportive

of the present research project. In some cases their inability to help was more related to the substantial time pressures or work loads they were experiencing, particularly for those who worked only part time in problem gambling services and who were involved in multiple research projects at the same time. It is advisable to: 1) continue to promote the value and importance of research and/or evidence-informed practice in the sector; 2) allow counselling or clinical practitioners time and acknowledgement for their involvement in research projects; 3) support staff exchange programme whereby practitioners are seconded to a gambling research centre, or university-based researchers work at a problem gambling service provider for a defined period of time; and, 4) support practitioners who are interested in pursuing research as their career, to undertake further training as they are in the ideal position to bridge the divide between practice and principles or theories.

Recommendations for future studies

Contextual issues The following contextual factors will need to be taken into consideration when planning further problem gambling intervention studies in New Zealand. They are based on both the research team’s reflections or learning gained from conducting this project and specific comments gathered from key stakeholders towards the end of the project.

1. Coordination of research projects; 2. Sufficiency of allocated time for research; 3. Partnerships with providers/practitioners; 4. Partnerships with ethnic/population groups; and, 5. Partnerships with members of the gambling industries.

Specific methodological considerations The following methodological issues have been identified as requiring consideration in future research of this type.

1. Use of a control group; 2. Managing and minimising drop outs; 3. Selecting appropriate and sensitive outcome measure(s); 4. Description of the interventions; 5. Training of practitioners; 6. Recording number of intervention sessions; and, 7. Assessing intervention effectiveness.

Priorities for future studies The following have been identified as priorities for future study: • Evaluate outcomes of different modes of problem gambling service deliveries:

individual, small groups, workbook, web-based, phone contacts, text-based mobile phone, self-help group, couple and family therapies.

22

XGAMBLE Report

• Study practitioners’ characteristics; future research should attend to this potentially important determinant of gambling outcomes.

• Improve understanding of the mechanism of intervention efficacy; the mechanism of action of various techniques may generate further hypotheses regarding the aetiology of the condition and may lead to further improvement of therapies; determine which elements have had specific efficacy, only behavioural treatments have been tested sufficiently to be judged as “effective”.

• Compare different types of interventions - simple support or specific treatment techniques e.g. cognitive, cognitive-behavioural, or motivational enhancement techniques.

• Improve the matching between interventions provided by services and the needs and presentation patterns of potential clients.

• Study factors that contribute to treatment failures and relapse in clients with problem gambling.

• Study various sub-groups of the population: o Youth; o Older adults; o Women; o Families affected by problem gambling; o Indigenous/native people; o Ethno-cultural minorities; o Refugees; o Patients with comorbid mental disorders; o Inmates; and, o Specific high-risk occupational groups (e.g. food industry, shift workers).

23

XGAMBLE Report

CHAPTER 1 INTRODUCTION

This research was commissioned by the Ministry of Health in March 2006 to provide a better understanding of the range of treatment approaches that are utilised in New Zealand and whom they are effective for.

This report is the final report required by the National Problem Gambling Team, Mental Health Directorate, Ministry of Health, in accordance with the contract specifications (Contract 414953/303496/01).

The overall research consisted of two phases: Phase One was a qualitative study and Phase Two was a pilot/vanguard randomised clinical trial.

Phase One of the research sought to identify the range of services provided and the theoretical approaches used by problem gambling intervention services in New Zealand. It also aimed to understand what practitioners, service users and families considered as helpful and less helpful in terms of reducing harms caused by problem gambling. Phase Two of the study aimed to investigate the effectiveness of two intervention modalities: face-to-face interventions and phone interventions. Standard measurement instruments and an intervention guideline that was developed from Phase One, and based on actual practices in New Zealand, were utilised.

This final report includes an overview, analysis and interpretation of the data from Phase One interviews and the Phase Two pilot randomised controlled trial. It includes recommendations, and a conclusion chapter that summarises the achievements and issues that have arisen throughout the project. The manual that informed Phase Two is attached in an Accompanying Document.

1.1 Brief background of the study The Gambling Act 2003 was introduced to address concerns relating to the rapid expansion of gambling opportunities which has occurred in New Zealand over the past two to three decades, especially in relation to electronic forms of gambling and ‘pokie machines’ (more formally known as Electronic Gaming Machines (EGMs)). The Act aims to control, regulate and monitor gambling and to prevent and/or minimise gambling related harm.

Problem gambling is estimated to affect approximately 2-3% of New Zealand’s adult population (Abbott & Volberg, 2000) and 4-5% of our adolescent population (Rossen, 2005). It is also known that certain population groups (Maori, Pacific and Asian Peoples, and youth) are disproportionately affected by problem gambling. Problem gambling is recognised as having widespread negative socioeconomic, social and health effects. It has consistently been associated with bankruptcy, crime and incarceration, depression, suicidality, substance use/abuse, breakdown of family units, and disruption to the family

24

XGAMBLE Report

and community of which the problem gambler is a member (Brown & Raeburn, 2001; Katterns, 2006; Rankin & Haigh, 2003).

In New Zealand, problem gambling has been declared a significant public health issue, a stance which is reflected in the Gambling Act. As such, the Ministry of Health has developed an overall gambling strategy for public health, prevention, treatment and research. The Ministry of Health funds a range of problem gambling services that cover the continuum of gambling related harm across the population. The Ministry has identified the need for a range of intervention services to be available primarily in secondary and tertiary prevention areas and to be appropriate for all population groups. These services aim to work with people affected by gambling problems (including problem gamblers, family/whanau and significant others) to identify and manage problems, thus reducing gambling related harm. Currently in New Zealand, the Ministry purchases both face-to-face counselling services and a national telephone helpline. Intervention services provide screening and early interventions in primary care settings, assessment, brief interventions, full psycho-therapeutic interventions, follow up, and maintenance/support.

‘Objective Four’ of the Ministry of Health’s six year strategic plan, states it will “maintain and develop accessible, responsive and effective interventions”. Actions to achieve this objective include implementing a range of responsive services, identifying and validating screening tools, developing outcome measurement tools and brief interventions, developing culturally responsive services for Maori, Pacific, Asian and young people, and to develop a knowledge base around effective interventions.

The research detailed in this report (see Chapter 2. Phase Two Study), is the first systematic study of intervention effectiveness undertaken in New Zealand. It will provide the beginnings of the evidence or knowledge base for maintaining and developing successful and appropriate interventions within the emerging field of problem gambling in the unique context and settings of Aotearoa, New Zealand.

25

XGAMBLE Report

1.2 Literature Review and Context The aims of this brief review are twofold. Firstly, it highlights the most commonly used interventions for individuals with problem gambling and the evidences available. Instead of reproducing what has previously been covered in other comprehensive reviews, this brief review focuses on how the design of the present study was informed by the existing literature. Secondly, this brief review summarises the key methodological issues related to intervention outcome studies in the gambling field. Throughout the review, explicit comments are made to explain how the present study attempts to address the specific issues raised in the literature.

At present, there are a range of interventions available worldwide for individuals with problem gambling. These range from public health and health promotion programmes aimed at early detection and screening, to acute and crisis interventions, relapse prevention and rehabilitation programmes (Brown & Raeburn, 2001). The focus of this review is on therapeutic and counselling interventions for individuals or small groups (usually the family members/ whanau).

1.2.1 Interventions and Evidence Available The Cochrane Library review (Oakley-Browne, Adams & Mobberley, 2001) noted that by the year 2000, 17 gambling outcome studies had been conducted, only four of which were randomised clinical trials (RCTs). RCTs are frequently considered to be the ‘gold standard’ of treatment outcome studies and to provide the strongest form of evidence to support the efficacy of interventions. Korn and Shaffer (2004) used a three-level rating system (strong, moderate and low evidence) to classify intervention approaches by the strength of the scientific evidence that is available to support the use of the methods. In other words, the absence of evidence from a RCT does not mean that other interventions have little or no utility. Rather, the paucity of RCTs means that most of the clinical interventions in problem gambling have not been studied.

This brief literature review employed Korn and Shaffer’s classifications of evidence and framework to highlight major types of problem gambling interventions and the evidence available. The three levels of evidence are: • Strong evidence: outcome studies based on randomised clinical trials, usually had

6 to 12 months follow up period, clear outcome measures and adequate sample size;

• Moderate evidence: interventions and treatment protocols were fully documented and tested within a strong research design that included adequate follow up and carefully measured outcomes; and,

• Weak evidence: reflecting the absence of studies or studies with poor design.

• Korn and Shaffeer (2004, p.18) added another category of ‘evidence’, the so-called ‘Promising and complementary services’ which represent the recently emerging treatments to help client reduce harms caused by problem gambling.

26

XGAMBLE Report

Other relevant references include the work by Hodgins and colleagues (2001), Ledgerwood and Petry (2005), McCown and Howatt (2007), Petry (2006), Toneatto and Ladouceur (2003), Toneatto and Millar (2004).

Interventions: Strong evidence 1. Cognitive and Cognitive-behavioural therapy (CBT) and related interventions

Cognitive models involve identifying clients’ thought distortions in relation to gambling and attempting to modify these thoughts (e.g. client illusion of control). Cognitive therapies involve cognitive correction (e.g. identifying cognitive distortions regarding gambling) and relapse prevention (e.g. restructuring thought patterns to find alternative activities to replace gambling). Sylvian, Ladouceur and Boisvert (1997) studied the effectiveness of Cognitive Correction (CC) on problem gambling. This treatment involved cognitive correction of core erroneous gambling cognitions (irrational beliefs about gambling), training in problem solving and social skills training. This study reports that treatment was effective but could not identify which aspects were effective. A later study by Ladouceur and colleagues (2001) concentrated on CC consisting of four key components. These were: understanding the concept of randomness, understanding the key erroneous beliefs that individuals with problem gambling have about gambling, helping clients become aware of their own inaccurate beliefs about gambling, and correction of erroneous beliefs through discussion and information-giving. This approach was found to be highly effective at one year post-treatment follow up.

In CBT, behavioural components are added to reinforce non-gambling behaviour (Ledgerwood & Petry, 2005). The behavioural approaches include imaginal or cognitive desensitisation which assumes that problem gambling involves a pattern of arousal signalled by an initial stimulus. The treatment involves learning to relax in the presence of several imagined gambling related episodes or situations. Other research into desensitization suggests that the effectiveness of this approach is enhanced by acting it out in a real life situation as well as in the client’s imagination. A similar approach is imaginal relaxation which involves relaxation, but in this case the client learns to relax and remain relaxed in the presence of general relaxing scenes (also known as ‘progressive muscle relaxation technique’). Other approaches have included aversive therapy: the pairing of the compulsion to gamble with electric shocks. McConaghy, Blaszczynski and Frankova (1991) have reported a long term evaluation of these treatments which suggest that imaginal desensitization is more effective than imaginal relaxation alone and both are superior to aversive therapy.

However, there is evidence that integrating behavioural components with cognitive therapy is the most effective treatment for people with problem gambling to date. Ladouceur and colleagues (2002) developed a treatment programme based on cognitive behavioural principles that cover four components: i) correcting cognitive distortions about gambling; ii) developing problem solving skills; iii) teaching social skills and iv) coaching on relapse prevention techniques. Within the problem and social skills components, Korn and Shaffer (2004) included the learning of the following life skills:

27

XGAMBLE Report

assertive skills, goal setting, impulse management, learning the signs and symptoms of problem/pathological gambling, self-monitoring of money and time spent on gambling. In one of the recent reviews on treatment effectiveness of pathological gambling, it is concluded that:

“…interventions that fall within the cognitive-behavioral spectrum, even when delivered via a manual and involving only minimal therapist contact, have the most empirical support. The available scientific data suggest that the cognitive and behavioral therapies are more effective than no treatment; however, it is not possible to determine which specific type of cognitive-behavioral therapy is most effective or whether it is more effective than other approaches to treatment.” (Toneatto & Ladouceur, 2003, p. 291)

On the other hand, Jackson, Thomas and Blaszczynski (2003) pointed out that CBT neglects other areas pertinent to the person’s life such as comorbidity, substance dependence, marital problems, court orders, financial hardship, isolation and loneliness, family problems, relationship, post-immigration adjustment and employment issues.

It is imperative to mention that cognitive behavioural techniques have been clearly referred to as one of the intervention approaches in the Intervention Manual (see XGAMBLE Accompanying Document) which was written specifically for the present study. In addition to CBT, the Manual also covers other commonly used techniques and a range of gambling-related problems (e.g. marital problems, financial hardship, isolation and loneliness) that they might need to deal with.

Echeburua et al. (1996) proposed a treatment approach they call exposure-response prevention. This is primarily a behavioural treatment in which clients are taught to better manage money, avoid gambling situations, and remain in the presence of high risk gambling situations but resist gambling. The cognitive component challenges the illusion of control and memory biases, and facilitates communication skills in a group format. In this evaluation all participants were abstinent at the end of the behavioural treatment and almost 50% remained completely abstinent after a year. However, the group component did not improve the chances of success in this treatment. In a later study these authors (Echeburua et al., 2000) combined in vivo desensitisation with relapse prevention, consisting of teaching clients to identify high risk situations for gambling and develop effective coping strategies. These interventions were then tested in individual and group settings. Once again, the method was shown to be effective but the group component did not increase effectiveness.

Interventions: Moderate evidence 2. Relapse prevention and recovery training

Relapse prevention and recovery training centres on strengthening the person’s ability “to identify and cope with high-risk situations that commonly create problems and precipitate relapse” (Korn & Shaffer, 2004, p.21). The risk situations for gambling include

28

XGAMBLE Report

environmental settings (e.g. gambling venues, pokie machines in restaurants), intrapersonal stress (e.g. anger, depression, boredom), and interpersonal triggers (e.g. relationship difficulties at work, home, financial pressure). In the Intervention Manual written for the present study, situations that clients find themselves in or put themselves in, that increase their likelihood to lapse into destructive behaviour, could be: • people who condone problem gambling; • venues with gambling opportunities; and, • other known stressful situations that trigger dysfunctional behaviour (e.g. work-place,

family arguments).

Relapse prevention seeks to develop coping methods to deal with the identified high-risk situations without relying on maladaptive gambling behaviour. Similarly in the Intervention Manual, it refers to the formulation and use of a “safety plan”. A safety plan should be written with clients for them to retain. They may wish to put copies in high-profile places such as their car or fridge, if appropriate. It should be short and in bullet points. The plan should cover: • what they will do if they find themselves in high risk situations; • what they will do to control thinking errors; • if relapsing on key goals, what to do to get back on track; and, • when and how to use significant others.

The bullet points of the safety plan can be arranged to form a single word or acronym that summarises the safety plan. Clients can be encouraged to put them on cards and carry them with them. They can also be put in visible places.

This set of techniques have been well developed and widely used in the alcohol and drug treatment field (Korn & Shaffer, 2004).

Interventions: Weak evidence 3. Psychodynamic psychotherapy

There is very limited psychodynamic research in the gambling field and very little evidence in the outcome literature to support the effectiveness of its application for individuals with problem gambling.

4. Aversion Therapy

Use of punishment as a form of therapy is rarely used in contemporary addiction treatment centres or problem gambling services.

Ogborne, Wild, Braun and Newton-Taylor (1998) did not support the use confrontational approach and psychodynamic processes.

In Korn and Shaffer’s review (2004) ‘Self-help’ and ‘Gamblers Anonymous’ were included as one single form of intervention under the sub-heading of ‘Self-help: Gamblers Anonymous’ whereas they should be reviewed separately (e.g. see McCown &

29

XGAMBLE Report

Howatt, 2007; Petry, 2006). Furthermore ‘Self-help: Gamblers Anonymous’ and ‘Self-Exclusion’ were put under ‘Weak Evidence’ in Korn and Shaffer’s review. However given the increasing number of publications (some were published peer-reviewed outcome studies), the research team choose to put these interventions in the next category of interventions: Promising and complementary services.

Interventions: Promising and complementary services This category of intervention was used in Korn and Shaffer’s review (2004):

“…it is useful to consider the promise evidenced by empirically supported interventions from related mental health and addiction fields (Chambless & Ollendick, 2001). These applications appear theoretically attractive, are widely used by gambling treatment practitioners.” (p. 18).

5. Self-help interventions

Self-help interventions for individuals with gambling problems, often involve a workbook that is based upon cognitive and behavioural techniques. Other intervention modalities include a home-based, self-help manual derived from a similar package developed for people with alcohol problems (Dickerson, Hinchy & Legg England, 1990). Component ingredients of this manual included training in self-monitoring, functional analysis of gambling behaviour, goal/limit setting, self-reinforcement and maintenance of long term gains. Self-help interventions have been known to be effective in conjunction with motivational interviewing, but not alone.

One example of a self-directed approach to problem gambling is a minimal treatment regime that consisted of telephone contact that supported clients to work through a self help manual (Dickerson, Hinchy & Legg England, 1990). The treatment manual consisted of cognitive/behavioural self-control strategies such as self-monitoring of cravings, triggers and effects of gambling, goal-setting and self-reward for not gambling. The manual-only group improved more than a control group, and the effectiveness of these interventions was improved where clients had practitioner contact. Dickerson’s work again illustrated the importance of studying the effectiveness of different modes of delivery of problem gambling services.

6. Gamblers Anonymous

Gamblers Anonymous (GA) is a self help group which advocates abstinence from gambling. It is the most popular choice of treatment in the United States. Despite GA being reported as only having an eight percent success rate, Ledgerwood and Petry (2005) reported that GA is a good adjunct to other therapies, such as community-based treatment and CBT.

30

XGAMBLE Report

7. Brief Intervention

Brief intervention is also known as ‘Solution-focused brief therapy’ (SFBT) which was designed for use with people with substance abuse problems and has been adapted for the gambling field (Korn & Shaffer, 2004). This approach signals a shift from a more professional driven, medically (oriented) treatment to a more client-centred, solution focused intervention. The evidence to date suggests that SFBT is cost-effective and may be particularly useful for those in the early stages of problem gambling and individuals who are not committed to the full scale of counselling for their gambling problems (McCown & Howatt, 2007; Petry, 2006).

8. Integrated approaches to interventions

Integrated approaches may include cognitive and behavioural approaches with a motivational interviewing style or motivation enhancement strategies. For example, the trans-theoretical model incorporates 15 models, and is mainly comprised of cognitive and behavioural components, the application of motivational interviewing and the ‘stages of change’ model (DiClemente, 2003).

Other intervention models include the approach proposed by Walters (1994), described as a lifestyle interpretation of problem gambling activity, based on work with ‘incarcerated gamblers’ - individuals who have recovered from problem gambling. This model is guided by three primary objectives: to cease those lifestyle activities which lead to the problem gambling activity; to develop skills in order to manage gambling-related conditions, choices, and cognitions; and to implement an effective follow up and support programme. The second stage of the intervention involves behavioural techniques common to many behavioural intervention programmes. This stage of establishing a vehicle for change suggests a variety of activities that are designed to expand the client’s repertoire of social, coping, thinking, and general life skills. Included in these activities are such cognitive behavioural therapies as cue-control, substitution, limiting access to gambling opportunities, cognitive reframing and rational restructuring. In this way, lifestyle theory provides a broad range of possible therapy techniques that may be effective for a range of individuals. This lifestyle-based intervention also builds on the premise that for individuals who reduce the amount of time and money spent on gambling, there will be a need to fill this time with other ‘activities’.

9. Pharmacological interventions

Pharmacological treatments have primarily involved serotonin reuptake inhibitors (SSRI), naltrexone, and mood stabilisers.

Serotonin plays a role in the problem gambler’s aetiology regarding impulsiveness, obsessive-compulsive behaviours, and depression (Crockford & el-Gruebaly, 1998; George & Murali, 2005; Ledgerwood & Petry, 2005). In short the SSRIs were not effective as treatment for problem gambling and there was no evidence that they had anything other than an antidepressant effect on clients.

31

XGAMBLE Report

Naltrexone is a promising drug that regulates reward and reinforcement impulses within the brain. This has been found to be responsible for illnesses related to impulsivity, such as problem gambling or kleptomania. Naltrexone was found to be as effective as psychological approaches; however a high proportion of clients experienced side effects with this medication that included decreased libido, nausea and dry mouth. On the other hand, medications (e.g. Naltrexone) can benefit specific groups of clients, for instance, clients with coexisting depression or where counselling is proven to be unsuccessful. However, this will not be an option for most people in New Zealand; Naltrexone is not currently funded for the treatment of problem gambling.

Mood stabilisers such as lithium or carbamazepine regulate similar mechanisms which typically treat bipolar disorder. With this in mind, there appears to be a close symmetry between problem gambling and bipolar disorder (George & Murali, 2005; Hollander et al., 2005). Consequently, these stabilisers which ameliorate symptoms of bipolar disorder also result in a decrease in problem gambling behaviour (Moskowitz, 1980).

10. Residential Therapy Programmes

A small number of residential therapy programmes have been established in the United States and United Kingdom. For example, the Gordon House Association (GHA) (Griffiths, Bellringer, Farrell-Roberts & Freestone, 2001), which is the United Kingdom’s only specialist and dedicated residential facility for people with problem gambling. Their therapeutic programme is centred round a nine-month period of residency involving progression through a number of phases: initial assessment; ‘coping with today’ (Phase One); ‘coping with yesterday’ (Phase Two); ‘coping with change’ (Phase Three), ‘coping with tomorrow’ (Phase Four), and ‘coping on my own’ (Phase Five).

There is no evidence supporting more intensive treatments (e.g. inpatient treatments) over community treatments, except where clients have multiple serious mental health issues. Group based treatments have been found to be less effective than individually based treatments. For problem gambling treatment in New Zealand, we continue to focus on the provision of individual community-based treatment via face-to-face, telephone, and web-based (and more recently mobile-phone text) contacts, perhaps augmented with support groups.

Aside from these professionally provided intervention measures, some individuals affected by problem gambling go through a ‘natural recovery’ and stop or reduce their gambling without professional help (e.g. McCown & Howatt, 2007; Hodgins & el-Guebaly, 2000). Research participants in a recent New Zealand study talked about how they shifted from recreational gambling to a more regular, intense form of gambling and how they then reverted the shift in some cases (Tse, Abbott, Clarke, Townsend, Kingi & Manaia, 2005). Other people benefit from self help programmes (e.g. Gamblers Anonymous or workbooks). Overall, the more severe the gambling is, the more structured the intervention required (Hodgins et al., 2001).

32

XGAMBLE Report

1.2.2 Telephone and Face-to-Face Counselling Telephone counselling was initially set up to help those in a crisis. The Samaritans started the first telephone counselling service in London in 1953 and this service has flourished worldwide (Varah, 1973). The accessibility of this format of counselling reaches a myriad of specific population groups with specific needs including problem gamblers and their significant others (Adams, 2004; Coman, Burrow, & Evans, 2001; Scarfe, 2001). Telephone counselling offers accessible emotional support for those who cannot attend face-to-face sessions (e.g. agoraphobia, social phobia, disabled individuals, those living in remote areas, or individuals with time limitations) (Cheers, 2000; Hass, Benedict & Kobos, 1996; Jackson, 2005; Shepherd, 1987).

Telephone counselling is usually the first mode of contact that a problem gambler or significant other has with a problem gambling helping agency. While it is usually crisis driven, many helplines also offer ongoing counselling, information, and referrals (Marotta, 2005; Watson, McDonald, Pearce, 2006). Although some callers initially ring for information, they may also use this route as a way to start talking about their problems (Hunt, 1993). This mode of counselling provides easy access to help and it allows anonymity for those who feel threatened by face-to-face contact. Telephone counselling may also provide a ‘stepped care approach’ to face-to-face counselling or as an adjunct to face-to-face counselling (e.g. relapse prevention, follow up) (Coman et al., 2001; Marotta, 2005). Telephone counselling often involves an integrated approach, inclusive of but not limited to, cognitive behavioural techniques, motivational interviewing approaches, client-centred counselling, narrative therapy, and cultural models (when applicable).

Face-to-face counselling for problem gamblers is usually brief counselling lasting anywhere between 1 to 6 sessions on a weekly or fortnightly basis lasting 50 minutes per session. Followed by the brief time-limited sessions, there are follow up sessions with the practitioner or one can attend follow up groups who meet weekly. Face-to-face counselling also offers opportunities for referrals for budgeting, housing problems, drug or alcohol issues, gambler anonymous meetings, and various health issues. Face-to-face counselling also accommodates those who require couples or family counselling. Practitioners in New Zealand employ an integrated approach, inclusive of cognitive behavioural techniques, motivational interviewing approaches, client centred counselling, narrative therapy, sand-tray therapy and cultural models (when applicable) (personal communication, Problem Gambling Foundation, 2007).

The similarities and differences between face-to-face and telephone counselling are found within the dynamics of these services. Many of the therapeutic approaches offered in face-to-face counselling can be executed within a telephone session with similar success. The differences between telephone counselling compared to face-to-face sessions include the lack of visual cues such as facial expressions and body language; these components are considered an integral part of the therapeutic dynamics in face-to-face counselling (Coman et al., 2001; Hunt, 1993). Consequently, telephone counselling requires more emphasis on verbal cues from the caller, and the practitioner must execute empathy verbally (rather than employing head nods) (Hunt, 1993). Therefore the present study

33

XGAMBLE Report

seeks to investigate whether there will be any differences in outcomes between face-to-face and phone interventions for individuals affected by problem gambling.

Another difference between the two modes of help include the fact that these services may attract individuals with different characteristics. For instance, there are different characteristics between young people and older problem gamblers who ring a helpline. This would demand different counselling needs from the practitioner (Potenza, Steinberg, Wu, Rounsaville & O’Mallery, 2006). Consequently, many helplines have set up helplines for young people with gambling problems (e.g. Youthline) to cater to these needs (Marotta, 2005). As Jackson and Shane (2005) have identified different needs in face-to-face counselling between men and women, there is a need for different clinical interventions for these cohorts.

1.2.3 Methodological issues related to interventions studies for problem gambling This section focuses on reviews of the methodological issues for intervention studies, as documented by key researchers in the field, namely the work by Ledgerwood and Petry (2005); Toneatto and Ladouceur (2003) and lastly Jackson, Thomas and Blaszczynski (2003). Each subsection or discussion is concluded by reflecting on how the present study was able (or not able, in some cases) to address those issues raised in the previous reviews.

Ledgerwood and Petry (2005) concluded that while recent research highlights some promising treatment interventions, the inability to determine the effectiveness of these treatments may be due to a lack of scientific sophistication and application of rigorous standards. Methodological flaws include lack of controlled studies, small sample sizes and lack of adequate follow up. They also note that finding and retaining research participants with personal experience of problem gambling is difficult and that future research would need to develop better methods of tracking and contacting participants for follow up. Recommendations for improving intervention research are: • Researchers should ensure that there are enough multiple contacts to successfully

follow up clients over time; • Always use a control group for comparison studies; • Use large enough samples; • Address comorbidity with intervention planning; and, • Report variables for intervention failures.

Addressing the methodological issues raised by Ledgerwood and Petry In the present study, contact with research participants was maintained by practitioners as much as practically possible. The researchers also followed up with participants on completion of their sessions, or when the practitioners informed the Project Team that the client was no longer attending the scheduled sessions. Another follow up was conducted 6-weeks after the termination (planned or unplanned) of the intervention. The Project Team developed and maintained spreadsheets and systems to closely monitor the progress of clients and to help ensure that data was collected in a timely manner. A

34

XGAMBLE Report

control group was not employed as this is the first study on the effectiveness of problem gambling interventions in New Zealand, and the main purposes of the Phase Two study were to learn more about the counselling interventions provided over the telephone and through face-to-face contact, and about designing and implementing a randomised clinical trial on problem gambling in the context of New Zealand. The concern about low uptake of participants in the study did not allow the inclusion of a control group at the present developmental stage of outcome studies in New Zealand. The issue of comorbidity for example, use of mental health and addiction services and history of head injury was routinely collected in the baseline measure in the present study. Chapter Three of this report on the Phase Two of the study will discuss the variables which might be associated with the intervention outcomes.

Toneatto and Ladouceur (2003) reviewed interventions for problem gambling and rigorously outlined the many methodological limitations with reference to effective validated treatment (EVT). EVT are treatment approaches which undergo rigorous clinical trials to prove treatment effectiveness. EVT is necessary for some countries for example the United States where rely on insurance payments for treatment. Insurance companies will only pay for treatment if the treatment provider employs EVT. In practice, the authors of the study had to analyse treatments on a less rigorous basis, as there are currently no treatments for problem gambling that meet the strict EVT criteria. Toneatto and Ladouceur lowered their standard to include any study which randomly allocates clients to one of at least two conditions, where one condition is a control group. However, despite various treatment successes indicated in their review, they pointed out that many researchers have different definitions of: • Abstinence (e.g. cutting down gambling significantly rather than complete abstinence

from gambling); and, • Drop outs (e.g. drop outs in some cases are defined as failures to avoid inflated rates

of success).

These inconsistent definitions within studies greatly challenge the validity of intervention effectiveness. Overall, the most important point that Toneatto and Ladouceur (2003, p. 290) addressed is:

“almost all studies reviewed lacked theoretically relevant measures to assess the mediators of change posited by the conceptual model underlying the treatment being evaluated.”

Addressing the methodological issues raised by Toneatto and Ladouceur In the present study, both complete abstinence or reduction in time/money spent in gambling following the intervention, was used to gauge the effectiveness of interventions provided in this project. The ‘drop outs’ in the present study were reported separately, but not treated as failure. Instead, consistent with best practice in clinical trials the ‘intention to treat’ method was used for analysis whenever appropriate.

Toneatto and Ladouceur (2003) suggested that measuring tools, such as the Gambling Attitudes and Beliefs Survey (GABS), should be utilised as mediators of change. The GABS was used in the Phase Two of the present study. GABS has been proven to assess

35

XGAMBLE Report

treatment outcomes of cognitive therapies and high scores on the GABS correlate significantly with the South Oaks Gambling Screen (SOGS) for problem gambling.

Jackson, Thomas and Blaszczynski (2003, p. 8-9) reviewed over 60 intervention studies and suggested that firm conclusions on treatment effectiveness are compromised by a variety of methodological flaws in many of the reported studies. These include: • Poorly delineated selection criteria and procedures for the inclusion of individuals

into intervention programmes; • Failure to take into account improvement in other areas of functioning in programmes

where criteria for success is based on whether or not the client abstained from gambling;

• Lack of distinction between intervention effects in relation to different forms of gambling;

• Varying levels of motivation to change within the sample which makes generalisation of results problematical;

• Lack of reporting of data on client intervention rejection or attrition; • Difficulty in identifying the impacts of primary interventions when a number of

interventions are used simultaneously; • Lack of clarity about whether reliable and valid measures of change are being used,

or how concepts such as ‘improvement’ are measured; • Lack of a clear cut definition of what constitutes lapse or relapse in terms of gambling

behaviour; and, • Variation in post-intervention and follow up intervals indicating a lack of a system-

wide approach to tracking the efficacy of interventions.

Addressing the methodological issues raised by Jackson, Thomas and Blaszczynski In the present study, a set of clear inclusion and exclusion criteria were defined and a series of primary (i.e. self-reported time and money spent in gambling) and secondary (e.g. cognition/attitudes towards gambling, level of satisfaction, anticipated level of success from intervention, and experience about counselling intervention; including some standardised psychological instruments) outcome measures were used to deter any intervention effects. In terms of specific forms of gambling activity, the present study recruited only individuals who had concerns about their gambling on pokie machines. Furthermore Phase One of the study sought to explore the range of intervention techniques used by practitioners in the problem gambling field; then during the course of the trial in Phase Two, practitioners had to report on their intervention sessions, including keeping records of the techniques they used, and appraisal of clients’ progress and level of risk in relation to loss of control of gambling behaviour.

Summary In summary, it is generally known that treatment for problem gambling is more effective than no intervention at all and that a combination of intervention models is generally more effective than one model on its own, as concluded by Jackson and colleagues (2003). This conclusion resonates strongly with the outcome of reviews drawn by Toneatto and Millar, 2004 (p. 523):

36

XGAMBLE Report

“…the pathways model of problem gambling synthesizes the role of the multiple determinants of problem gambling identified above and may be a useful guide to treatment.” (italics added)

The pathway model identifies three main sub-groups of individuals with problem gambling that may further support the eclectic approach to problem gambling (Toneatto & Millar, 2004): 1. Gambling controlled by conditioning, cognitive variables and poor judgement. 2. Those who suffer from pre-morbid mental health problems such as depression,

anxiety and those with limited coping skills or disturbed personal histories such as abusive backgrounds.

3. Those who demonstrate features of impulsivity and character/personality disorder or difficulties.

Research conducted by Echeburua et al. (2000) has demonstrated that mode of delivery of intervention is an important factor that warrants consideration in treatment research. As such, it was recognised that the present study should move further than merely retrospectively reviewing and evaluating different intervention models and their effectiveness, to also looking at ways of delivering problem gambling interventions, for example, provision of problem gambling counselling through face-to-face versus telephone based contacts.

1.2.4 Conclusion There are various modes of intervention services available for people with gambling problems, ranging from public health interventions through to acute interventions and relapse prevention. Types of interventions include cognitive and cognitive-behavioural therapy, motivational interviewing, self help programmes, pharmacological treatments, natural recovery and combinations of all the above. Recent research indicates that a combination of interventions work best, especially those incorporating elements of cognitive-behavioural strategies.

Future research designs need to produce systematic, credible data on intervention effectiveness and should include both qualitative and quantitative (including the use of standard measurement instruments for reliability and validity purposes) research to provide more information and richer results.

New Zealand is a unique environment (e.g. pokie machines are not limited to large gambling venues but are also available in pubs, clubs and restaurants) with diverse cultures that require different approaches (see Herd & Richards, 2004 for an example of a kaupapa Maori intervention programme). It is important that research methodologies be modified to suit the cultural context. It is envisioned that the results from this proposed research will go someway to providing more comprehensive information of gambling interventions in New Zealand.

37

XGAMBLE Report

CHAPTER 2 PHASE ONE STUDY

Phase One of this research sought to identify and analyse the range of problem gambling intervention services and approaches utilised in New Zealand. It also aimed to identify some of the elements of successful interventions, as perceived by practitioners, service users and family/whanau.

2.1 Research aims This phase provided stand alone qualitative information on problem gambling intervention services in New Zealand, as well as informing and enhancing the methodology of Phase Two. The specific aims of the Phase One study were to:

1. Identify different problem gambling intervention services as representative of the range of services in New Zealand (including services located outside Auckland);

2. Analyse the range of tasks and therapeutic techniques undertaken by practitioners in problem gambling services;

3. Identify important features of these problem gambling intervention services, as perceived by clients (people who gambling, and their family/whanau); and,

4. Explore differences in terms of perceived effectiveness, between clients who attend one session only, and those who attend three or more sessions.

2.2 Methodology The main task of Phase One involved briefly reviewing and analysing a range of the different intervention services available. This included both generic and population group specific services in Aotearoa, New Zealand: i.e. Problem Gambling Foundation of New Zealand (including their Asian services), Gambling Helpline New Zealand, Oasis Centres/NZ Salvation Army, Pacific Mental Health, Alcohol, Drugs and Gambling Services – Tupu Team, Pacificare Trust, and Te Rangihaeata Oranga. These service providers were approached by the Project Team in the early planning stages of the project and all agreed to take part in Phase One of the study.

In order to provide a comprehensive analysis and overview of problem gambling treatment services, a number of data collection methods were employed: • Administration of a standardised questionnaire to problem gambling service

practitioners; • Interviews with problem gambling service practitioners; and, • Interviews with problem gambling service past clientele (both family/whanau and

individuals with problem gambling).

38

XGAMBLE Report

2.2.1 Recruitment process Problem gambling service practitioners and clientele were recruited through the problem gambling intervention services listed above. These services all accepted invitations to participate in this phase of the research and also sat on our Technical Advisory Group (with the exception of the Gambling Helpline, who were consulted separately). The Technical Advisory Group (TAG) provided technical and clinical service advice to the Project Team throughout the course of the research.

Individuals were sought for three research groups: 1. Problem gambling intervention practitioners. 2. Clients who had sought intervention services in the past, in relation to their own

gambling issues. This included clients from three categories; those who have attended:

a. One session; or, b. Three or more sessions; or, c. Ten or more sessions.

3. Clients who had sought intervention services in the past, in relation to someone else’s gambling issues (i.e. family/whanau members).

The sampling methods of Grounded Theory were used as the theoretical framework to guide the recruitment process (and data analysis) of the Phase One study. Recruitment of participants, data collection and data analysis occurred simultaneously. Three basic procedures were used to direct the recruitment of research participants, namely: Open, Axial and Selective Coding (Creswell, 1998). • ‘Open Coding’ refers to the process of discovering and categorising phenomena

under study and then developing properties that will describe each category (e.g. ‘how effectiveness is defined?’ and ‘what makes an intervention effective as perceived by practitioners and service users?’). At this stage, the researchers did not pre-select any particular individuals for the study as it was unclear which concepts or experiences would be relevant to the topic of study. As such, openness, rather than specificity, guided the initial sampling process (Strauss & Corbins, 1990). Once the initial category outlines had been formed and their dimensions developed, the coding procedures of ‘Axial Coding’ and ‘Selective Coding’ were applied.

• ‘Axial Coding’ aims to uncover and validate any relationships that exist between the categories or concepts developed so far.

• In the final step of ‘Selective Coding’ (also known as Discriminate Sampling), individuals were specifically chosen in order to maximise research opportunities (e.g. individuals who only saw the practitioner once as opposed to those who attended three or more sessions) for verifying emerging relationships between categories and/or for filling in poorly developed categories (Creswell, 1998; Strauss & Corbin, 1990).

39

XGAMBLE Report

The following inclusion/exclusion criteria were developed for Phase One and were based on the advice from the TAG and the Project Team:

Inclusion criteria • Individuals who were able to give informed consent; • Individuals who were aged 16 years or above; • Clients who had used the problem gambling services in the past, attended one session

as a pre-determined ‘single’ session (as perceived by practitioners /therapists) and those who attended one session and not returned, or, clients who attended three or more sessions, or, the long term clients - represented as ten or more sessions; and,

• Individuals who were able and willing to reflect on their experiences in providing and/or using problem gambling intervention services.

Exclusion criteria • Individuals who were at immediate risk of harm to self or others; • Individuals who were not interested in participating in the research process; • Individuals who were not able to engage in a 20-60 minute interview; and, • ‘Sensitive cases’ - such as individuals who were involved in a court case against a

gambling industry member or treatment agency.

During the recruitment stage, a standard letter of invitation or email note was sent to relevant agencies and individual managers/CEOs to request assistance with participant recruitment (i.e. practitioners and clients with whom practitioners worked with in the past). Eligible participants were given a one-page study summary and asked if they were interested in participating. If the participant expressed interest, they would be referred to the research field worker, who would answer any questions she or he may have and provide the written information sheets. The individual then indicated their consent by signing the consent form (or alternatively not consent to participate). After obtaining the consent, an interview session was completed. Participants were interviewed in person at a location convenient to them and the interviewer, or via the telephone if they received support through telephone services. The interview took about 40 minutes for clients whereas practitioner interviews took approximately 60 minutes. To maximise reliability and strengthen the rapport between interviewers and research participants, ethnically-matched interviewers were engaged to conduct the data collection. Qualitative interviews were not audio-taped but field notes were written. All the participants, including agencies practitioners, clients with personal experience of problem gambling and family members/significant others were reminded that their participation was voluntary. The following wordings were inserted in the research Information Sheet:

“Your assistance in this matter is greatly appreciated and complete confidentiality and anonymity is promised. Your name will not be used and your data would be identified by codes only. If the information you provide is reported or published, this will be done in a way that does not identify you as its source.

40

XGAMBLE Report

Please note that you do not have to take part in this interview if you do not feel comfortable and that you can refuse to answer any particular question. You can withdraw from the research at any time and ask any questions about the research at any time during participation. You have the right to withdraw your information/data up to August 31, 2006. Whether you continue to participate or not will have no reflection upon your treatment.” (abstracted from Clients Information Sheet, italics added)

2.2.2 Participants: Sample size and distribution The research team decided, with advice from the TAG members, to delete the split between one and more than three sessions for the family/whanau participants and to introduce a long-term client category. Service providers were asked to identify and access clients to ensure an even balance of males and females, telephone and face-to-face clients and match the ethnic quotas where possible. In the Phase One study, “telephone clients” were individuals who accessed problem gambling counselling services through a dedicated gambling helpline; whereas “face-to-face clients” were service users who were seen by practitioners primarily in person, face-to-face, although they might also receive occasional, brief telephone calls for ongoing support.

Recruitment was monitored by the Project Team and direct communications were maintained between the services, the interviewers and the project manager in relation to accessing clients during this recruitment stage.

2.2.3 Data collection instruments/semi-structured interview guidelines Semi structured interviews were conducted with both service users and practitioners. Interviews involving Maori, Pacific, Asian and New Zealand European service users and practitioners were conducted with ethnic specific interviewers to ensure cultural safety. Early consultation with representatives from each of these population groups had indicated that it is desirable to collect data through individual interviews (as opposed to focus groups, particularly for family members of their own culture). This enables people to feel safe, and prevents them feeling ashamed or condemned when sharing their experiences of how gambling unfolded as an issue in their life and/or for their family. Interviews were conducted either at service premises or locations that were convenient, safe and comfortable for the clientele (e.g. Tupu practitioners accompanied the interviewer to the clients’ homes).

Demographic data A set of questions was also included to ascertain selected demographic details. For clients these related to gender, ethnicity, location, agency/mode of counselling received etc; and for practitioners, years of experience in field and qualifications/training.

41

XGAMBLE Report

Counsellor Task Analysis (Problem Gambling), CTA (PG) (Jackson, Holt, Thomas & Crisp, 2003):

The CTA (PG) questionnaire gave practitioners the opportunity to document the practice and theories they use when dealing with a problem gambler, a member of the problem gambler’s family and the community at large. It aims to provide a broad overview of the complexity of the practitioner’s role, specify the range of tasks they perform, and document the relationship between the frequency of task performance and the practitioner’s beliefs about the importance of the tasks performed.

The CTA (PG) was chosen as a data collection instrument for the present study, primarily for two reasons. Firstly, the CTA (PG) captures the practice of the problem gambling practitioner where clients represent a heterogeneous group of men, women, older people, younger people, people from Anglo and non-English speaking backgrounds, with a wide array of clinical diagnoses, who live in a variety of geographical locations including rural, regional and metropolitan settings (Jackson, Thomas & Blaszczynski, 2003). The CTA (PG) can be used for a number of objectives, as relevant to the present research questions: • “Determine whether there is a commonality of counselling practice across counselling

sites in cases where they have a generic label, such as in state-funded gambling treatment programs, and that variation occurs where this is justified by demonstrable regional differences in client characteristics.

• Assist in our understanding of the influence of different theoretical orientations on detailed counselling practice. For example, do all practitioners identifying their interventions as cognitive-behavioural, perform the same range of tasks? Similarly, variations in counselling practice may be correlated with variations in client characteristics as a way of determining the extent to which practitioners are capable of engaging in differential practice.”

(adapted from Jackson, Thomas & Blaszczynski, 2003, p. 85)

Secondly, the CTA (PG) instrument displays adequate psychometric properties in terms of internal consistency, content validity, concurrent validity, construct validity and test-retest reliability. Some selected findings are:

“The results of the factor analysis provided preliminary support of the content and factorial validity for the nine CTA (PG) subscales. In addition, when estimates of internal consistency reliability were computed for the factor scales, the results were comparable to those computed for the scales as conceptualised.” (Jackson, Thomas & Blaszczynski, 2003, p.82)

Furthermore the reliability of the instrument was evaluated using internal consistency methods, based on the coefficient alpha, as well as the test-retest method. The overall mean Cronbach alphas for the Frequency subscales were .80 for year 1 and .76 for year 2; and for Importance subscales, the alphas for year 1 and year 2 were .87 and .81 respectively.

The CTA (PG) addresses nine dimensions of practice activity through nine subscales:

42

XGAMBLE Report

• “Assessment: Tasks that relate to assessing case situations to determine need, risk, urgency and engaging clients either in making use of services or preparing them for transition or termination.

• Treatment Goals: Tasks that relate to developing counselling goals for the outcomes of treatment. These may include change goals or choice goals.

• General Interventions: Tasks that use basic helping skills such as interviewing, questioning, and counselling to assist individuals and/or families in understanding the problems they experience in social functioning and help them to examine possible options for resolving those problems.

• Gambling Interventions: Tasks that use a variety of therapeutic techniques and strategies with individuals (and family members if appropriate) to improve their social functioning through abstinence or controlled gambling.

• Family Interventions: Tasks that use clearly defined formal treatment modes to help families to improve their social functioning or resolve social problems.

• Interventions for Related Problems: Tasks that deliver a variety of services, which enable individuals, families and employers to cope with legal, financial and employment problems due to problem gambling.

• Referral: Tasks that coordinate service planning with internal agency staff and providers from other agencies in order to provide any additional services that the client may require.

• Education: Tasks that inform and educate individuals, the general public, the gambling industry, and community agencies about problem gambling. In addition, engaging in activities that strengthen one’s own practice effectiveness and expand one’s professional competence.

• Research/Policy: Tasks to collect, analyse and publish data so as to influence public opinion, public policy and legislation in the development of a knowledge base of accountable, professional and accessible counselling services for problem gamblers and their families.”

(Abstracted from Jackson, Holt, Thomas & Crisp, 2003, pp.70-71)

Semi-structured interview data Semi-structured interview schedules were developed for use with practitioners and clients (both problem gamblers and their whanau/family members, relatives and significant others) of problem gambling services. These schedules were based upon the reviewed literature and the interview group schedules (see below), which have been previously developed, piloted and employed by Jackson, Thomas and Blaszczynski (2003) in Victoria, Australia. Interviews focused on items such as:

Practitioners: • What do you consider is meant by a ‘successful’ intervention with a person(s)

with a gambling-related problem? • How do you currently measure success in your service? • What do you consider to be the key elements of ‘successful’ methods of

intervention for people with gambling problems? Prompts included: o theoretical model; o individual/couple/family intervention orientation; o intervention intensity (actual vs. ideal number of sessions);

43

XGAMBLE Report

o intervention timing in problem onset; and, o location of service delivery, e.g. centre based, outreach, home based.

• How do the ‘best’ methods differ for different groups? Prompts included: o men/women; o people from different cultural backgrounds; o type of gambling preferred; o severity of the gambling-related problem; and, o length of time the person has experienced the problem.

Clients: • What do you think affects a person’s wish to gamble? • What would make someone go from being a regular gambler to being a problem

gambler? • What would you do to better protect problem gamblers? • Did you have experience of self-exclusion, and did it work? Please elaborate. • What are the features of a good problem gambling intervention service? • (If applicable) How do current problem gambling services give you what you

need? • What are your aims of your involvement with a problem gambling programme -

“abstinence” or “control of”? Please elaborate.

The questionnaires were presented to the TAG and research interviewers for appraisal and were then piloted. Following the pilot study (involving clients, family members and practitioners), a couple of questions were deleted to make the interviews briefer and the wording on the client/family interviews was modified to be more understandable to the general public and to incorporate a cultural perspective(s).

2.2.4 Data analysis • Counsellor Task Analysis (Problem Gambling)(CTA)(PG) Descriptive analysis was used to create a profile of the types of activities performed by practitioners and the frequency and level of importance allocated to each of those activities/tasks identified. The analysis focused on identifying trends and distribution of the collected data set (e.g. the clustering of the items). The New Zealand CTA results were also compared to the earlier study conducted in Australia. As there were only 27 practitioners involved in this survey, no further analytical or inferential analyses were performed.

• Semi-structured interviews Qualitative data in the form of detailed information provided by practitioners and clients were formed into categories and analysed thematically (Creswell, 1994). According to Patton (1990), there are two ways of representing the patterns that emerge from analysis of such data. First, the researcher can use the categories developed and articulated by the people studied to organise presentation of particular themes. Second, the researcher may also become aware of categories or patterns for which the people studied did not have labels or terms, and the researcher develops terms to describe these inductively generated

44

XGAMBLE Report

categories. It was found that broad categories were suggested by the structure and purpose of the interviews, while sub-categories emerged from within the data (Jackson, Thomas & Blaszczynski, 2003).

The preliminary data analysis and interpretation obtained from interviewing practitioners and service users were presented in the third meeting of the TAG. The general comments included: • The data presented at that stage were rather basic and required further careful

examination and cross-checking with interviewers involved in the study. • It was important to interpret the data with caution given the small sample size of

clients involved. • The recruitment process and data analysis procedures had to be documented clearly in

the Final Report. • The data obtained from this qualitative study were valuable in describing what the

general issues were, the general pattern of practice, and respondents’ comments on the topic under investigation.

Following the TAG meeting, the data were further analysed. A few follow up actions were taken: • A complete set of data including the original responses/comments from research

participants and the coding/themes developed were sent to all four ethnic interviewers to verify the accuracy of recording and interpretation and to check if there was any omission of critical themes or findings.

• When the discussion section and commentary included in this report was drafted, it was also sent to all interviewers for verification and comments.

• All comments were collated by two members of the Project Team and integrated into the final report.

To conclude, Phase One of the research was guided by following principles:

Comprehensive approach: As outlined above, a comprehensive approach to investigating the existing problem gambling interventions in New Zealand was undertaken. For example, rather than concentrating solely on information collected from practitioners, data was also collected from clients attending these services (including family members/whanau, relatives, and significant others). The research team also recognised the diversity of clients with regard to duration of help-seeking behaviour and cultural context. It is possible that clients who only attend one session with a practitioner may differ substantially to those who attend a number of sessions. For the purposes of this phase of the research, it was of interest to investigate if these groups differed according to their perceptions regarding the role of their therapist.

Community approach: The investigators endeavoured to obtain the support of various individuals and community groups in the ethnic populations of interest to the study (e.g. Pacific Tupu Clinical Services team, NIU Development, Maori National Reference Group – Te Herenga Waka O Te Ora Whanau, Te Rangihaeata Oranga, and PGF

45

XGAMBLE Report

National Asian Services). These people liaised between the research team and potential participants, promoting the study and encouraging community member involvement.

Systematic and theory-based approach: Members from the Project Team and the TAG have a strong theoretical background in psychosocial interventions generally and problem gambling intervention models specifically. This allowed the Manual Writing Group to articulate a framework to capture most of the range of existing interventions provided for individuals affected by problem gambling.

2.2.5 Research rigour The research rigour for Phase One can be summarised from the above methodology section in Table 1.

Table 1: Research rigour in Phase One study Category Activities/ remarks Instrumentations • Used standardised, reliable and valid measurement

instruments • Used established semi-structured interview guides

Triangulation of data sources

• Used quantitative and qualitative data from practitioners and different client groups (the weave of quantitative and qualitative data is reflected in the logo used for the study- XGAMBLE)

• Collected data from multiple sources namely, service users, whanau/ family or significant others and problem gambling practitioners

Being culturally • Included various community/ethnic groups; and the project responsive was able to obtain support from those groups

• Used ethnic matched interviewers to interview and analyse data

Technical support • Used a specialist Technical Advisory Group for intervention services input, knowledge, expertise, recruitment of practitioners and clientele, and data interpretation

Data verifications Presented the findings in community meetings and problem gambling conferences to verify the accuracy of data interpretations (for details also see Chapter 4, Section 4.5 Dissemination Plan)

46

XGAMBLE Report

2.3 Results The framework of the semi-structured interview guidelines has been used to report the key findings from Phase One. Furthermore, as one of the features of the present study was to engage four population groups (Europeans, Maori, Pacific Island and Asians) and explore how their experiences in receiving counselling services might differ, the data are presented under the four population groups. In Phase One, Pacific peoples included participants from Samoa, Tonga, and Niue; and Asian peoples were participants of Chinese and Korean descent. On the one hand caution must be advised in the interpretation of these results and the subsequent generalisations due to small sample sizes, on the other hand findings emerged from the present qualitative study provide a wealth of knowledge on the topic under investigation and resonate with experts in the field. Lastly, as the number of participants who attended more than ten sessions was small and their responses did not seem to vary significantly with those who used more than two sessions, the findings from these two categories were combined in the following report (this decision was supported by members of the TAG). To demonstrate how the data were interpreted and themes were developed, a sample of the Maori data analysis record is attached in the XGAMBLE Accompanying Appendices.

In total, Phase One semi-structured interviews were conducted with 58 past service users (including individuals with personal experience of problem gambling and family members/significant others). Table 2 presents details relating to the ethnicity, gender, intervention mode, number of sessions, and client category (gambler, family/Whanau) of these participants. In the rest of this section, participants’ verbatim responses have been used to represent their voice and the data that emerged from the Phase One qualitative study. A coding system was used to provide basic but unidentifiable information about the research participants, which might provide the readers with a clearer context to interpret the quotes. An example of a participant code is: “#4F/T/HAM/PG/RS”. Codes can be interpreted through application of the following information: • #4 fourth client interviewed by that particular field researcher • F gender: F- female; M- male • T type of services received: T- telephone; F- face-to-face services • HAM city location where the research participants were recruited e.g. HAM -

Hamilton, AKL - Auckland, CHC - Christchurch, HB - Hawkes Bay • PG background of the participants: PG - individuals with personal

experience of problem gambling; W - Whanau or family or significant others

• RS initials of the researcher

47

XGAMBLE Report

Table 2: Demographic and background information for service users who completed Phase One semi-structured interviews

European Maori# Pacific Asian

Individuals with problem gambling Sub-total 11 13 6 9

Women 7 8 2 2

Using telephone services@

Using face-to-face services Using 1 or 2 sessions

Sub-total

8

3

4

3

3

10

3

6

2

4

3

5

Nil

9

4

5

Women 3 3 4 3

Family members affected by problem gambling≠

Total (58 participants) 14 19 11 14

Note: # As there was only one male in the original Maori stream, the TAG advised the research team to recruit a few more men to ensure a broader representation of the range of possible experiences. Seven additional Maori men (including both service users and family members) were successfully recruited and interviewed; new relevant data were added to the Maori Data Tables in the appendices and to the results/discussion. @ “Telephone services” denote services where clients receive help solely through a dedicated gambling helpline; whereas “face-to-face services” mean that service users are seen by a practitioner primarily in person, face-to-face but they might receive occasional, brief telephone calls for ongoing support.≠ No information on service type and number of sessions for the family members was collected for the present study

48

XGAMBLE Report

2.3.1 European experiences of utilising problem gambling services

Description of Participants This section of the report presents the data collected from the European participants in the service user semi-structured interviews. It describes the fourteen Europeans participants’ experiences in utilising problem gambling intervention services, particularly with regard to themes of why people may gamble, access services, and the successfulness of the intervention from clients’ perspectives. The relevant demographics of the European participants are presented in the table below.

Table 3: Demographic and background details for European service users who completed Phase One semi-structure interviews

Gender Attending sessions Type of services Male Female 1-2

Sessions Multiple sessions

Telephone Face-to-face

Individuals with problem gambling (11)

4 7 4 7 8 3

Family Members/ Significant Others (3)

0 3 Not applicable (NA) 3 NA

Gambling Participants were invited to discuss their perspectives on gambling generally. When asked what gambling is, problem gambler responses included taking risks and betting money for a chance to win:

“Taking the chance of making money” (#4F/T/HAM/PG/RS)

“Placing money for a return” (#6F/T/HAM/PG/RS)

Only family members responded that it was an “addiction” (#2F/F/AKL/W/RS). Another family member elaborated:

“It is an addiction like drinking or smoking pot” (#1F/F/AKL/W/RS)

There were no differences in responses between clients attending one session and those who had attended multiple sessions, nor differences between telephone and face-to-face clients.

When asked why people gamble, the responses of telephone clients centred on escape (from boredom, loneliness, depression, stress) and stimulation:

“Escape from reality” (#6F/T/HAM/PG/RS)

“the sound of the machine” (#2M/T/AKL/PG/RS)

49

XGAMBLE Report

Face-to-face clients referred to excitement, escapism and making money:

“Thrill of a win” (#10M/F/AKL/PG/RS)

Family members, cited a number of reasons to gamble, including hope for a better life, to chase losses, and for enjoyment/socialising. For example:

“Win money” (#2F/F/AKL/W/RS)

“dream lifestyle, easy way out” (#3F/F/AKL/W/RS)

“possibility of winning a lot of money, chasing the losses, escapism, relaxation, socialising” (#1F/F/AKL/W/RS)

The general categories among all groups included escapism, making money, excitement, hope for a better life and to chase their losses.

From the clients’ perspectives (for both telephone and face-to-face), gambling becomes problematic when one is not managing financially, losing control of oneself and starting to lie to family:

“You are out of control and the gambling is controlling you” (#4F/T/HAM/PG/RS)

“Start to lie and defend self. You make excuses. My bank balance was going down, spending money I shouldn’t be spending…” (#1F/T/HAM/PG/RS)

A number of problem gambling clients also referred to the attributes of the machines (pokies) and being driven by them:

“Machines motivate someone to develop a problem with gambling. The need for a big win and when you get a big win it drives you to spend more money” (#2M/T/AKL/PG/RS)

Family members focussed on the gambler’s unreliable behaviour (lying, deceit, wasting money):

“Gambling becomes the single most important past time and they start to lie about what they do” (#3F/F/AKL/W/RS)

Again, there were no differences between clients that only attended one or two sessions and those that attended multiple sessions.

Participants were asked for suggestions on how to better protect individuals and family against the impact of gambling. The main responses were to remove or reduce access to gambling opportunities (i.e. pokie machines), introduce modifications to the machines, remove access to money, self exclusion, provide information on gambling as a health and behavioural issue, and reduce the stigma attached to problem gambling for all client groups. Some examples of these responses include:

50

XGAMBLE Report

“to reduce the number of pokie machines and its availability, take pokies out of poor areas, don’t allow bartenders to give free drinks to make them stay longer, take away noise from pokies as it is trigger...” (#4F/T/HAM/PG/RS)

“time limits, money limits & pop ups” (#8M/F/AKL/PG/RS)

“greater understanding of gambling as a health issue and addiction as opposed to giving the behaviour a stigma” (#10M/F/AKL/PG/RS)

The multiple session clients also added counselling and that family members should be informed of one’s gambling behaviours so they can offer support:

“Let family know” (#5F/T/DDN/PG/RS)

The family members emphasised the need to learn coping skills:

“Focus away from money and stop the financial impact of gambling. Practice responses to problem gambling” (#3F/F/AKL/W/RS)

Participants were asked how they regarded their own gambling. While the majority had either stopped or were in control of their gambling, some were still in the process of gaining control/stopping gambling completely. Some of the multiple visits clients were still finding gambling problematic and the face-to-face clients in this group mentioned gaining control by changing to a discontinuous form of gambling like lotto and poker games:

“100% control, I play poker once a month with 8 friends and bet only $60” (#10M/F/AKL/PG/RS)

Interventions The clients all stated that their aims of involvement with problem gambling services were to gain information and either control gambling or to abstain from it completely. Family members were looking for support and information.

Participants were primarily motivated to seek help because of financial and life pressures, losing control of their gambling behaviour, pressure from family, or their own dishonest behaviours becoming too much. The problem gambling clients responded mostly with financial issues:

“No money for daily living needs” (#2M/T/AKL/PG/RS)

“Out of control, uncomfortable” (#7M/T/CHC/PG/RS)

Multiple visit clients in face-to-face services mentioned the dishonest behaviour and family pressure:

“Had enough…wasn’t being honest” (#10M/F/AKL/PG/RS)

Family members all mentioned the dishonest behaviour that was discovered by the family:

51

XGAMBLE Report

“when I discovered my husband was gambling and lying about it” (#3F/F/AKL/W/RS)

Some of the participants made use of the self exclusion scheme:

“yes, it worked because I made it work. It is symbolic for making a commitment to control my gambling. I don’t think I would get caught if I walked into a casino and started gambling (unless I won a lot of money)” (#10M/F/AKL/PG/RS)

Of those that did not self ban, some commented that the process was too difficult:

“No, it would be too difficult to do this. I would have to go every pub and casino. Plus, you get fined if you break it twice” (#1F/T/HAM/PG/RS)

When the participants were asked what they saw as features of a good service, they all discussed the importance of being provided with good information and advice (including websites and internet message boards), service availability when needed, the good attributes and abilities of practitioners and being able to deal with other life issues as well as the gambling. The single session clients focussed on information, resources and service availability:

“I rang…she was very good telling me what can be done and gave me information about resources” (#2M/T/AKL/PG/RS)

Multiple session clients also focused on the content of the counselling and the client/ practitioner relationship:

“identified triggers, deal with feelings, deal with loneliness” (#7M/T/CHC/PG/RS)

“good caring counsellors, they never give up on me, good listeners, relationship important, empathetic” (#11F/T/AKL/PG/RS)

The face-to-face clients also stressed the importance of involving family to help them to understand and the utilisation of group therapy:

“bring my wife to understand” (#8M/F/AKL/PG/RS)

“individual counselling plus weekly group work” (#9F/F/AKL/PG/RS)

The family members mentioned couple counselling, identifying the harm and the importance of having a practitioner for themselves:

“someone there to listen to me and he knew what I was feeling” (#1F/F/AKL/W/RS)

Participants were mostly pleased with current services and felt that their needs were met (as alluded to in the previous quote), especially with the availability of telephone support and returning calls:

“someone called back – phone contacts” (#3F/T/AKL/PG/RS)

52

XGAMBLE Report

It was also noted that:

“different counsellors had different approaches” (#7M/T/CHC/PG/RS)

According to all the participants their cultural and special needs were met if required (gender, age) but most did not need any special services.

Recommendations to improve services received by participants included the “need for more funding to have more counsellors” (#10M/F/AKL/PG/RS) available and at different times, actively contacting clients and utilising text messaging.

Successful Interventions All participants’ agreed that a successful intervention involved stopping or controlling gambling and having someone who will listen, empathise and provide information. Family members added that acceptance of the problem and positive changes in the gamblers behaviour were important; a typical example is:

“Husband accepts and changes” (#2F/F/AKL/W/RS)

Successful outcomes for multiple session participants specifically included a good sense of overall wellbeing, stopping gambling, being able to deal with life issues and acknowledgement that recovery is a process/journey:

“…feeling well, complete abstinence, deal with life problems, recovery as a journey” (#7M/T/CHC/PG/RS)

Family members acknowledged the importance of support when it is difficult to talk to others:

“to gain support, can’t talk to friends or relatives” (#2F/F/AKL/W/RS)

In summary, all participants agreed that the factors that contribute to achieving successful interventions include having someone available who will listen, empathise, understands gambling/problem gamblers, and will provide information, advice and referrals (e.g. budgeting) to support people to stop or control gambling. The one session clients also indicated the importance of the service being available when required and that being contacted by the services also helps. For example:

“the Helpline is backup support and they are always available” (#4F/T/HAM/PG/RS)

The multiple session face-to-face clients also mentioned the use of individual, couple and group therapies. Family members commented on learning to support and protect the family from gambling harm while supporting the family member’s recovery:

“Understanding of the gambler and the gambling is not a matter of just ‘will power’. To learn not to take it personally and to find ways to support myself and protect the rest of the family. To also learn to support my husband’s recovery” (#3F/F/AKL/W/RS)

53

XGAMBLE Report

Multiple session clients noted that seeing a person who that does not listen, empathise, or offer specific advice can hinder success. Being judgemental was also noted as a hindering factor:

“someone does not listen, discounts what you said, minimises your experiences, not feeling heard, being judged” (#7M/T/CHC/PG/RS)

Face-to-face clients also mentioned that the changing of meeting locations and having no commitment for therapy can hinder success.

54

XGAMBLE Report

2.3.2 Maori experiences of utilising problem gambling services

Description of Participants This section of the report presents qualitative information that was collected as part of Phase One of the study. The information collected relates to how Maori: utilise gambling treatment services, what support is provided to Maori, treatment options offered to Maori and outcomes desired from gambling intervention services. The relevant demographics of the Maori participants are presented in the table below.

Table 4: Demographic and background details for Maori service users who completed Phase One semi-structure interviews

Gender Attending sessions Type of services Male Female 1-2

Sessions Multiple sessions

Telephone Face-to-face

Individuals with problem gambling (13)

5 8 3 10 3 10

Family Members/ Significant Others (6)

3 3 NA NA 6

Gambling Participants were invited to discuss their definition and general perspective on gambling. Various responses were given.

Those who sought help through the telephone helpline considered gambling as a form of addiction and as a result you “splash all your money on gambling” (#8F/T/HB/PG/SJ) and as a consequence it ruins your family and is a “transgression against your whakapapa” (#2F/F/HB/W/SJ).

For those who had sought help through face-to-face counselling, similar views were expressed, such as:

“gambling involves spending money you cannot afford” (#3F/F/HB/PG/SJ)

“betting/winning – a fools game” (#1F/F/HB/PG/SJ)

“an escape from problems and chasing loses”(#2F/F/HB/PG/SJ)

“a form of addiction” (#7F/T/HB/PG/SJ)

[and can lead to] “a form of “sickness” (#9M/F/HB/PG/SJ)

People gamble for many different reasons and for some individuals it is an opportunity for:

“having time out for me” (#8F/T/HB/PG/SJ)

55

XGAMBLE Report

“habit, boredom” (#2F/F/HB/PG/SJ)

“way to manage depression and anger” (#8F/T/HB/PG/SJ)

“to make money” (#2F/F/HB/PG/SJ)

For others, gambling is a means of coping with “stress, dealing with a low sense of self esteem” (#4F/F/HB/PG/SJ) and avoiding problems within a family or with key personal relationships. Gambling is also often used as a means to escape from personal or family problems or as a “quick fix to meet financial needs and wants” (#9M/F/HB/PG/SJ).

The environment was also mentioned as a reason to gamble, as it can be stimulating with “lights on the machines, winning a jackpot, and sounds” (#1F/F/HB/PG/SJ). Participants noted that gambling increased levels of arousal and excitement to make them feel different. Also the availability of gambling opportunities was mentioned by most of the male participants, “Availability of venues & advertising (#1F/F/HB/PG/SJ, as a reason why they gamble.

The Whanau participants considered that gambling is used for many different purposes, and in general, is a “means of coping and dealing with different triggers” (#2F/F/HB/W/SJ) and that “their Whanau (#1F/F/HB/W/SJ)” can be a reason why some people start gambling in the first place.

Participants considered that problem gambling occurred:

“when you spend money and don’t think about it, you just keep going and going and going and continue to spend money” (#8F/T/HB/PG/SJ)

Others said it was when they are “unconsciously spending money they don’t have” (#8F/T/HB/PG/SJ), which should have been used to purchase other items such as, “rent and groceries” (#2F/F/HB/PG/SJ). Problem gambling occurs when you are no longer in control. This then affects how you behave, such as:

“dishonest…stealing money…not caring about yourself” (#4F/F/HB/PG/SJ)

Problem gambling also affects other areas of life:

“social and work functioning” (#5F/F/HB/PG/SJ)

“affects family friends and work” (#9M/F/HB/PG/SJ)

“can cause depression” (#1F/F/HB/W/SJ)

Participants indicated that problem gambling leads to debt, it affects communication with others and denial of problems is common. Overall, problem gambling affects all aspects of a person’s total health and wellbeing and that of their family.

Participants identified that information conveyed through “pamphlets about gambling” at venues were helpful, as well as “television advertisements” (#8F/T/HB/PG/SJ). It was

56

XGAMBLE Report

suggested that the following message should be promoted to protect against gambling: “never start” (#7F/T/HB/PG/SJ). Maori television was identified as an ideal medium to provide information to Maori about gambling, risks associated, the signs of problem gambling, and where help can be obtained.

It was considered that more information was needed and that it should be available and delivered in different settings, such as, on the “marae or at wananga” (#4F/F/HB/PG/SJ). Participants also felt that school children should be a key focus and that they should be fully informed of the risks associated with gambling. In relation to families, it was thought that people affected by problem gambling should be encouraged to seek help, so that there is more openness about problem gambling and a collective approach can be taken to providing support.

It was felt that good counselling should be available, and that this should be supported by appropriate information. Some participants suggested that venue operators should also provide information, that it should be clearly visible with better signage, and that operators should be held accountable if do not comply with their statutory requirements.

Finally, most of the male participants suggested that the best protection from gambling would be to limit gambling opportunities:

“Ban all machines” (#1F/F/HB/W/SJ)

“Get rid of the pokies” (#3F/T/HB/W/SJ)

“limiting availability of machines, restricted advertising and actually enforcing legislation” (#1F/F/HB/PG/SJ)

Interventions Participants who sought help acknowledged that they wanted to gain better control of their gambling and that this was the primary motivating reason for seeking help. The level of control over gambling varied with participants covering the full continuum: some had stopped for a number of years, while others were still out of control. Reasons for seeking help included concerns for the wellbeing of their children, the possibility of getting a mortgage over their house or the sale of their car. The difficulty of not being able to pay bills was a constant issue, as was pressure from family or a partner to seek help. Verbal and emotional abuse about their gambling was also common. Participants’ gambling patterns also impacted upon their work, and for one participant, it led to dishonesty and the disintegration of key relationships. One participant had been motivated to seek help by a television advertisement about gambling.

Participants reported on their experience of self banning: while it had been helpful for six participants, three found it had limited success and encouraged the gambler to go to another area where they are not excluded from gambling. Four of the gambling participants did not try self banning.

57

XGAMBLE Report

Participants identified features of a good problem gambling service, from their experience and perspective. In relation to the telephone helpline, participants made the following positive comments:

“anonymity, makes you more open and honest, had tried groups and counselling services before” (#7F/T/HB/PG/SJ)

“follow up calls” and “access to a 24 hours contact helpline, counselling and website” (#5F/F/HB/PG/SJ)

Participants who had used face-to-face counselling commented that they appreciated:

“the buddy system, and counsellor who comes to you and who understands and cares” (#3F/F/HB/PG/SJ)

In addition to being able to involve other family members and ensure that children were safe, the incorporation of Tikanga in counselling sessions was identified as helpful:

“counsellors willingness to listen…[being] non judgemental…having an open door policy” (#9M/F/HB/PG/SJ)

“individual and couple sessions, Tikanga was used” (#1F/F/HB/W/SJ)

The ability to refer participants on to other services was noted as helpful and some of the telephone helpline and face-to-face clients appreciated that services were offered free of charge.

Both services did meet participants’ personal needs in different ways. For those who used the telephone helpline, they valued “being able to ring any time” and to speak freely and openly with a practitioner while “being anonymous” (#7F/T/HB/PG/SJ). For one gambler who attended face-to-face counselling, the ability to go somewhere just for them self was appreciated:

“where I could nurture my own self, as being involved with problem gambling can eat away inside you. Being in the group is one day when I can nurture myself. Massage and heath has been very beneficial to me. It is stress relief for me.” (#3F/F/HB/PG/SJ)

Overall, both groups of gamblers using either help line or counselling services liked the unconditional positive regard and non-judgemental approach they received and help in developing positive solutions to their personal issues, such as how to pay debts/budget and establish and utilise support networks. The inclusion of cultural strategies such as learning waiata was valued, as was as learning new relaxation exercises.

Participants’ personal objectives in seeking help were discussed. Irrespective of the service chosen, all participants had sought help in order to control their gambling. They also wanted to understand more about problem gambling, and to learn to manage their gambling and other problems so that it was not problematic to them personally, or detrimental to other important relationships. The importance of having someone who

58

XGAMBLE Report

understands and will listen was important to participants and for Whanau it was about getting help for their family and helping them to understand the impacts of their gambling on the family.

For one participant affected by her gambling, she sought help in dealing with the situation, so that she would:

“stop being violent to her husband” (#2F/F/HB/W/SJ)

In gaining control, participants also commented then they could then decide if they wished to stop gambling completely.

A suggestion for improving the help offered by telephone services was to have more time to spend with practitioners.

In relation to face-to-face counselling sessions, participants’ identified a need for more local programmes and information about how to identify/access help earlier. The mobility of practitioners was particularly appreciated. However two participants noted that messages left were not passed onto the practitioners. Attending group sessions was considered useful but the confidentiality of issues discussed needed to stay within the group session. Listening to other stories was also considered helpful. Another participant mentioned the time delay between first contact and getting a face-to-face appointment and that timely contact between the services is essential. Also letting clients know they have the ability to change practitioner would be helpful.

Cultural needs were discussed and appeared to be met in telephone services and all but two of the face-to-face clients. Face-to-face practices in particular, had tailored their services to meet the cultural needs of participants which assisted clients to engage and feel understood and supported. This was accomplished through the inclusion of Tikanga, the offering of a Maori practitioners, “help with provision of food supplies”, and support to the wider Whanau.

Successful Interventions Participants had different perspectives on the meaning of a successful intervention. For those using the telephone helpline service, their measure of success was:

“to get control of gambling and minimize it” (#8F/T/HB/PG/SJ)

In contrast, those using face-to-face services requested a service where they could speak honestly about their gambling and lifestyle generally without any form of judgement. They wanted help to stop gambling, and wanted services to provide ongoing support and a range of services, including preventative strategies, group sessions, as well as services that affirmed who they were in terms of Tikanga Maori. They also wanted a place to where people could go to get together.

59

XGAMBLE Report

Individuals’ differed in their views of desired service outcomes. For those using the telephone help line, help was requested in ensuring that gambling was no longer part of their life, never going back; preventing relapse was important. Similar aspirations were desired by those using face-to-face services. Help was wanted with stopping gambling, completing treatment, getting out of debt, being able to pay bills and buy things, involvement of the whanau in their treatment, restoring disintegrated relationships and achievement of inner peace or happiness:

“Self-discovery, reinforced identity, reinforced Tikanga, restored relationship, daughter was educated” (#1F/F/HB/W/SJ)

Those using the telephone help service recognised that achieving these outcomes required on-going support and the commitment of the gambler to work hard and focus on personal goals. Those using face-to-face services identified the need for ongoing support and unconditional love from their family, willingness, honesty and commitment from the gambler to implement action plan, to develop good communication skills and coping strategies, a buddy system, an approach which involved the whole family, and a practitioner who had: skills in communication and empathy/rapport, could build the self confidence of clients, provide financial interventions, and utilise Kaupapa Maori approaches in their service delivery. They also requested services that are welcoming, safe, non judgemental, accessible, and able to support social interaction so that alternatives to gambling could be explored.

Barriers to successful outcomes were also explored. Those using the telephone help line identified a lack of self control and being in denial and services that were cut back as pertinent issues. Those using face-to-face services identified a different set of issues, including: practitioners not engaging with clients properly, a lack of compassion and understanding of the needs/circumstances of clients, lack of willpower, unfinished treatment, having easy access to money, presence of mental illness or other addictions, dishonesty, and a family which is not supportive of help seeking. Other issues that were raised included the normalisation of addiction within families (considering it a family trait instead of a response to the environment they lived in), a lack of self confidence and unresolved personal issues. In addition to these factors (which have centred on client issues or service delivery), some participants considered that a lack of funding and understanding (of the relevant issues and their interactions) on the behalf of those funding the services, also created barriers. Again here the male participants talked about too many gambling opportunities being available (including money availability) and the environmental stimuli:

“Machines available longer, bar person lending out money, bright lights, cool music” (#3F/F/HB/PG /SJ)

A sample of the full set of Maori participants’ responses is attached in the XGAMBLE Appendices Booklet.

60

XGAMBLE Report

2.3.3 Pacific (Samoan, Tongan, Niuean) experiences of utilising problem gambling services

Description of Participants This section of the report presents data gathered from the Phase One interviews with 11 participants of Pacific descent. The semi-structured interviews, highlighted the experiences of the Pacific participants in relation to problem gambling intervention services, why people choose to gamble, reasons for accessing services. The relevant demographics of the Pacific participants are presented in the table below.

Table 5: Demographic and background details for Pacific service users who completed Phase One semi-structure interviews

Gender Attending sessions Type of services Male Female 1-2

Sessions Multiple sessions

Telephone Face-to-face

Individuals with problem gambling (6)

4 2 3 3 2 4

Family Members/ Significant Others (5)

1 4 NA NA

Gambling When asked to discuss their perspectives on gambling, several themes began to emerge. Themes associated with gambling (for both face-to-face and phone participants) included, spending money to make money, the concept of hope (hoping to win - as explained by participants), time away from family and work, and fun. Participants went on to describe gambling as a form of addiction or disease that can be self destroying and have devastating affects on family. One participant explained that gambling is a:

“Sickness affecting, destroying family, and relationships” (#1F/F/AKL/W/LA)

When asked why people gamble, responses were linked to a desire for social interaction and to escape boredom. Participants also highlighted the increased exposure to gambling opportunities in New Zealand for recent migrants whereas fewer gambling forms are available in the Islands. Other reasons described by the research participants included: concept of luck, risk taking, and creating alternative methods to fix or meet financial needs and wants. Furthermore Pacific participants regarded being connected to family and the maintenance of relational arrangements as of utmost importance and this had been reported as influencing some peoples gambling behaviour and practices. For instance, some participants noted that they gamble with family, and many were introduced to gambling by other family members. Gains through gambling were also seen as a means to contribute to household obligations. One participant discussed wanting to make more money, and viewing gambling as a means to bring money for the family:

“When we are struggling but it just makes things worse. Every time we are going through financial problems, he will go to gamble.” (#1F/F/AKL/W/LA)

61

XGAMBLE Report

From the participant’s point of view, gambling becomes problematic when they people are unable to control the amount of money they spend on gambling, when most of their time is consumed in gambling, when they start blaming others for their lack of money, when their family becomes dysfunctional and they are no longer reliable:

“When you spend more than you’re supposed to and you get carried away” ( #5F/T/AKL/PG/LA)

“Don’t do any normal thing, your time spent gambling” (#6M/T/AKL/PG/LA)

Family members of problem gamblers also noted when they perceived gambling to be problematic. They related it to ongoing beliefs in the gamblers’ mind that they could win and when the gambler had obvious personality changes. One Pacific participant stated that gambling became problematic:

“When they don’t realise their gambling is a problem and start blaming other for their lose of money” (W1F/F/AKL/W/LA)

Clients were also asked to make suggestions for ways to protect individuals and family against the hazards of gambling. The main responses centred on self exclusions from venues, for example:

“Making it okay to ban yourself or family from casino so they don’t go back” (#5F/T/AKL/PG/LA)

Other suggestions included education on limiting the amount of money and time people spend at gambling venues, placing limits on money credit, and better advertisement and promotion on problem gambling issues. Offering services was also suggested by the participants in this study; however, a challenge facing the delivery of services, is the poor awareness (according to the participants) in many Pacific communities of the available services by.

Intervention The Pacific respondents in this study stated that the issues driving them to seek help in the first place included a self realisation that they had a problem, being sick of the lies and secrecy, and wanting to break away from being dependant on family for extra money. A commonly mentioned motive for seeking help amongst the Pacific participants was family. As one participant illustrated, becoming aware of the effects of their behaviour on loved ones and encouragement from family, were often catalysts for seeking help:

“I wanted to put a stop to it and my daughter directed me to do it” (#5F/T/AKL/PG/LA)

“…sick of lies, talked to my wife, I rung the 0800 number.” (#6M/T/AKL/PG/LA)

Family members talked about recognising personality or behavioural changes in their loved ones. For example:

“Deterioration in my loved ones personality. I saw physical and mental changes. He was not the person I always knew” (#5F/T/AKL/W/LA)

62

XGAMBLE Report

“Dad’s behaviours change, aggressive towards mum and she called me all the time” (#2F/F/AKL/W/LA)

Overall, Pacific participants in the present study identified the insistence of family members as a motive for seeking help.

Some of the participants made use of the self exclusion scheme and found it helpful, for example:

“Yes it worked, because when I banned myself, I didn’t think about it” (#5F/T/AKL/PG/LA)

Others did not self bar themselves but made a decision not to attend gambling venues and thus, were able to keep it. However, others did not self bar as they felt it would make them want to gamble even more. As one face-to-face interviewee noted:

“no, it makes me want to do it more” (#2F/F/AKL/PG/LA)

When participants were asked what they perceived to be characteristics of a good service they discussed the significant role of practitioners. These participants wanted practitioners who were non judgemental, culturally aware (of what it meant to be a Pacific gambler), culturally competent and who could understand their circumstances. On the whole, the participants seemed to suggest that establishing trust in the services/workers of the services is a key to the utilisation of services.

Another theme that emerged from questions based around good services was the importance of service availability and ongoing support. Some participants said that services can only be utilised if people know that they exist and if hours of operation are not confined to the traditional 9am – 5pm working hours. As many Pacific people are either working during these hours or do not want others to know that they have gambling problems, it would be more convenient to have services that operate around the clock.

Participants within the study were pleased that they could express their needs and concerns openly and that those catering to their needs were culturally aware of these issues:

“They allow me to be open and they have men of my culture who can help” (#4M/F/AKL/PG/LA)

“Understand my culture, importance of status and fa’alavelave, always willing to listen, can ring them if I need anything, accessible most of the time…They support me, show me the right way to go. This was a new thing for me but it gave me a good feeling for me and it makes me fight for it. I’m not 100% yet but I m getting their. I’m recovering. I am doing things, speaking up and I can face up to people now. I couldn’t do that before.” (W3M/F/AKL/W/LA)

“Having an island counsellor is good, she knows me, she understands all the stuff that other people may not, you know like cultural expectations – church and stuff.” (W1F/F/AKL/W/LA)

“Culture is met. I’m PI and I am different from Palangi and Maori. They don’t know what I am thinking and understand the way I do things. I used to go to groups and be stuck with a Maori or Palangi counsellor. I didn’t feel comfortable. That’s why I didn’t take them seriously I they don’t understand what they want me to do. Also the counsellor speaks Nuiean and English and the other one speaks Samoan. I understand both of these and feel a connection with them more. They

63

XGAMBLE Report

also take me away from home and that makes me feel good cos I get bored sometimes” (W3M/F/AKL/W/LA)

Another benefit of using services was the information that provided the clients with options and the skills needed to make gambling decisions. The participants also added that they could move at their own pace and they were able to discuss non-gambling related issues, which may indirectly affect their gambling behaviour.

Within this study, Pacific service users discussed their aims; what they hoped to achieve with the assistance of the services. All generally wanted to stop gambling, with some highlighting family as a reason to stop gambling, for example:

“I wanted to stop for me and my families sake.” (#5 F/T/AKL/PG/LA)

Another outcome achieved through utilising the services was the desire for better self control and changes in their individual behaviours. From the families’ perspectives, their aims were similar to those of the gamblers themselves, self control being a key aim. As one family member noted:

“I want him to have control of his gambling. We have taken drastic measures to control it and he, myself and our separate counsellors agreed that I would have control of his money. His wages go into my account and the bills are paid and the money divided after that.” (W5 F/T/AKL/W/LA)

For the family members that were interviewed, the need to build trust and re-establish the relationships that they once had prior to the gambling becoming problematic were stressed.

Recommendations to improve services received by participants included forming an action group. Participants using phone services requested more phone practitioners on the gambling helpline, more Pacific staff who have adequate training, and more direct assistance with self exclusion. From the interviews conducted with Pacific family members, a noticeable recommendation was improvements to the availability of services in the weekends. All participants were happy with the fact that there were culturally specific services that were able to relate to their experiences, speak their native tongue and understand the significance of family in relation to gambling behaviours and outcomes.

Successful Intervention All participants agreed that a successful intervention involved accessibility to services – particularly those that were free and available when needed. Services that were able to assist individuals and families with managing finances were also seen as beneficial. From the family members that were interviewed, services which provided support to the family and having practitioners that understood also contributed to a services success.

The key feature of successful intervention, as identified by a majority of participants who gambled, was services that enabled them to stop gambling. Other themes were better self control, better management of money, when participants are seen to accept advice and attend therapy. From their family’s point of view, success for the individual with

64

XGAMBLE Report

gambling related problems is also measured in their ability to admit they have a problem and to stop blaming others.

Contributions to success were again strongly linked to the role of the practitioners within the services:

“A good counsellor. One on one counselling for me and family as well. The counsellor has to be someone you can talk to about the problem and the money lost without judging you. Also support from the counselling team and family as well.” (#5F/T/AKL/PG/LA)

Services that were readily accessible, that offered advice and direction, helped build confidence and provided support for their family and friends were also mentioned by the Pacific participants. The re-emphasis on services that provide for the family gives some indication of the significance of family to the participants. Thus, it is important for services to understand family situations and work alongside families to support individuals with problem gambling.

Factors noted by the participants and family members of problem gamblers that hindered the success of services included procrastination, not dealing with the issues, the inability to be honest with the practitioners or open up to practitioners and friends, continual secrecy, communication problems with clients and their families, and lack of motivation. This suggests that dealing with problem gambling from a Pacific viewpoint takes time. It means building a trusting relationship between the service and the clients so that they can open up.

65

XGAMBLE Report

2.3.4 Asian (Chinese and Korean) experiences of utilising problem gambling services

Description of Participants A total of 14 Asian people were interviewed in this part of study; they were all of Chinese and Korean descent. For the relevant demographic background, please see Table 6 below.

Table 6: Demographic and background details for Asian service users who completed Phase One semi-structure interviews

Gender Attending sessions Type of services Male Female 1-2

Sessions Multiple sessions

Telephone Face-to-face

Individuals with problem gambling (9)

7 2 4 5 Nil 9

Family Members/ Significant Others (5)

2 3 Not applicable Nil 5

Gambling When asked why Asian people gamble, there were no differences in responses between clients attending one session to those who had attended multiple sessions. Note also, there were no participants who received telephone counselling specifically therefore no comments on the perceived usefulness of a telephone approach or ideas for improvements in telephone services can be made from the view of Asian clients. Five main themes emerged, the first of which related to using gambling to avoid or escape from negative emotions:

“Gambling can help temporarily forget unhappy things, like smoking… Escape stress from study and family problems” (#4F/F/AKL/PG/YL)

Another participant explained that gambling can help reduce:

“…Loneliness…very few good friends…(they) do not want to talk, feel bored” (#2M/F/AKL/PG/YL)

Having too much spare time was another reason cited for gambling. One family member also mentioned post immigration issues:

“For Asian people, they have difficulties to live in a foreign country, such as language difficulties, employment problem, different custom and thoughts, anything is different for them. Try to forget worries and troubles” (#2F/F/AKL/W/YL)

The second theme centred on gambling for excitement, feelings of success, and to show off one’s wealth. Positive feelings also emerged from pleasant experiences in the gambling environment, rather than gambling per se:

66

XGAMBLE Report

“People can experience satisfaction and respect in casinos, for example, those waiters/ waitresses serve them with very good manner, make gamblers feel they are the one in charge” (#3 F/FAKL/W/YL)

Thirdly, gambling is about winning money, as put by one of the participants in a very succinct way:

“Most of people gamble for money. If those games are only with chips involved, the chips cannot be exchanged to cash, I believe that not many people will be interest in going there (gambling venue)” (#6M/F/AKL/PG/YL)

The fourth theme related to people’s gambling behaviours being influenced by friends and peers:

“My boy friend gambles often, so I follow him” (#4F/F/AKL/PG/YL)

“Friends, colleagues often talk about their exciting experiences of winning money, but never mention about their loss…” (#1F/F/AKL/W/YL)

Fifthly, both a problem gambling client and family member explained how gambling industries influence one’s gambling behaviour:

“Gambling industry misleads people, passes wrong messages that gambling can give a hope to improve the quality of life” (#5M/F/AKL/PG/YL)

“Casinos are legal gambling venues in NZ, they offer VIP gamblers discounts in accommodation, car park…” (W3 F/FAKL/W/YL)

The progression from social to more intense/problematic levels of gambling was attributed to the desire to have a big win, chase lost money, and people being attracted or drawn by the gambling industry. Some examples include:

“Human being’s psychological change: when winning money, they taste some sugarplums, gain easy money; when losing money, they regret but forget quickly…Such a cycle repeats, they gamble more and more, and then become problematic” (#6M/F/AKL/PG/YL)

“Wanted to win back. The more I lost, the more I bet. I am an emulative person, generally, my initial attitude towards gambling had changed, I became out of control of my gambling behaviours. Whenever I had money, I would immediately go to casino, and wanted to win back my loss” (#1M/F/AKL/PG/YL)

“Gambling in casino is totally different from playing Majong which Chinese/Asian people usually play before they come to New Zealand. Gambling in casino involves a lot of money, which is really dangerous... X (name of gambling venue) cunning marketing strategies, when they know someone have trouble in gambling, they will offer free/ discounted meal, special cards for special customers on some certain days, cheaper parking via mail to lure people...” (#2F/F/AKL/W/YL)

Interventions The first question explored in this section relates to those factors which motivated Asian people to seek help from problem gambling services. There were five main categories of responses explaining why people sought help for their gambling problems. Firstly, Asian people sought help because they were directed to by an authority. For example, some

67

XGAMBLE Report

casino self-exclusion schemes require attendance at a prescribed number of counselling sessions before the clientele can return to the gambling venue.

“I was forced to attend XXX [name of agency] counselling service because I illegally entered casino after self-barring” (#4F/F/AKL/PG/YL)

“After self-barring, XXX [name of agency] called me for counselling. So I came” (#1M/F/AKL/PG/YL)

“I self-barred myself with my girl’s insistence. And then I was compulsory to attend a counselling session” (#2M/F/AKL/PG/YL)

“I have been issued “trespass”, and (it) was compulsory to attend the counselling session ... I am not allowed to enter the XXX [name of gambling venue] for 2 years” (#5M/F/AKL/PG/YL)

Secondly both problem gambling participants and family members mentioned how they were prompted by what they saw or heard in the media about the availability of professional services for individuals affected by problem gambling. Examples of responses are:

“I have noticed the labels of XXX [name of agency] on the casino’s gaming tables at a very early stage, and I knew there was such an organization existed…” (#3M/F/AKL/PG/YL)

Other participants had similar experiences:

“The gambling hotline number attached on the casino tables, the free brochures displayed within the casino (by the entrance)” (#9M/F/AKL/PG/YL).

“I found the XXX [name of agency] number from the labels attached on the pokies machines, I realized that my gambling was out of control, and then I dialed the XXX [name of agency] phone number. No one answered the phone, so I left a voice message, the next day, a counsellor contacted me” (#6M/F/AKL/PG/YL)

Family members had very similar accounts of how they came to know problem gambling services and subsequently contacted them:

“Saw the advertisement on some Chinese paper, and then my husband and I rang up” (#5F/F/AKL/W/YL)

“I know XXX [name of agency] through broadcast, Chinese newspaper, and some other resources with Chinese text. I realised that I could not find any way to go further, therefore I asked for help from this organisation” (#4F/F/AKL/W/YL)

Thirdly, although less common, some individuals sought help when they were asked or encouraged to by their family members or close friends. For instance:

“With some friend’s suggestion, and also found the self-bar pamphlet in XXX (name of gambling venue). I called up XXX [name of agency] by myself” (#8F/F/AKL/PG/YL)

A family member added:

“I was suggested by my classmate to seek help from XXX [name of agency]” (#1F/F/AKL/W/YL)

68

XGAMBLE Report

The fourth theme (also less common) related to a realisation of the harms caused by problem gambling. One participant recalled vividly:

“I have read lots of books about the gambling psychology and behaviours, and gained a good understanding of them. But every time I gambled, I could not control myself. I browsed XXX [name of agency] website and did their gambling test, I found that I had the gambling problem” (#3M/F/AKL/PG/YL)

Fifthly, a family member said she contacted problem gambling services because she knew the practitioner personally:

“I know some gambling counsellors in person through social activities, so I called the counsellor for help” (#2F/F/AKL/W/YL)

Regardless the number of sessions attended by problem gambling clients, they all defined ‘successful interventions’ as having someone who understand their background and could explain why they became addicted to gambling. The typical responses were:

“Understand the gamblers’ psychological status which is very important” (#2M/F/AKL/PG/YL)

“The staff needs to understand the gamblers’ psychology, the triggers behind. Re-establish gamblers’ hope/goal. Good listening skills: when a person was in a hardship that every thing was lost, he/she might have some thought to commit suicide, my counsellor was listening to me, understood what we need, gave me a hope to live on” (#6M/F/AKL/PG/YL)

“The family-like kindness and care offered by staff. Staff cares gamblers with a real heart…” (#7 M/F/AKL/PG/YL)

These sentiments were also shared by family members who used gambling counseling services; they also described good intervention as:

“Help client analyse, give clients comfort...(have) effective communication skills. Understand clients’ emotional feelings. Work as a bridge between gamblers and their family” (#1F/F/AKL/W/YL).

Family members also referred to a list of qualities that practitioners should have in order to deliver good interventions. The range of qualities included professional knowledge, friendliness, caring, ability to bridge between gambling clients and family, and maintaining confidentiality:

“Can help the gamblers admit their gambling behaviours. Help the gambler recognise the harm he has brought to his family. Kind staff makes clients feel like at home” (#5F/F/AKL/W/YL)

“Knowledgeable staff, so that they can speak to clients with their knowledge. Friendly, helpful. Meeting with family and know the clients expectation. Maintain confidentiality” (#2F/F/AKL/W/YL)

“Counsellors should be competent, have qualification and knowledge, successful personal profile, their words should be convinced...Help gamblers establish their value concepts, guide them think of the influences caused to others when they decide to doing a thing. Care about the emotional change of gamblers. Organise support groups, so that gamblers can meet, share and understand each other through peer support. Work as a bridge between gamblers and their family. Maintain confidentiality” (#3F/FAKL/W/YL)

69

XGAMBLE Report

Participants who attended one to two sessions also commented that ‘good interventions’ would result in clients being referred to group meetings and receiving suggestions about activities other than gambling:

“[Counsellor] refers [clients] to recreation activities to help gambler escape boredom in New Zealand. Only talking [counselling] is not enough” (#1M/F/AKL/PG/YL)

“A good problem gambling service should have some group discussion with other gamblers together” (#2M/F/AKL/PG/YL)

Those who attended multiple sessions, indicated that ‘good interventions’ meant that they could attend appointments with little hassle:

“Convenient location, easy parking. Good follow-up services” (#7 M/F/AKL/PG/YL)

Furthermore, a number of clients mentioned how counselling services are seen as ‘useless’ if gambling is still readily available. Examples of responses include:

“The current problem gambling service cannot do much for gambling, the power is too little and passive - only a ‘baby sitter’ for casinos, to look after those whom casinos abandon” (#5M/F/AKL/PG/YL)

“Nothing can help, until there are other entertainment activities/places operate all the time, 24 hours a day, 7 days a week. And the excitement of the activity must need to match the level the gambling does” (#4F/F/AKL/PG/YL)

Successful Interventions Problem gambling participants and family members shared very similar views in terms of the outcomes of receiving assistance and support from counselling services. All participants mentioned that counselling had helped them to stop or control their gambling:

“Admit the fault in the past, can bravely face and realize, make new changes, how to walk to out the situation…Make control over gambling with personal volition, and with professional advice/assistance”(#6M/F/AKL/PG/YL)

“Realise the harm caused by gambling…Help my husband stop gambling” (#1F/F/AKL/W/YL)

Another common outcome related to regaining positive emotions about oneself, an improved sense of health and wellbeing, and resuming a normal life:

“Make myself feel better, reduce emotional effect, want to go back to normal life as soon as possible, don’t think those bad things any more, those bad experiences can make me feel negative, regret and confused” (#1M/F/AKL/PG/YL)

“Get rid of addiction. Can control himself, psychological health can be improved. Walk out the shadow of gambling caused to my boy friend and me” (#3F/FAKL/W/YL)

“My husband can stop gambling, and we can go back to our normal life” (#5F/F/AKL/W/YL)

70

XGAMBLE Report

Participants who had attended multiple counselling sessions mentioned that rebuilding relationships was one of the outcomes of receiving support:

“More care about the family, keep meaningful relationship with them” (#8F/F/AKL/PG/YL)

“She also offered counselling services to my wife, so that we can have gain better understanding of each other which help effective intervention” (#7M/F/AKL/PG/YL)

When participants were asked to describe what contributes to successful interventions, their responses could be grouped into seven main themes. Firstly, a very strong theme related to the qualities of practitioners and the practitioner-client relationships:

“A good counsellor whom treat clients with kindness and patience. Friendly attitude” (#5F/F/AKL/W/YL)

“The counsellor is honest, keeps confidentiality, cares about my family’s feeling. The counselor asked me to write self evaluation, summarise my strength, and let me gain confidence through doing these” (#3M/F/AKL/PG/YL)

“My counsellor is very patient for me, let me talk and guided me, made me reduce the self-blaming, and shame. Encourage me and help me build up a hope in life” (#6M/F/AKL/PG/YL)

“Respect me as a man; the counsellor treated me as her brother, did not blame/ judge me” (#7M/F/AKL/PG/YL)

Secondly, it was fairly common for participants to refer to finding alternative activities to replace gambling. This response was not mentioned by family members:

“Comprehensive entertaining activities/ places in New Zealand” (# M/F/AKL/PG/YL)

“Organize activities, such as art performances to build up friendships with other international students/ people from the same hometowns” (#6M/F/AKL/PG/YL)

“More gambler discussion group, recreation activities, art performance” (#2M/F/AKL/PG/YL)

“Develop healthy interest, and make good use of the time” (#3M/F/AKL/PG/YL)

Thirdly, it was imperative for clients, in particular international students, to learn basic life skills in order to regain good control over their life; a few participants gave specific elaborations:

“Investment and budgeting courses to teach students to manage money because most of the Chinese students are the only child in their family, and we don’t have any idea how to make use of money” (#6M/F/AKL/PG/YL)

“Students need to learn managing money, budgeting control” (#3M/F/AKL/PG/YL)

The fourth set of common responses related to gaining a better understanding about gambling behaviours and the subsequent influence on the person’s behaviour:

“The staff need to understand the gamblers’ psychology, the triggers behind” (#6M/F/AKL/PG/YL)

71

XGAMBLE Report

“Pre-warning/ prevention system will play a successful and effective role for people with a gambling-related problem” (#5M/F/AKL/PG/YL)

“Provide systematic analysis of gambling psychology. Understand the reason(s) of each gambler’s gambling behaviour, adopt/ customize/ individualize methods for every gambler” (#3M/F/AKL/PG/YL)

Fifthly, good follow ups were identified as contributing to a “successful intervention”. Responses include:

“Keep on-going contact by phone calls at a regular basic, saying fortnightly to ask about the life status” (#8F/F/AKL/PG/YL)

“Understand every one individually, offer focusing support, and follow-up services which would be ideal once fortnightly to understand the situation. If follow-up is longer than two weeks, the gambler might have already lost a few times” (#6M/F/AKL/PG/YL)

Some participants indicated that seeing someone who had stopped gambling and participating in religious activities were important elements to their recovery from problem gambling. The following is an example of this sixth theme:

“One of my friends who has longer gambling history than me, he has successfully stop gambling after believing in Jesus. I saw his witness and want to help myself too. I went to his church twice, and can feeling the spiritual work which made me cry and regret” (#4F/F/AKL/PG/YL)

A family member added:

“Spiritual support through religion, someone who cares. Publish the real stories, examples to encourage gamblers stop gambling” (#3F/FAKL/W/YL)

From the family members’ point of view, they emphasized the importance of forming an alliance between the practitioner and family to make therapy work:

“Communication between gamblers and significant others” (#2F/F/AKL/W/YL)

“Cooperation from family, friends and relatives. Family should insist, not give up. Support from girl/boy friend” (W4 F/F/AKL/W/YL)

The seventh, and final theme, was voiced by both problem gambling clients and family members. They expressed strong opinions on how the government should ban or regulate the gambling industry to ensure that counselling really works:

“Government should ban international students getting into casinos” (#1M/F/AKL/PG/YL)

“Government should assign an independent agency to implement relevant gambling laws, and not allow the casinos continuously make use the power the government gave to protect themselves, but hurt gamblers. Internal Affairs should look at the current legal system and review its effectiveness. A well-designed computer system will certainly help prevent gambling problems. XXX (name of agency) should be given more power to approach gamblers, its intervention should start as early as possible” (#5M/F/AKL/PG/YL)

“Government’s cooperation, in terms of making policies and implementing laws. Success cannot only rely on a few peoples’ effort” (#3F/FAKL/W/YL)

72

XGAMBLE Report

Along a similar vein, participants discussed how problem gambling services should work with government and relevant community bodies to outreach the community to promote the message about harms caused by problem gambling:

“XXX [name of agency] can hire some community movie theatres to run workshops, and show movies after the workshops, thus, it will attract more students/ people to come and learn…The contents of workshops should include study, life, personal safety, dealing with emergency, discussions about NZ news and international news, and introduction of different study institutions” (#6M/F/AKL/PG/YL)

“Strong and wide promotion to raise public awareness, such as issuing pamphlets at schools, workshops/ activities involved with international students, focus on prevention and protection” (W3 F/FAKL/W/YL)

What hinders good interventions? Problem gambling clients gave a range of responses including:

“Loan-sharks are around in XXX [name of gambling venue], hunt for gamblers (particular young, student-look gamblers). They force them borrowing money, doing things for them. Once people get into their palms, they cannot get freedom. They force them to do lots of illegal, unethical things” (#1M/F/AKL/PG/YL)

“Face is extremely important for Asian. Asian people need to overcome this barrier (to seek help)” (#3M/F/AKL/PG/YL)

Family members agreed:

“Face issue disturb gamblers seeking help” (#1F/F/AKL/W/YL)

Another category of factors hindering effective intervention centred on the lack of services or resources:

“Insufficient human resource in XXX [name of agency], not much flexibility in time. XXX [name of agency] lacks group work for gamblers. More psychological tests to measure the effectiveness of stopping gambling” (#3M/F/AKL/PG/YL)

Family members tended to worry about how denial of gambling problems and the affiliations with gambling friends might hinder effective interventions:

“Gambling people around in the community” (#1F/F/AKL/W/YL)

“Denial of gambling behaviours. Not wanting to face the fact of being a gambler. Casino can be a disaster” (#2F/F/AKL/W/YL)

There were many questions exploring different aspects of practice that were designed for this phase of the study, however in this report, only the critical aspects and sections of the results were included here. Full records or transcripts of the results are available from the Project Team upon specific request.

73

XGAMBLE Report

On reviewing the data provided by participating service users for Phase One of this study, it is apparent that the four population groups expressed very similar views on the issues being investigated. There were more similarities between the groups than differences, with most differences attributable to different levels of emphasis for the issues. For a comprehensive discussion of the observed differences and similarities, see the Phase One discussion (Section 2.4).

The next section reports results relating to practitioners’ experiences of providing problem gambling services. As with part one, the results have been reported according to the four population groups that the practitioners’ identified as the primary client population that they work with. Moreover, each practitioner who took part in this research primarily worked with clients of their own ethnicity. A total of 27 practitioners were interviewed from a range of the current problem gambling services in Auckland, Hamilton and the Hawkes Bay region of New Zealand. They included the Problem Gambling Foundation of New Zealand (and their Asian Services), New Zealand Salvation Army Oasis Centres, Gambling Helpline, Te Rangihaeata Oranga, and the Tupu team at Pacific Mental Health Alcohol, Drug and Gambling Service. This means that the two primary modes of service delivery in New Zealand, that is face-to-face and telephone, were represented. “Telephone counsellors” were practitioners who provided problem gambling counselling services solely through telephone calls; for the present study, practitioners for this group were recruited from the Ministry funded Gambling Helpline. The “face-to-face counsellors”, were practitioners recruited from Ministry funded problem gambling services. They provided their services primarily through face-to-face meetings with clients, although they might also give their clients occasional, brief telephone calls (or in some cases, text-messaging was used as well) for ongoing support. It is worth noting that that some, but not all, practitioners who participated in the Phase One interview were treating practitioners for the clinical trial of the Phase Two study.

The practitioners were drawn from New Zealand’s four main population groups (10 NZ European, 7 Maori, 6 Pacific and 4 Asian). Pacific practitioners included those from Niue, Samoa, and Tonga, and Asian practitioners were of Chinese and Korean origin. The practitioners’ working experience ranged from one to twenty years in the gambling and addictions field. Approximately 63% of the practitioners were female. The majority of practitioners (74%) engaged in face-to face counselling activities while the remaining practitioners primarily used telephone counselling as a medium of intervention.

As with the participating service users in Phase One, a simple coding system was used for participating practitioners. An example of a code is: “#16/E/7/06”. This code can be interpreted with the following information: • #16 sixteenth practitioner who was interviewed by that particular researcher

for the present project • E denotes the ethnicity of majority of clients seen by that practitioner: E –

European; M – Maori; P - Pacific; and, A - Asian (Chinese & Korean) • 7 the interview was conducted in July • 06 the interview was conducted in the year 2006

74

XGAMBLE Report

2.3.5 European practitioners’ experiences of providing problem gambling services

Practitioners’ basic demographic information A total of 10 European practitioners participated in the Phase One Practitioner Semi-structured Interviews: six face-to-face practitioners (two of which also conducted group based work) and four telephone practitioners. All practitioners had between one and five years of experience in the problem gambling field, although the average experience was slightly lower in face-to-face practitioners (2.1 years) compared to telephone practitioners (3.5 years). While the range (one to five years) and average (3.5 years) of experience in the addiction field was comparable for telephone practitioners, face-to-face practitioners reported a more diverse range (1-20 years) and more than twice the general addiction experience than the telephone practitioners (11.3 years).

The European practitioners reported a wide range of qualifications, from formal Post Graduate studies, Diplomas, Bachelor Studies and Certificates in Counselling & Psychotherapy, Health Sciences, Alcohol and Drug studies, Social and Community Work, Supervision, Religious studies, Telephone counselling, and Banking and Money Management, to experiential qualifications in internship training, budget counselling, and Gambling Helpline training. A greater range of formal qualifications were reported by the face-to-face than telephone practitioners.

The Policy Context, Philosophy and Objectives of the Service Kaupapa/Purpose of your organisation? When asked about the kaupapa/purpose of their service, the European practitioners’ primary perspective was harm minimisation, in terms of reducing gambling related harm. They viewed their work as being:

“To reduce harm with people affected by gambling” (#16/E/7/06)

“To reduce problem gambling” ( #17/E/6/06)

One face-to-face practitioner alluded to a diverse range of skills and tasks in order to achieve these objectives:

“Contract with MOH, 1 to 1 work, intervention and screening people, community treatment and public health” (#2/E/6/06)

Face-to-face and telephone practitioners referred to helping both problem gamblers and affected others:

“To support gamblers and people affected by gambling” (#16/E/7/06)

“Help problem gamblers and their families” ( #11/E/6/06)

Key service features When asked about the key features of their services, some practitioners discussed the holistic nature and the importance of acknowledging and incorporating culture and

75

XGAMBLE Report

spirituality. Both face-to-face and telephone practitioners mentioned various service features that relate to accessibility (such as the service being free and readily available). Some examples include:

“Holistic, community based, provide education, employ preventive and intervention measures” (#17/E/6/06)

“Home visits, spiritual component, holistic based, structured to meet people’s needs e.g. Maori group leads Maori group” (#11/E/6/06)

“Free counselling, belief system of individual and Christian based with human touch and spiritual aspect. Considers treaty of Waitangi” (#10/E/6/06)

“No limit amount of contact service. Telephone and nationwide which reaches a wide area” (#16/E/7/06)

The roles of education, prevention and health promotion were also seen as key service features by some face-to-face practitioners. For instance:

“Provide presentations, prevention work and public health promotion” (#2/E/6/06)

Practice Theoretical/therapeutic orientations A wide range of theoretical/therapeutic orientations were reported by each practitioner, demonstrating that an eclectic approach is routinely adopted. A cognitive-behavioural orientation was cited by each practitioner and the following theoretical/therapeutic orientations were also frequently mentioned: psychodynamic (explore the past but not childhood material), systemic family therapy, motivational interviewing, biopsychosocial, problem solving, Gestalt therapy, structural family therapy and eclectic (Maori counselling/holistic models, marital counselling, human growth and attachment approach, motivational interviewing, wheel of change, Rogerian, strengths approach especially if relapse, narrative, Logotherapy, spiritual approach).

Practice – specific techniques and strategies used when counselling clients When asked to provide specific examples of the techniques and strategies used by practitioners when counselling clients, the most frequent response, was reflective listening. A number of other responses were cited by both face-to-face and telephone practitioners; these included role playing, visualisation, and narrative reporting.

In addition to these common techniques/strategies, face-to-face practitioners also reported using reflective listening, art therapy, motivational interviewing, psychodrama, sand tray, identifying strengths, talking existentially, teaching relaxation techniques, time management and empty chair techniques.

Overall, a less diverse range of techniques were reported by the telephone practitioners. As well as the commonly utilised techniques mentioned above, decisional balance, dealing with triggers, protection of self, and budgeting strategy were cited by telephone practitioners.

76

XGAMBLE Report

Utilisation of measurement tools when assessing clients Practitioners utilised a variety of tools to gain information on their clients’ problem gambling and amount of money lost, alcohol use, and depression. Regardless of mode (face-to-face / telephone), all practitioners utilised a problem gambling diagnostic tool (either the SOGS or the DSM). The next most frequently measured issue was depression: four of the six face-to-face practitioners administered the BECKS depression screen, and two of the four telephone practitioners administered a depression checklist or suicidal screen. One practitioner from each mode administered the AUDIT, and one face-to-face practitioner indicated that they measured the amount of money lost on gambling.

To what extent is the practitioner-client relationship considered important? The importance of the practitioner-client relationship was endorsed by all the practitioners; the majority indicated that it was ‘extremely important’, and two indicated that it was ‘very important’.

Utilisation of the practitioner-client therapeutic relationship A number of relationship aspects were frequently cited by both the face-to-face and telephone practitioners, these included: building trust, developing a therapeutic relationship, and building rapport. However, these relationship aspects were used by a greater proportion of the telephone practitioners, who also reported helping the client to feel comfortable in dealing with the problem. Face-to-face practitioners also utilised the following additional relationship aspects: proposing appropriate interventions and building clients’ self confidence.

The five most frequent problems that cause problem gamblers or their families/partners to contact a service Relationship problems and financial difficulties/stress were cited by most of the face-to-face and all of the telephone practitioners. Issues relating to the wellbeing of the gambler were also reported by all of the telephone practitioners, and lack of self control and legal implications were the next most common responses across both groups. Problems less frequently cited by face-to-face practitioners included difficulty managing emotions, history of abuse, isolation & loneliness, secrecy issues & shame, crisis, employment issues, fear of behaviours, and pressure to attend. These differed to those less frequently reported by telephone practitioners: client’s confidence, lack of self control, shame, personal awareness, and other addictions.

The five problem areas that take up the most of practitioners’ time As illustrated in Figure 1, relationships, mental wellbeing, and empowering the client, were the three problem areas most frequently cited as taking up most of the practitioners’ time. It is interesting to note that practitioners rated these areas more frequently than communication issues and gambling concerns. Financial aspects and lack of support were the next most frequently rated problem areas, followed by isolation, existentialism, lifestyle balance, stress, social implications, relapses, client follow up, setting boundaries, inability to cope, and environmental concerns.

77

XGAMBLE Report

Figure 1: Problem areas taking up most of the practitioners’ time - European

7

Problem Area

6

5

4

3

2

1

0

Relationships Isolation Existentialism Lifestyle balance Stress Mental wellbeing Social implications Communication issues Empowering client Gambling concerns Financial Relapses Client follow up Setting boundaries Inability to cope Lack of support Environmental concerns

Freq

uenc

y

‘Successful’ service outcomes Practitioners were asked to define a ‘successful’ outcome in their service. They were also asked to comment on what factors contribute to and hinder successful outcomes.

How to Define? Gaining more self control and stopping gambling were cited by more than half of the face-to-face and telephone practitioners. For example:

“…when the client can get his/her gambling in control or stop gambling” (#17/E/6/06)

Following through with recommended referrals was also cited by half of the telephone practitioners as a successful outcome:

“Follow up on referral and more help” (#16/E/7/06)

“Quite hard to define since you don’t always hear feedback. Those who go to the referrals when given” (#15/E/6/06)

Less frequent responses by face-to-face practitioners referred to the clients’ level of coping and satisfaction with life. The following quotes illustrate these factors:

“When client can cope e.g. gains autonomy and independence” (#17/E/6/06) “Life is better…” (#2/E/6/06)

What Contributes? The practitioner-client relationship, coping strategies and interventions, client centred service, and support networks were mentioned by approximately half of the face-to-face practitioners.

78

XGAMBLE Report

The face-to-face practitioners mentioned a wide range of contributing factors. In particular, the practitioner-client relationship, coping strategies and interventions, client centred service, and support networks (all of which were mentioned by approximately half the practitioners). The following aspects were less frequently cited: intervention intensity, psychoeducation, narrative techniques, client empowerment, and focus only on gambling.

Although responses from the telephone practitioners were somewhat consistent with the face-to-face practitioners, they were more diverse (i.e. each one was mentioned by a small proportion of practitioners): relationship with the service, support networks, use of CBT, understanding self and what triggers them to gamble.

What Hinders? Comments relating to ‘what hinders successful outcomes’ were largely consistent across both face-to-face and telephone practitioners and can be grouped into three main categories. The role of governmental legislation, gambling policies, and agendas were all identified as relevant factors, and can be illustrated by the following quotes:

“Power and balance of gambling industry (they don’t care), government contributes to this since they gain tax money from gambling, community don’t see gambling as a problem because they benefit from it, and view of minimise the size of problem gambling in society” (#8/E/6/06)

“New Gambling Act e.g. host responsibility makes it better but there are still venues that don’t offer self exclusion (e.g. pubs)” (#14/E/6/06)

Concerns were also raised around the potential lack of confidentiality for client’s personal information. These concerns related to the conveyance of this information to the Ministry of Health:

“confidentiality with the Ministry of Health .The records of clients go on record for the MOH and not sure how secure this is” (#10/E/6/06)

Issues relating to professional training, access to professional supervision, and lack of resources were also identified by both face-to-face and telephone practitioners.

Morale and Satisfaction The majority of practitioners indicated that their work was ‘extremely satisfying’ or ‘quite satisfying’.

When asked to identify the particular aspects of their work that are satisfying/ unsatisfying, a number of themes emerged. Satisfaction centred on successful outcomes for clients (problem gamblers and significant others), such as, seeing clients achieve positive changes in their thoughts/behaviours, and seeing clients achieve a better quality of life. For example:

“When someone who comes in here leaves with a fuller richer quality of life” (#17/E/6/06)

“See hopelessness change to hope and take control of life, clients start to make good choices”

79

XGAMBLE Report

(#2/E/6/06)

“Making a difference in someone’s life” (#15/E/6/06)

While the aspects of work identified as unsatisfying were more diverse, both face-to-face and telephone practitioners expressed frustration with clients who are unwilling to participate or change and/or clients who don’t achieve positive outcomes:

“Human wasting their lives and see the self destructive patterns. They lose track of what they can become.” (#10/E/6/06)

“When significant others don’t take the steps to change” (#16/E/7/06)

Other factors included internal politics and a lack of genuine client-practitioner therapeutic relationships (face-to-face practitioners), and negative experiences when referring to other agencies and an inability to follow up with clients (telephone practitioners).

80

XGAMBLE Report

2.3.6 Maori practitioners’ experiences of providing problem gambling services

Practitioners’ basic demographic information In total, seven Maori practitioners participated in the Phase One Practitioner Semi-structured Interviews. Five of these were based at face-to-face services and two were based at telephone services. On average, face-to-face practitioners had 6.4 years of experience working in the problem gambling, with a range of 3-12 years. These practitioners also had an average of 12.8 years experience in the addiction field, with a range of 6-20 years. As only one of the two telephone practitioners responded to this set of questions, it can only be reported that they had almost two years of experience in the problem gambling and other addiction fields.

A wide range of qualifications and training backgrounds were reported, including alcohol and drug training, Certificate in Social Work Practice, marriage guidance tutoring and counselling, psychodrama psychotherapist training, Maori development and research training, short courses on CBT, motivational interviewing, mental health & addictions, coexisting disorders, management training, Diploma in Maori Counselling, Graduate Diploma in Psycho-social studies, and IT and communications technology training.

As data were collected from only two Maori telephone practitioner, it should be noted that the remaining sections will report findings for Maori practitioners as a whole (i.e. findings from the five face-to-face and two telephone interviews will be combined) and no comparisons will be made between face-to-face and telephone practitioners.

The Policy Context, Philosophy and Objectives of the Service Kaupapa/Purpose of your organisation? A common kaupapa/purpose referred to by practitioners related to assisting those affected by problem gambling (through interventions and support) and enabling people to be free from gambling. The following practitioners’ statements illustrate these views:

“To counsel problem gamblers and those affected by them” (#17/M/6/06)

“Vision is to have people free of gambling problems.” (#2/M/5/06)

“Freedom from problem gambling for all people.” (#3/M/5/06)

The roles of community education, harm reduction, and transforming lives through Christianity were also discussed by some practitioners.

Key service features When asked about the key features of their services all practitioners responded with one consistent theme - the role of ongoing and accessible (24 hours, 7 days) support for individuals (people who gamble and family members/significant others) affected by problem gambling. The following participant quotes illustrate this theme:

“Offer support to pg and also parents, partners, children affected by gambling” (#12/M/6/06)

81

XGAMBLE Report

“7 day a week, ongoing support, call backs to clients” (#17/M/6/06)

The importance of operating within a kaupapa Maori framework was frequently discussed by practitioners. Aspects such as flexibility, mobility and adaptive and comprehensive approaches were also identified as being key features of their services. For example:

“It is a specific service for gambling incorporating Maori models of practice, Ttikanga and western models and we also practice these things e.g. kaumatua. Try our best to practice it. We have a flexible approach, go out to people – this is a key feature of what we do. We offer a social work component and health promotion, not just clinical.” (#3/M/5/06)

“We are a mainstream service utilising a kaupapa Maori framework. We work with the whole family and the individual. We do our own follow-ups and go out to the community.” (#4/M/6/06)

Practitioners also identified a focus on maintaining up to date problem gambling knowledge (through research) and an ability to adapt to different ethnic groups and cultural protocols as key features of their services.

Practice Theoretical/therapeutic orientations A wide range of models were cited by each practitioner. These included cognitive-behavioural, biopsychosocial, problem solving, psychodynamic (explore the past but not childhood material), eclectic (Maori counselling/holistic models i.e. Te Whare Tapa Wha and Tikanga), gestalt therapy, structural family therapy, and systemic family therapy.

Practice – specific techniques and strategies used when counselling clients Practitioners were asked to provide specific examples of the techniques and strategies used when counselling clients. The most common response was reflective listening. The next most frequent response was to incorporate Maori Tikanga and protocols (i.e. powhiri, karakia, waiata). Additional examples provided included: role playing, schema therapy, empty chair techniques, strengths based cards, visualisation (e.g. an exercise to invite clients to imagine “going to the bush”), active listening. “Empathy” as a form of technique was also cited by one practitioner.

Utilisation of measurement tools when assessing clients Most practitioners administered some form of problem gambling screen (SOGS, DSM, 8-Screen) and a number indicated that they also assessed mental health, depression (through tools such as the BECKS depression screen, a suicidal screen, a depression checklist), other addictive behaviours (i.e. engagement with alcohol, other drugs), and the amount of money lost on gambling. Anxiety, coping and the impact of gambling on self and others were each mentioned by one practitioner.

To what extent is the practitioner-client relationship considered important? Each of the Maori practitioners indicated that the practitioner-client relationship is “extremely important”.

82

XGAMBLE Report

Utilisation of the practitioner-client therapeutic relationship Building trust and building rapport were the most frequently endorsed aspects of the practitioner-client therapeutic relationship. Practitioners also cited refocusing priorities (i.e. short and long term goals), building clients self confidence, and empowering the client.

The five most frequent problems that cause problem gamblers or their families/partners to contact a service All practitioners identified relationship and financial problems as prompting problem gamblers or significant others to contact a service. Other frequently cited responses included legal and judicial issues, lack of self control, secrecy issues and shame, and being at crisis point. In addition to these factors, practitioners also referred to the committing of crime to gamble, fear of losing house/job/status, inability to provide basic necessities, pressure to attend, depression and being ‘fed up with it’, and one practitioner identified other referrals (from a gambling venue or service provider).

The five problem areas that take up the most of practitioners’ time Figure 2 illustrates that relationship problems, financial difficulties, legal and judicial issues, and lack of self control were the four most frequently cited areas that practitioners identified as taking up most of their time. Secrecy issues and being at crisis point were the next most frequently cited problem areas. Less frequently endorsed problem areas included: committing crime to gamble, fear of losing house/job/status, inability to provide basic necessities, pressure to attend, depression, being fed up with it, and referrals.

Figure 2: Problem areas taking up most of the practitioners’ time - Maori

0

1

2

3

4

5

6

7

Problem Area

Freq

uenc

y

Relationship problems

Secrecy issues & shame

Committing crime to gamble

Financial difficulties

Fear of losing house/job/status

Inability to provide basic necessities Legal and judicial issues

Lack of self control

Pressure to attend

At crisis point

Depression

'Fed up with it'

Referral

83

XGAMBLE Report

‘Successful’ service outcomes Practitioners were asked to define a ‘successful’ outcome in their service. They were also asked to comment on what factors contribute to and hinder successful outcomes.

How to Define? Responses to this item centred on client wellbeing. In particular, practitioners felt that changes made by clients, such as reductions and/or cessation of gambling and improved wellbeing were important. Client satisfaction was also important:

“That it meets the client’s needs or expectations” (#3/M/5/06)

What Contributes? Practitioners differed in their identification of contributing factors. Some practitioners indicated that client awareness and accessibility of services were contributing factors. Other practitioners identified a range of contributing factors, including: client centred, non invasive, non judgemental practitioners, affordable, simple to implement, theoretical model, family involvement, environment of services, and intensity of counselling. As summed up by one practitioner:

“That is based on client needs, is non-invasive, non-judgmental, is affordable, no cost, simple and easy to implement.” (#3/M/5/06)

What Hinders? All practitioners identified clients being unprepared to make changes as a hindrance to successful outcomes. Practitioners also indicated that dealing with other issues, a sole focus on debt reduction, maintaining goals and motivational levels, clients returning to a destructive environment, and the availability of services and resources hindered the achievement of successful client outcomes.

Morale and Satisfaction Each of the Maori practitioners stated that their work was ‘extremely satisfying’.

The multifaceted, flexible nature of the work, and the opportunity to utilise a wide range of skills and work with a wide range of clients was appreciated by all practitioners. For instance:

“The creative, spontaneous approach. Flexibility in work.” (#3/M/5/06)

“The work I do is multi faceted, I am involved in legal, medical, can use kaupapa Maori and taha Maori” (#16/M/6/06)

“Working with people, families, challenge of working with the community and problem gamblers, utilising insight we can apply to our work and to promote a better understanding in the community” (#4/M//6/06)

Although less frequently endorsed, practitioners cited opportunities for self development/ awareness, the availability of a wide range of services, and positive changes in clients as satisfying aspects of their jobs.

84

XGAMBLE Report

The unsatisfying aspects identified by practitioners were the lack of training to help clients who present with difficult situations (e.g. sexual abuse), community perceptions of gambling and their reluctance to change, a lack of resources, bad pay, a lack of agencies working together, too many screens, and an emphasis on outcomes by management and the Ministry of Health. The following quotes illustrate some of these points:

“Community sees problem gambling as entertainment, resistance to change” ( #4/M//6/06)

“The click system (Client management software). The whole thing around outcomes. Managing quality and quantity.” (#3/M/5/06)

“The bums on seats philosophy from Ministry of Health, it’s a numbers game but quality isn’t the same as quality” (#16/M/6/06)

85

XGAMBLE Report

2.3.7 Pacific Island (Niue, Samoan, Tongan) practitioners’ experiences of providing problem gambling services

Practitioners’ basic demographic information A total of six Pacific Island practitioners completed the Phase One Practitioner Semi-structured Interviews. Five of these were face-to-face practitioners (two of which also conducted group based work), and one was a telephone practitioner. The telephone practitioner reported nine years experience in the problem gambling field, and twelve years experience in the addiction field. Overall, the face-to-face practitioners had between seven months and four years work experience in the problem gambling field (with an average of approximately two and a half years), and between seven months and twelve years work experience in the addiction field (with an average of five and half years).

A wide range of qualifications and training were reported by the Pacific Island practitioners, from a Graduate Diploma in Teaching, Diploma in Health Sciences, Diploma in Social Work, Bachelor of Arts, Bachelor of Health Science, Post Graduate Diploma in Business Studies, Masters in Health Sciences (papers in mental health, health promotion, Maori studies, youth development and dual diagnosis), to Gambling Helpline training, alcohol and drug training, addictions training, and supervision training.

As data was collected from only one Pacific telephone practitioner, it should be noted that the remaining sections will report findings for Pacific practitioners as a whole (i.e. findings from the six face-to-face and one telephone interviews will be combined) and no comparisons will be made between face-to-face and telephone practitioners.

The Policy Context, Philosophy and Objectives of the Service Kaupapa/Purpose of your organisation? When asked about the kaupapa/purpose of their organisation, four of the six practitioners’ responses centred on providing counselling and consultation for Pacific people. Other themes included the incorporation of Pacific culture and religion, harm reduction, and providing support to individuals (both people who gamble and family members/significant others) affected by problem gambling. As two participants explained:

“Is a Pacific Island service, families and community, incorporate traditional and modern approaches, aim to reduce harm, based on growth, use the coconut to reflect the image.” (#5/P/6/06)

“Individual and family, wellbeing and addiction, offer culture and religion” (#4/P/6/06)

Key service features The incorporation of Pacific concepts, culture, language and religion were all identified as key features of the practitioners’ services. The ability to match Pacific practitioners with clients according to language, gender and ethnicity were also mentioned:

“Cultural approaches, go out to visit clients, language, cultural awareness, acknowledge

86

XGAMBLE Report

spirituality, gender/ ethnicity sensitivity, accessible to people (different models), provide info, free services.” (#6/P/6/06)

Issues relating to the accessibility of services, in terms of cost and mobility, were also discussed.

Practice Theoretical/therapeutic orientations In total, four theoretical/therapeutic orientations were cited by the Pacific practitioners: Cognitive-behavioural, biopsychosocial, holistic, and eclectic (Pacific counselling/ holistic models i.e. Fonofale). As with practitioners from other ethnic groups, a variety of orientations were employed by each practitioner - the majority (five of the six) cited using cognitive-behavioural and biopsychosocial models, and three also reported using an eclectic model. The following quote illustrates one practitioner’s views of their theoretical/therapeutic orientations:

“Cognitive-behavioural (change their attitudes), Biopsychosocial, other- holistic, the Fonofale model looks at how all these areas affect them and they stick them into time and context” (#5/P/6/06)

Practice – specific techniques and strategies used when counselling clients A wide range of techniques and strategies were reported by the Pacific practitioners. Motivational interviewing was cited by five of the six practitioners, and the wheel of change and reflective listening were also frequently cited. Other specific examples of techniques and strategies included humanities, support role, Fonofale, Ipumahua, Puahauila, Popao, Hanga Tamaki, continuum line, patterns systems thinking, genuine empathy, and crisis modelling

Utilisation of measurement tools when assessing clients A variety of tools were employed by the Pacific practitioners. In particular, each practitioner reported using at least one of the problem gambling screens (SOGS, COGS, DSM IV, Eight Screen). Less commonly reported measurement tools included the wheel of change, alcohol, drugs, mental health audits, money lost, and sense of control.

To what extent is the practitioner-client relationship considered important? Five of the six practitioners indicated that the practitioner-client relationship was ‘extremely important’, and one said it was ‘very important’.

Utilisation of the practitioner-client therapeutic relationship The most frequently cited aspects of the practitioner-client relationship that were utilised were: an emphasis on professionalism, provision of professional support and punctuality. Other relationship aspects included developing mutual respect and building trust, ensuring a safe and comfortable counselling environment, ensuring confidentiality, and enabling feedback between the client and practitioner.

87

XGAMBLE Report

The five most frequent problems that cause problem gamblers or their families/partners to contact a service The problem most frequently cited by Pacific practitioners was financial difficulties: all six practitioners felt that financial issues prompted problem gamblers or significant others to contact a service. Relationship problems, suicide or self harm, everyday functioning being affected, and legal and judicial issues were also frequently referred to. Less frequently cited problems included the safety/wellbeing of significant others, loss of material possessions, domestic violence, secrecy issues and shame, alcohol and drug concerns, affect on mental wellbeing, and access to loans to replace lost money.

The five problem areas that take up the most of practitioners’ time The Figure 3 below demonstrates that financial difficulties were the problem area most frequently endorsed by practitioners as taking up their time (as endorsed by all six practitioners). Relationship problems, issues relating to suicide or self harm, effects to everyday functioning, and legal and judicial issues were also frequently indicated (i.e. by at least half of the practitioners). Less frequently endorsed problem areas included domestic violence, secrecy issues and shame, significant other wellbeing/safety, alcohol and drug concerns, affect of mental wellbeing, loss of material possessions, loan to replace money, and interruptions to everyday life.

Figure 3: Problem areas taking up most of the practitioners’ time - Pacific

0

1

2

3

4

5

6

7

Problem Area

Freq

uenc

y

Relationship problems

Financial difficulties

Suicide or self harm

Domestic violence

Secrecy issues & shame

Affects to everyday functioning

Legal & judicial issues

Significant other safety/wellbeing

Alcohol & drug concerns

Affect on mental wellbeing

Loss of material posessions

Requiring a loan to replace money Interruptions to everyday life

‘Successful’ service outcomes Each practitioner was asked to define a ‘successful’ outcome in their service, as well as to comment on what factors contribute to and hinder successful outcomes.

88

XGAMBLE Report

How to Define? A number of the factors identified as successful outcomes were client centred and reflected changes to gambling behaviour, for example, reducing or stopping gambling, achieving a restored sense of control, achievement of goals by the client, translation of plans/thoughts into actions, and client satisfaction. The following statements illustrate some of these points:

“Client feels happy with the outcomes, does not mean abstinence but it means a decrease in behaviour” (#6/P/6/06)

“Achieved reasonable control, harm is reduced to family and individuals” (#3/P/6/06)

High utilisation of services and a high proportion of clients recovering were also identified as defining factors of successful service outcomes.

What Contributes? Support networks and compassionate, directive practitioners who incorporate culture into a treatment plan were mentioned by two practitioners as contributing factors to a successful service outcome. Other aspects included culturally competent staff, working alongside the community, being responsive to client needs, and intervention intensity. Some of these points are illustrated by the following statements:

“Cultural aspects, having family involved, running in conjunction with the community i.e. church” (#1/P/7/07)

“Being responsive to needs, appropriate framework, work with family, regular meeting, client-counsellor relationships” (#3/P/6/06)

What Hinders? Workforce capacity and access to good staff, organisational characteristics, lack of resources, and funding contracts were identified by a number of the Pacific practitioners as factors that hinder successful outcomes in their services:

“Work force capacity, organisational characteristics, government contracts renewed and staff/ clients turn over” (#1/P/7/07)

“Resource and training (mostly Western), need for practical training” (#4/P/6/06)

“Not enough resources, not trained up workforce, need more Pacific Island services, not enough money, research on Pacific gambling is good” (#6/P/6/06)

Other factors included Pacific perceptions of gambling, including clients unwillingness to engage, and a lack of governmental restrictions on gambling.

Morale and Satisfaction Half of the Pacific practitioners indicated that their work was ‘extremely satisfying’; the other half indicated that it was ‘quite satisfying’.

A wide range of work aspects were cited as satisfying by practitioners. Some of these related to working with Pacific populations, including: the development of Pacific

89

XGAMBLE Report

methods, the ability to mould current Pacific models, utilising and incorporating culture into work, working with ‘own’ people:

“Being a helper to our people” (#4/P/6/06)

“Working with Pacific, learning about impacts - positive and negative, learning and using traditional and western theories and finding new ways to beat the challenges” ( #5/P/6/06)

“Identify needs and gaps, develop model that fits with Pacific people, able to mould our own models” (#6/P/6/06)

Other satisfying aspects related to the practitioners personal satisfaction, client/ practitioner relationships, witnessing positive change in clients and being an agent for change. For example:

“Knowing that you have helped” (#1/P/7/07)

“Being there to help, have knowledge and confidence and experiences to support and provide treatment options” (#2/P/6/06)

“Seeing people recovering and changes for good” (#3/P/6/06)

Practitioners also identified a number of unsatisfying aspects to their work, including: a lack of resources (particularly manpower), large workloads, organisational/funding pressure, the narrow scope of funding, and political and racial implications. For instance:

“Small Pacific Island team, limited resources” (#2/P/6/06)

“Stress, lack of resources- workers and cares, stretched to limits, many areas to work on but we are paid for treatment (gambling related), work retention is low because of workload” (#5/P/6/06)

“Political and racial stuff, old does not respect the new” (#3/P/6/06)

Client issues such as an unwillingness to change and clients failing to attend appointments were also cited as unsatisfying job aspects.

90

XGAMBLE Report

2.3.8 Asian (Chinese and Korean) practitioners’ experiences of providing problem gambling services

Practitioners’ basic demographic information In total, four Asian practitioners completed the Phase One Practitioner Semi-structured Interviews: three face-to-face practitioners (one of which was also doing some group based work, and another who was doing some telephone counselling), and one telephone practitioner. The telephone practitioner reported two years experience in both the problem gambling and addiction fields, and the face-to-face practitioners had between one and approximately four (3.8) years work experience in both the problem gambling and addiction fields (with an average of 2.4 years).

The following wide range of qualifications were reported: Graduate Certificate in Group Supervision, Diploma of Social Work, Bachelor in Psychology, Bachelor in Engineer Chemistry, Bachelor in Theology, Bachelor of Sciences, Youth Ministry Degree, Graduate Diploma in Psychosocial Studies, Postgraduate Diploma in Guardian Studies, Postgraduate Diploma in Economic Law, Masters in Social Work, Masters in Divinity, Masters in Psychotherapy.

As data was collected from only one Asian telephone practitioner, it should be noted that the remaining sections will report findings for Asian practitioners as a whole (i.e. findings from the three face-to-face and one telephone interviews will be combined) and no comparisons will be made between face-to-face and telephone practitioners.

The Policy Context, Philosophy and Objectives of the Service Kaupapa/Purpose of your organisation? When asked to describe the kaupapa/purpose of their organisations service, all of the Asian practitioners stated harm reduction:

“Eliminate harm of gambling in the community” (#4/A/6/06)

“Minimise gambling harm by delivering Telephone and Face-to-face counselling services to public.” (#3/A/6/06)

Advocacy services, offering a range of services to a range of populations, and working with government to shape policies and practices were also mentioned. As one practitioner stated:

“Work with government agencies to make effective policies/ rules by providing/ participating in conference, including Ministry of Health which is essential. Advocacy: self-bar, self-help to protect clients from suffering further harms” (#3/A/6/06)

Key service features When asked about the key features of their service, a number of service elements were discussed, including: the need for multi-skilled approaches, the incorporation of public and mental health, cultural awareness, commitment to the work, building clients self

91

XGAMBLE Report

confidence, and holistic approaches for Asian clients. The following quotes illustrate some of these points:

“Culturally appropriate. Native Asian languages - deep down to spiritual and mental level, not just gambling issues which are only the tip of iceberg. Holistic approach: spiritual/ spiritual dimension/ life skills/ stress management skills.” (#3/A/6/06)

“Strengths, in terms of the number of staff, service coverage, local qualifications. The working team successfully combines every worker’s strengths, characteristics, and working experiences in mental health, counselling and social work.” (#4A/6/06)

Practice Theoretical/therapeutic orientations The Asian practitioners referred to a number of theoretical/therapeutic orientations. These included: cognitive-behavioural therapy, Gestalt therapy, structural family therapy, a biopsychosocial approach, problem solving, an eclectic approach, and transitional analysis.

Practice – specific techniques and strategies used when counselling clients When asked to provide specific examples of the techniques and strategies used by practitioners when counselling clients, the most frequent responses were desensitisation and reflective listening. Other examples included relaxation techniques, visualisations, preparing the client to get involved, practise how to stop, combining culture and problem, drawing, communication skills, and flexibility in working with clients. The following examples illustrate some of the Asian practitioners’ use of desensitisation:

“Ask clients to do some visualisation exercise, imagine the light in Casino from bright to dark, casino is like a tiger there to bite them.” (#2/A/6/06)

“You are going to a gambling venue, when you touch the pokie machine, what do you feel?” (e.g. the faces of casino manager, other gamblers, cameras around casino, review… gambling is not only between you and the machines). Prepare-> prevent ->get involved -> practice how to stop.” (#3/A/6/06)

Utilisation of measurement tools when assessing clients The SOGS was the most frequently used measurement tool to assess clients. A number of practitioners also assessed for depression/suicide, anxiety, coping skills, money lost from gambling, and working with the family.

To what extent is the practitioner-client relationship considered important? Each of the Asian practitioners indicated that the practitioner-client relationship is ‘extremely important’.

Utilisation of the practitioner-client therapeutic relationship The Asian practitioners described utilising the following aspects of the practitioner-client therapeutic relationship: gaining trust and helping clients feel comfortable with the counselling environment, convincing clients of the usefulness of counselling, building self esteem, respecting client’s rights, and deciding the extent of the problem and exploring interventions. For instance:

92

XGAMBLE Report

“Convincing skills: Should be able to provide information for them; to believe to achieve their goals. Human approach: no judge, just a listener.” (#3/A/6/06)

“Respect clients, treat them like friends, but show the profession of a counsellor; Establish clients’ self-esteem; Talk with them with a soft tone; make them feel not being criticised.” (#1/A/6/06)

The five most frequent problems that cause problem gamblers or their families/partners to contact a service A wide range of problems were cited by practitioners in response to this question. A number of these were specifically concerned with aspects of gambling behaviour: concerns with gambling behaviour, loss of control, chasing losses in the hope of winning, financial difficulties, self-barring, shame and secrecy, and interruptions to everyday functioning (e.g. work, study). Familial issues were also frequently mentioned, such as, family problems, concerns about children’s welfare, family relationships, and family referral.

Issues relating to post-immigration settlement adjustment were also seen as a frequent problem causing clients to present. For example, issues relating to a lonely life in New Zealand with a lack of activities and places to go (things to do), an unfamiliar culture, and a mismatch of expectations for life in New Zealand.

The five problem areas that take up the most of practitioners’ time As illustrated in Figure 4, understanding/assistance, relationship problems, and financial difficulties were the most frequently cited problem areas for taking up most of practitioners’ time. Lack of accountability, disruption to everyday life, mental wellbeing, self confidence, problem solving, managing time/money, preventing relapses, and loneliness were all less frequently cited.

Figure 4: Problem areas taking up most of the practitioners’ time - Asian

0

1

2

3

4

5

6

Problem Area

Freq

uenc

y

Financial difficulties

Relationship problems

Lack of accountability

Disruption to everyday functioning

Affect on mental wellbeing

Understanding/assistance

Self confidence

Problem solving

Managing time/money

Preventing relapses

Loneliness

93

XGAMBLE Report

‘Successful’ service outcomes Practitioners were asked to define a ‘successful’ outcome in their service. They were also asked to comment on what factors contribute to and hinder successful outcomes.

How to Define? Responses to this item can be sorted into two categories. The first relates to gambling behaviour: control of gambling, stopping gambling, clients understand their gambling behaviour, and clients goals being met. The second relates to other aspects of life, in particular, increases in self confidence and improvement of relationships.

What Contributes? Various intervention aspects were identified as contributing to ‘successful’ outcomes. These included the use of theoretical models (as detailed earlier), the practitioner-client relationship, the intensity and timing of the intervention, and teaching clients to manage behaviours and relationships. Two individually based client factors were also mentioned: increases in self confidence and personal growth.

What Hinders? Practitioners identified a number of hindrances to ‘successful’ outcomes. The first related to the need for a more holistic approach to be utilised and the recognition that different treatment modalities are utilised in other organisations. The second, related to clients lacking a commitment to achieving change, through either an unwillingness to change, or a focus on different outcomes (e.g. a quick fix for financial problems):

“Clients are not willing to change: they do not believe they have any problem with their gambling. Urgent need: clients focus on finding money, but not solving the problems.” (#1/A/6/06)

Another set of issues that were discussed by a number of the practitioners included the role of advertising in enticing people to gamble, and the inadequacy of monitoring systems (i.e. host responsibility) in gambling venues such as casinos and bars. For example:

“[X] does not play its role well as a responsible host: self-barred clients/ problem gamblers can still access the venue. Casino advertisements lure people to gamble and convey a strong message that gamblers can win. Bars, gambling venues do not have strict supervision.” (#1/A/6/06)

“The rights of gamblers are not clearly stated/not been seen by clients. They should be written down on the board in Casinos, so that gamblers can see). Monitoring system: Host responsibility.” (#3/A/6/06)

Morale and Satisfaction A range of responses were received to this question: two of the practitioners stated that their work was ‘extremely satisfying’, one responded that it was ‘quite satisfying’, and one indicated that their work was ‘quite satisfying’ to ‘neutral’.

Satisfying aspects of the work included agency supported training programmes, good team relations, a multi-disciplinary team environment, successful improvements, personal achievements, the opportunity to match discipline and practice, and the relationships formed between client and practitioners. For instance:

94

XGAMBLE Report

“Training programmes are fully supported by the agency, in terms of tuition fees, and study time. My discipline matches my practice, the effectiveness is good.” (#2/A/6/06)

“Good team environment, every one is with passion and commitment, and supports each other. Multi-discipline: combining research and practice.” (#4/A/6/06)

Unsatisfying aspects centred on organisational factors, such as, a lack of support and respect when suggestions are made, lack of advancement in job position, heavy workloads and hard work, no clear practice guidelines, and a lack of clear job outlines.

95

XGAMBLE Report

2.3.9 Results from the Clinical Tasks Analysis (Problem Gambling) (CTA) (PG)

Background The CTA (PG) was administered to problem gambling practitioners in New Zealand to document their tasks in counselling practice as part of the Phase One of the study. This sort of functional job analysis covers a range of analytic procedures used to describe human service work in terms of tasks or activities and has proven to be a useful tool for administrators and trainers because it quantitatively describes existing tasks, which can be used to assess the fit between practice and purpose and to assess the utility of training programmes.

Scoring method The scoring system for the CTA (PG) required practitioners to rate the frequency of performance of their tasks and simultaneously rate the importance of each of those tasks. Respondents were asked to rate the task statements in relation to their current practice in the course of a typical month, using a five-point scale, ‘How often’ they performed each task, from ‘not done’ (scored as 1) to ‘almost always’ (scored as 5). They were then asked to rate the task statements as ‘How important’ they are to their job. Responses range from ‘not important’ (scored as 1) to ‘extremely important’ (scored as 5). The 107 tasks span a number of practice areas, including: assessment; treatment goals; general interventions; gambling interventions; family interventions; interventions for related problems; referral; education; and research/policy.

Results A descriptive analysis of the data is presented in Table 7.

Table 7: Frequency of performing counselling tasks and importance attached to those tasks as per the CTA (PG) (n= 27)

Items Mean Freq

SD Freq

Mean Imp

SD Imp

% Freq. Done

% Very Imp.

1. Determine the urgency or risk in the individual’s situation in order to decide if emergency services or routine handling and referral are required.

4.48 .80 4.81 .58 81.5 100

2. Observe individuals and gather information from appropriate sources in order to decide whether there is a need for specialist counselling or mental health treatment.

3.74 1.40 3.85 .76 70.4 78.0

3. Interview client/family in order to gather information including gambling activities as part of a psychosocial assessment or to compile a social history.

3.93 1.27 3.93 .91 70.4 78.0

4. Assess specific aspects of client and/or family life in order to determine the need for mental health or medical services.

4.15 .72 4.33 1.07 81.5 96.3 5. Assess clients in order to determine

eligibility for service and/or referral where appropriate.

4.48 .65 4.42 1.33 92.0 88.5

96

XGAMBLE Report

6. Interview people, review applications and/or complete paperwork in order to determine initial or continued eligibility for services or financial help.

3.88 1.18 3.85 .99 73.1 73.1

7. Obtain information from individuals, their relatives or significant others in order to carry out admission or intake procedures for treatment or services.

3.58 1.39 3.81 .97 65.4 73.1

8. Ask specific questions relating to gambling when a client presents with symptoms of depression, insomnia, anxiety, and/or family conflicts without any mention of gambling.

4.20 1.26 4.40 .83 88.0 92.0

9. Gather information so as to assess whether client has other addictions as well. 4.22 .75 4.19 1.14 89.0 74.1

10. Observe individuals and gather information from appropriate sources in order to establish the existence of substance/alcohol abuse problems.

3.46 1.45 3.50 1.35 57.7 57.7

11. Gather information on gambling when evaluating other addictions affecting individuals and families.

4.00 1.22 4.16 1.52 76.0 80.0 12. Ascertain what functions gambling plays in

the client’s life. 4.81 .40 4.85 1.18 100 100 13. Identify high risk situations in order to

assess when gambling is most likely to occur.

4.85 .36 4.85 .76 100 100 14. Evaluate addictive behaviours including

substance abuse, eating disorders or relationship addictions of gambler’s partner in order to assess multiple addictions in families.

4.04 .27 3.93 1.39 59.3 66.7

15. Inquire as to the drinking, gambling and/or drug use of each parent to assess family patterns of addictions.

3.63 1.24 3.70 1.29 51.9 51.9 16. Where either or both partners are found to

have a sexual dysfunction or a sexual addiction, obtain more detailed sexual histories in order to refer to a sex therapist.

2.15 1.23 2.56 1.30 14.8 22.2

17. Interview client regarding employment in order to assess work impairment issues. 3.63 1.21 3.59 1.61 55.6 59.3

18. Use screening information in order to diagnose problem gambling and provide treatment or refer client to other sources.

4.56 .93 4.63 1.32 88.9 88.9 19. Enable the client to recognise the problem

and address it openly and positively within an established and trusted relationship.

4.70 .46 4.89 1.52 100 100 20. Discuss treatment options with individuals

in order to help them understand choices and/or resolve a particular problem.

4.59 .80 4.63 1.25 88.9 92.6 21. Develop individual service plans in order to

reflect the client’s goals, strategies and time frames to achieve these goals, and evaluation of these outcomes.

4.30 .91 4.37 1.50 88.9 88.9

22. Enable the client to analyse the benefits of continuing to gamble versus the costs of giving up gambling.

4.59 .64 4.67 .97 92.6 96.3 23. Ascertain whether abstinence or controlled

gambling is the goal of the gambler. 4.52 .70 4.52 1.21 88.9 88.9 24. Talk with individuals and/or relatives about

problems in order to reassure, provide support, or reduce anxiety.

4.27 1.00 4.22 1.0 80.8 81.5 25. Encourage and help people to discuss their

points of view, feelings, and needs in order to establish open and trusting relationships.

4.63 .49 4.67 1.30 100 96.3 26. Express and demonstrate and understanding

of peoples’ point of view, feelings, and needs in order to establish open and trusting relationships.

4.70 .54 4.67 1.26 96.3 92.6

97

XGAMBLE Report

27. Confront people about unacceptable behaviour in order to reassure, provide support, or reduce anxiety.

3.81 .83 4.07 1.17 63.0 85.2 28. Develop culturally appropriate intervention

plans for clients in order to meet the needs of individual clients.

4.41 .75 4.70 1.31 85.2 100 29. Use specific intervention techniques with

individuals in order to improve behavioural functioning and adjustment.

4.37 .63 4.56 1.32 92.6 96.3 30. Use a specific therapeutic method in a

group situation in order to improve the adjustment and functioning of the group members.

3.96 6.38 3.19 1.22 44.4 59.3

31. Advocate on client’s behalf in a manner that encourages self-determination and self-reliance.

3.52 1.42 3.85 .89 63.0 70.4 32. Negotiate arrangements in relation to clients

within the agency and with relevant agencies in order to ensure effective case management

3.38 1.10 4.81 .85 46.2 76.9

33. Work with individuals, family members, or significant other, in order to prepare them psychologically and socially for movement from one living arrangement to another.

3.65 1.06 3.85 1.42 57.7 69.2

34. Work with individuals, family members and significant others in order to prepare them for termination of treatment, services, or financial help.

3.58 1.21 3.93 1.49 61.5 65.4

35. Emphasise labelling client as a “problem gambler” in order to enable the client acknowledge that he or she problem with gambling.

3.85 1.41 4.33 1.50 77.8 77.8

36. Challenge client’s beliefs regards personal luck in order to get their thinking straight on the amount of money lost.

4.33 .68 4.42 .94 88.9 88.9 37. Suggest strategies to overcome the impulse

to gamble when confronted with gambling related stimuli.

4.59 .50 3.85 .91 100 96.3 38. Teach client to set limits regarding number

of gambling sessions in order to gain control of gambling.

3.70 1.07 3.81 1.16 48.1 59.3 39. Instruct clients to stop gambling when a

predetermined amount of money has been lost.

2.74 1.06 4.81 1.35 18.5 48.1 40. Teach client to set limits regarding length of

gambling sessions in order to gain control of gambling.

3.37 1.24 3.85 .69 44.4 55.6 41. Assist gambler to find new friends who do

not gamble in order to enable attention and energy be directed away from gambling.

4.74 .45 3.93 .98 100 96.3 42. Teach clients how to overcome “action” or

arousal or excitement stage by encouraging the client to take breaks between sessions of gambling.

3.81 1.07 4.33 .93 70.4 74.1

43. Advise clients that chasing ones losses in an attempt to get even will only produce irrational gambling and further losses.

3.15 1.38 4.42 .75 42.3 53.8 44. Provide advice on how to overcome the

need to raise money without resorting to gambling.

4.59 .57 3.85 1.50 96.3 88.9 45. Suggests strategies for coping with sadness

or depression that do not involve gambling. 4.00 .78 3.81 1.26 77.8 74.1 46. Develop with client ways to celebrate

occasions that do not involve gambling or gambling venues.

4.37 .74 4.40 1.49 92.6 96.3 47. Teach client to set limits on the amount of

money spent in order to gain control of gambling.

4.00 .78 4.19 1.50 77.8 81.5

98

XGAMBLE Report

48. De-emphasise labelling client as a “problem gambler” in order to draw upon the client’s awareness of the problem and his or her preparedness to change.

3.65 .94 3.50 1.18 57.7 69.2

49. Advise clients to limit alcohol and drug consumption while gambling in order to gain control over gambling.

3.81 1.04 4.16 1.29 55.6 66.7 50. Teach clients self-monitoring skills in order

that they may have greater control when gambling.

3.33 1.24 4.85 1.01 44.4 63.0 51. Instruct clients to keep a gambling diary of

all monies spent and won from gambling in order to keep a record of the amount of money spent gambling.

4.08 .74 4.85 1.23 76.9 76.9

52. Teach clients to reward themselves when they maintain their treatment goals. 3.37 .79 3.93 1.47 40.7 51.9

53. Teach clients various relaxation techniques like muscular relaxation training, exercise and yoga in order to keep to keep the urge to gamble under control especially when they feel tense and irritable.

4.30 .72 3.70 1.53 85.2 85.2

54. Encourage clients to change their gambling from games of luck to forms of games involving a degree of skill so as to minimise losses.

3.42 1.14 2.56 1.24 46.2 53.8

55. Teach client optimal strategies in order to maintain gains in the longer term. 2.08 1.09 3.59 1.12 7.7 19.2

56. Teach clients problem-solving skills which do not involve gambling. 3.08 1.50 4.63 1.44 44.0 52.0

57. Make clients aware of the possibility of relapse and suggest strategies to adopt should relapse occur.

3.96 1.10 4.89 1.04 76.0 80.0 58. Interview each partner alone in order to

establish a therapeutic bond with each. 4.38 .80 4.63 1.26 88.5 92.3 59. Use specific intervention techniques to

work with family members, individually or as a group, in order to strengthen the family as a unit.

2.64 1.35 4.37 .99 24.0 44.0

60. Facilitate emotional healing between partners in order to assist the gambler to become more emotionally sensitive to partner’s feelings and needs.

3.16 1.34 4.67 1.08 40.0 72.0

61. Resolve interpersonal conflicts through marital, family or group therapy in order to rebuild trust that has been destroyed.

3.36 1.19 4.52 .71 44.0 76.0 62. Provide individual and family counselling

so as to address the problem of gambling openly within an established and trusted relationship.

3.00 1.47 4.22 1.21 40.0 64.0

63. Assess and intervene directly in the dysfunctional couples system in order for a significant change to occur.

3.73 1.34 4.67 1.53 61.5 76.9 64. Facilitate partners, other family members

and significant others to confront the gambler.

2.96 1.55 4.67 1.25 37.5 56.5 65. Include the partner and key significant

others as early as possible in the assessment of a gambling problem.

3.00 1.12 4.07 1.50 32.0 54.2 66. Teach partners ways so as to reinforce non-

gambling behaviours. 3.08 1.52 4.70 .97 44.0 64.0 67. Encourage partners not to nag the gambler

about their gambling as nagging is counter productive.

3.80 1.00 4.56 1.21 72.0 80.0 68. Teach partners appropriate responses so as

not to reinforce gambling behaviours. 3.52 1.12 3.19 1.00 56.0 76.0 69. Analyse case background, consult with

appropriate individuals, in order to arrive at a plan for services and/or financial help.

3.68 .94 3.85 1.30 60.0 83.3

99

XGAMBLE Report

70. Provide specialist problem gambling financial counselling to individuals and families.

3.48 1.30 4.81 1.26 52.0 72.0 71. Assess client’s financial situation in order to

provide information regarding the legal implications of debt and government assistance, budgeting and negotiation with debtors.

2.92 1.35 3.85 1.17 42.3 65.4

72. Teach clients about money and budgeting in order to develop skills in the management of personal finances.

3.28 1.24 3.93 1.31 48.0 76.0 73. Develop a realistic family budget and a plan

for financial restitution in order to enable the family to gain or regain control over finances.

3.23 1.42 4.33 1.32 50.0 73.1

74. Discuss with client the need to make restitution for incurred debts in order to teach gambler responsibility for his/her indebtedness including avoidance of further bailouts.

2.48 1.42 4.42 1.22 28.0 60.0

75. Suggest spouse assumes responsibility for family checking and savings accounts in order to avoid financial crises taking place.

3.42 1.24 3.85 .89 57.7 64.0 76. Suggest to clients that access to cash must

be limited in order to restrict gambling. 4.04 .96 3.81 .85 84.6 92.3 77. Advocate for individuals in order to

persuade others that those people do qualify for services or financial help.

4.58 .58 4.40 1.42 96.2 96.2 78. Start a legal process in order to protect the

rights of an individual. 3.04 1.43 3.48 1.49 40.0 56.0 79. Testify or participate in court hearings in

order to provide information on which legal decisions can be based.

2.00 1.26 2.72 1.50 16.0 32.0 80. Make contact with other units/agencies, by

letters, memos, or phone calls, in order to refer people to appropriate services.

1.67 1.20 2.36 .94 12.5 28.0 81. Put individuals in touch with people of

similar backgrounds, cultures or ethnicity in order to make a move or change easier for them.

3.58 1.24 4.19 .91 61.5 84.6

82. Establish contact between the service and local provider networks in order to create linkages and referral networks for people affected by problem gambling.

3.81 1.13 4.23 1.16 61.5 84.6

83. Provide assessments and reports to referring agencies and courts. 3.85 1.35 4.15 1.35 69.2 80.8

84. Refer individuals with gambling issues to GA and Gam-Anon recovery programs. 2.96 1.31 3.64 .69 38.5 68.0

85. Refer client to a general practitioner or psychiatrist for treatment of depression or mood disorders.

3.42 1.24 4.35 .98 57.7 88.5 86. Provide client with a self help manual in

order to gain a better understanding of problem gambling.

3.27 1.43 4.00 .93 46.2 73.1 87. Refer client to medical practitioner for

treatment of health problems that are caused or exacerbated by gambling.

3.20 1.41 4.12 .75 52.0 84.0 88. Educate spouse, parents, children and other

family members about gambling. 3.81 .89 4.38 1.50 65.4 84.6 89. Develop a marketing strategy in order to

publicise and promote the gambling counselling service.

2.73 1.51 3.62 1.26 42.3 69.2 90. Educate individuals and groups of people

about legal issues relating to debts. 2.68 1.21 3.31 1.49 24.0 46.2

100

XGAMBLE Report

91. Develop and initiate regional skill development workshops regarding problem gambling for health and community services industry personnel in order to identify and respond to the needs of problem gamblers.

2.08 1.32 3.16 1.50 16.0 48.0

92. Deliver education programs on problem gambling to gambling facilities in order to assist the gaming industry to respond to the needs of problem gamblers.

2.04 1.37 3.48 1.18 20.0 68.0

93. Deliver community education programs to the general public and the gaming industry in order to promote an awareness of problem gambling.

2.80 1.22 3.84 1.29 28.0 72.0

94. Collate and distribute information regarding gambling trends and problem gambling in order to maximise the impact of community education.

2.92 1.47 3.80 1.01 40.0 76.0

95. Promote the problem gambling counselling service amongst the gaming industry, service providers and the community in order to publicise its existence.

3.48 1.19 4.12 1.23 56.0 84.0

96. Explain service programs and policies to people in public appearances of various kinds in order to inform the general public about issues and programs.

2.92 1.47 3.71 1.45 41.7 70.8

97. Attend Financial Counselling Induction training and regular in-service training with the Financial and Consumer Rights Council.

2.64 1.60 3.48 1.53 40 56

98. Respond to request or inquiries from the community regarding the Financial Counselling Program.

2.25 1.39 3.21 1.23 25 50 99. Gather and compile data about services

provided to people in order to prepare statistics for periodic reports.

3.38 1.28 3.83 1.12 41.7 66.7 100. Review and analyse data about service

needs and demands in order to establish workload and staffing requirements.

2.60 1.26 3.60 1.45 28.0 52.0 101. Take part or direct studies or research

projects in order to increase the knowledge base of social work either in education or service provision.

3.00 1.47 3.44 1.04 36.0 56.0

102. Liaise with other services in the regional network in order to compare and analyse agency trends or issues in relation to problem gambling.

3.36 1.32 4.08 1.26 48.0 80.0

103. Develop, maintain and analyse client data consistent with the state wide format, and anticipate in any research projects regarding the service model.

2.74 1.29 3.70 .99 26.1 65.2

104. Contribute to the agency’s policy development, planning, monitoring and evaluation of the problem gambling counselling service.

3.38 1.41 4.13 1.08 54.2 79.2

105. Work with gambling facilities in order to assist them in developing policies, practices and procedures to deal with problem gamblers.

3.40 1.35 4.08 .71 56.0 76.0

106. Identify and promote best practice models for problem gambling programs in order to provide an accountable, professional and accessible service to clients.

3.16 1.25 4.20 .78 40.0 84.0

107. Do you explore the place the Treaty of Waitangi has in gambling in New Zealand. 3.52 1.33 3.84 1.21 52.0 68.0

101

XGAMBLE Report

• The results overwhelmingly indicate that problem gambling practitioners ‘frequently to almost always’ engaged in direct practice tasks and considered these tasks as ‘very important to extremely important’ to their practice, in terms of:

o Assessment of risk, health, mental health, other addictions, role of gambling and its impact on the problem gambler and their families (see items 1, 4, 5, 8, 9, 11, 12, 13, 18). It is worth noting that the item on assessment of sexual history (item 16) was rated as very low in both practice (15%) and level of importance (22%);

o The establishment of treatment goals, considering treatment options for the clients in discussion with them so that they may take ownership of their own treatment (see items, 19, 20, 21, 22, 23);

o Using general interventions to assist clients with a number of issues, to deal with their anxiety, feelings, and needs (see items, 24, 25, 26, 28, 29); and,

o Using gambling specific interventions (see items, 36, 37, 41, 44, 46, 47, 51, 53), such as challenging clients’ beliefs, suggesting strategies to overcome the urge to gamble and the use of self-help strategies to minimize harm from gambling and to promote healthy behaviour.

• Approximately 70% to 75% of problem gambling practitioners taught their clients how to set limits to control their gambling, however, only 18.5% frequently ‘instructed their clients to stop gambling when a pre-determined amount was lost’ (item 39), while 48% of practitioners considered this task to be extremely important; and only a rather low percentage of practitioners focused on ‘teaching client optimal strategies in order to maintain gains in the longer term’ (item 55) - level of practice (8%) and importance (20%).

• Around 75% of problem gambling practitioners thought it was ‘very important to extremely important’ to their practice to use specific family interventions which focused on the development and building of relationships and healing within families (see items, 59-67), but their frequency in providing family-related intervention or support in their current practice was occasionally used or varied.

• The results indicate that close to 50% of practitioners frequently provided financial counselling (see items, 70-73) although most believed that it is an important task in their counselling practice as a problem gambling practitioner. One probable explanation for only 50% ‘frequently’ providing financial counselling was that practitioners referred their clients to specialised financial advisors. Interestingly, from a professional development perspective, of these practitioners only 40% had attended a Financial Counselling Induction course (see item 97) while 56% considered this to be very important to their continuing practice.

• Practitioners felt that being involved in legal processes was ‘only somewhat important’ to their role (see items, 78 & 79), although other key tasks related to advocacy for clients in need were provided and considered important by 96% of the problem gambling practitioners (see item 77).

• Tasks relating to referral (see items, 80-85, 87) were considered to be very important by a majority of the problem gambling practitioners and frequently done by almost two-thirds of problem gambling practitioners.

• Tasks relating to education (see items, 88-98) at an individual level appeared to be rarely completed but these tasks were considered to be ‘very important’ by most

102

XGAMBLE Report

practitioners at a community/key stakeholder level. This disparity between practitioners beliefs (i.e. they see the value of health promotion activities) and level of activity (i.e. they are not actively engaging in these areas) could be attributable to a number of reasons, including: 1) they did not have the time to complete this important aspect of work; or 2) they did not necessarily see it as their role; or 3) they did not see themselves having the skills to carry out those activities.

• Similarly, tasks relating to research/ policy (see items, 99-107) indicate that they are considered very important by most problem gambling practitioners but are only ‘frequently completed’ (by 25% to 50%) as part of their role. This reveals a polarity in the practice of problem gambling practitioners in terms of their lack of involvement in collecting and analysing data/information (e.g. item 100 - “Review and analyse data about service needs and demands”; item 103 - “Develop, maintain and analyse client data consistent with the state wide format, and anticipate any research projects regarding the service model”) at a research, policy, and community level.

103

XGAMBLE Report

2.4 Phase One Discussion

2.4.1 Client experiences of problem gambling services After reviewing the results section of the Phase One data, it becomes apparent that the four population groups have more similarities than differences in relation to the topics under investigation. Most of the differences were a matter of different emphasis paid by participants from the four population groups rather than were fundamental differences.

With regard to the motivating factors for gambling, three main themes were evident across all ethnic groups (European, Maori, Asian and Pacific), people gambled: 1) for fun/excitement; 2) for making/winning money; and, 3) for escapism. Most participants in the present study reported that gambling is a satisfying and enjoyable activity. Large numbers of participants said they gambled to win money or think/dream about winning. The generation of these positive mood states may well be a major reason many people continue to gamble despite being aware that they are likely to lose. In the 1999 New Zealand national survey, people with problem gambling reported excitement and relaxation while gambling much more often than people without problem gambling (Abbott, 2001). They also mentioned gambling to escape more often when feeling depressed. These findings are consistent with the view that using gambling to relieve negative emotional states is a significant factor in the development of problem development (Abbott, 2001; Blaszczynski & McConaghy, 1989). Some research suggests that moods also influence the choice of gambling activity, for example, anxious gamblers have been found to favour gaming machines while depressed gamblers favour forms involving greater skill and/or social interaction (Coman, Evans & Burrows, 1996). Other research has found that prior negative mood states contribute to regular gamblers continuing to gamble despite repeated losses and affect gambling decision-making, for example, depressed mood increases high-risk/high-reward choices (Raghunathan & Pham, 1999). Although gambling may act as a form of escape or distraction from a negative mood state, people with problem gambling also frequently report feeling depressed after losing and feeling guilty after completing a gambling session (Abbott, 2001). This suggests that people who are at-risk and people with problem gambling may often get caught in a circular process whereby they gamble to reduce negative mood states that, over time, increasingly result from their gambling behaviours, losses and associated adverse consequences.

Consistent with earlier work in New Zealand (Tse et al., 2005) and Australia (Blaszczynski, Huynh, Dumlao & Farrell, 1998), there were some differences across the four population groups in terms of other reasons for gambling. They included hope for a better life and chasing losses (mentioned more often by European participants), and enticing messages from the gambling industry (mentioned more often by Asian participants).

Social aspects of gambling were raised by Asian, Maori and Pacific participants but members from these population groups offered slightly different explanations. Asian participants referred to gambling with friends/peers, while Maori and Pacific clients

104

XGAMBLE Report

emphasised the role of family in gambling behaviour, including an indication that they were often introduced to gambling by family members. Pacific people also indicated that they saw gambling as a means of bringing money into the family and meeting household obligations (Bellringer, Perese, Abbott & Williams, 2006; Perese & Faleafa, 2000).

When asked to define problem gambling, matters relating to the financial aspects of gambling (not managing financially, spending more money than intended, chasing losses) were mentioned by each ethnic group. These findings were consistent with previous research in New Zealand (Clarke, Tse, Abbott, Townsend, Kingi & Manaia, 2007; Dyall, 2002; Tse et al., 2005). In addition, European, Maori, and Pacific clients all mentioned the effects of gambling on overall wellbeing/other aspects of life. Family was a particularly important area, with participants stating that problem gambling ruins relationships and results in dysfunctional families. Borrowing money, being unreliable and dishonest (including lying to your family) were also cited as defining factors of problem gambling. Significant others, from all cultural groups, emphasised unreliable behaviour (lying, deceit, and wasting money), the gambler believing they can win and personality changes which were captured by Pacific participants in particular. These findings correspond with those from focus group discussions held over a period of 16 months with 39 women in America who identified a range of negative health and social consequences (Heineman, 1992; 1987). Heineman’s work has identified the effects of problem gambling in spouses, siblings, children and parents and include: threatening phone calls, dealing with loan sharks, issues relating to trust, shortage of self-help groups or where to seek help, being overworked to repay gambling debts, dysfunctional relationships, loss of stable family incomes, experiencing a sense of neglect, and in some cases reporting of violence and abuse (Heineman, 1992; 1987).

There were noticeable differences across the four population groups in relation to the topic of reducing access to gambling and improving host responsibility. Participants were asked about ways to protect people against problem gambling. In particular, European & Maori participants suggested banning machines, modifying machines, removal of full access to machines and money, self exclusion, and better provision of information on gambling. Maori and Asian participants tended to call for venues to make information on gambling readily available, and for venues to be held accountable if they were not complying with statutory requirements. Pacific and Asian participants also felt that the process of self-exclusion from venues needed to be enhanced. The availability of information was another theme to emerge in this section. Maori and Pacific clients all felt that better awareness and education around problem gambling issues was needed. Maori suggested that information should be distributed through a variety of channels (e.g. television, Marae, schools) and mediums (e.g. pamphlets, advertisements, preventative efforts with school children). Previously, Dyall and Hand (2003) have argued the pressing need for a comprehensive Maori public health programme to minimise the disproportionate harms caused by gambling to indigenous people. Raising awareness amongst family members/communities, and encouraging them to seek help and support those with gambling problems was also endorsed by these groups.

105

XGAMBLE Report

The aim of seeking help for all population groups was very similar; to enable them to gain control of, or stop their gambling. The aims of European significant others seeking help included providing support for the problem gambling client and gaining information. Pacific significant others aimed to help the gambler rebuild trust and relationships within family and the wider social circles. These aims are consistent with those indicated in previous New Zealand research (Clarke, Abbott, DeSouza & Bellringer, 2007).

Regardless of ethnic group, the realisation of the harm caused by gambling motivated most clients to seek help. For some clients this was related to financial and life pressures, losing control of gambling, dishonest behaviour (including lies and secrecy), and abuse. Another very strong theme was the role of family in encouraging help seeking – all population groups cited issues such as the wellbeing of children, pressure from family, disintegration of key relationships, and encouragement by family/friends to seek help. These findings are very consistent with the results emerging from international studies (e.g. Hodgins, Makarchuk, el-Guebaly, & Peden, 2002; Jackson, Thomas, Holt & Thomason, 2005; Tang, Wu & Tang, 2007). Two differences were noted across ethnic groups with regard to reasons of seeking professional help. Firstly the present study indicated that the effects of gambling on family members and/or encouragement by family, acted as a very strong catalyst for Pacific clients to seek help. Another ethnic based difference that was particularly apparent was the propensity for Asian clients to initially seek help following the direction of an authority (e.g. as conditions of a self exclusion order).

Five main themes emerged in relation to good service features across the four population groups. Common to all ethnic groups, the first theme centred on perceived practitioner/service knowledge and attributes/abilities, including: provision of good information/advice/resources/referrals, good follow ups, confidentiality, content of counselling and the client-practitioner relationship, unconditional positive regard, non-judgemental, friendliness, caring, and an ability to bridge the gap between clients and family. This theme has been constantly referred to in the existing body of literature on the topic of working with individuals affected by problem gambling or addiction in general (e.g. Ogborne, Wild, Braun & Newton-Taylor, 1998; Najavits & Weiss, 1994). A second theme universal to all ethnic groups was the importance of considering family in the therapeutic process. European, Maori and Asian gambling clients commented on the importance of directly involving family in counselling (e.g. through couple or family counselling) and Pacific commented on the benefits of practitioners understanding the importance of family. On the other hand, recovery for family members is not an easy road. Wives or significant others of problem gamblers often experience a ‘developmental lag’ in recovery when contrasted to the gambler (Ciarrocchi & Reinert, 1993) and can reflect a strong sense of anger and resentment even after many years since the last gambling episode. It was noted that the negative emotions were related to financial matters. Two explanations are proposed: 1) the extent of financial ruin has just began to emerge and leads to not only anxiety but a severe level of uncontrollability and unpredictability and, 2) the effects and duration of long term financial devastation throughout the gambling episodes. Within the context of New Zealand, Morrison (2005) conducted a study to specifically investigate the impacts of Maori women’s problem

106

XGAMBLE Report

gambling on their partners and whanau members. The third theme related to the availability and accessibility of services. European, Maori, Asian and Pacific participants all referred to various aspects of service accessibility, including being free (including transport), anonymous, available after hours, have ongoing support, and websites (consistent with Clarke et al., 2007). The fourth theme endorsed by all ethnic groups was the opportunity to address a wide range of issues impacting upon gambling behaviour. For example, some participants cited financial issues, learning basic life and communication skills, finding alternative activities as important features of a good service. Griffiths and MacDonald commented (1999, p. 189) “An initial focus on relationship difficulties (preferably involving the partner of the client), and financial and legal difficulties is appropriate”. Not only focusing on problems, educating and counselling people on how to make best use of their free time can have a great impact on the physical and mental health of individuals (Leitner & Leitner, 2005). In the case of individuals with problem gambling, leisure counselling to assist people to find alternative activities other than gambling and using leisure activities as a form of distraction to reduce the gambling preoccupation is likely to be useful. For a comprehensive review of similar matters, see: Costello & Millar, 2000; Cultural Partners Australian Consortium, 2000; Goodman, 1995. The final theme centred on the inclusion of cultural strategies/methods. Maori, Asian and Pacific clients all cited the use of cultural elements such as tikanga and use of native language (for details of an example of kaupapa Maori intervention programme see Herd & Richards, 2004; for Pacific interventions, see Robertson, Pitama, Huriwai, Ahuriri-Driscoll, Haitana, Larsen & Uta’I, 2005; for Asian interventions, see Wong & Tse, 2003) or the practitioner’s ability to understand their background as important factors of good service.

Three areas were identified in relation to factors which hinder success. They can be categorised as practitioner skills and availability, individual client attributes, and policy issues. Different ethnic groups tended to have their own emphases. European and Maori participants indicated that practitioners who are judgemental and don’t listen, understand, empathise or offer specific advice were not conducive to good outcomes. Maori and Asian participants also commented on the negatives associated with limited availability of practitioners (not very flexible in appointment times, not enough group/family work, and the service not available 24 hours a day). Conversely, Pacific participants mentioned the negatives of communication problems with clients and family members. “The call is for therapies that are high-quality and accessible in a more consistent way, and that are practised in a way that is consistent with a recovery approach” (Peters, 2007, p. 7). With regard to individual client attributes, issues such as a lack of self control and motivation, denial, shame, and a lack of self confidence were mentioned by all groups except European. Finally, Maori and Asian participants had a specific concern about policy issues. Maori felt that those who fund services lack an understanding of the issues and interactions facing Maori, and that this leads to service barriers. Asian participants had a strong feeling that the availability of gambling (and associated issues such as loan sharks) whilst attending counselling hindered their attempts to address gambling issues.

Most suggestions on how to improve services addressed accessibility and availability, for example; more funding to provide more practitioners, 24 hour services, more

107

XGAMBLE Report

programmes, and text messaging. Both Asian and Pacific participants requested more health promotion activities (e.g. formation of action groups) and the promotion of messages highlighting the harms caused by gambling. One example of similarly aligned work is that of the Chinese Family Life Services of Metro, Toronto (1995).

Universal to all population groups, participants tended to define successful interventions in two ways. The first centred on changes to gambling behaviour (stopping/controlling) and improvement in other aspects of their lives (relationships, finances, emotions, health and wellbeing). The other aspect of successful interventions seemed to reflect a sense of satisfaction with the social support provided by the practitioner. In particular, it was important to have someone who could listen, empathise, provide information/support, and understand the issues facing the client and their family.

2.4.2 Practitioners experiences of providing problem gambling services The semi-structured interviews with practitioners revealed a number of common themes. Regardless of organisation, ethnicity, and mode of treatment delivery (face-to-face or telephone), a kaupapa/purpose relating to the reduction of harm from gambling was consistently expressed. Practitioners achieved their kaupapa through a variety of activities that engage with different levels of community, including counselling, health promotion, education, and advocacy.

All practitioners stated the importance of incorporating methods and approaches that are culturally appropriate for their clients. As also outlined in Tan and Wurtzburg (2004), this was particularly the case for the Maori, Pacific and Asian practitioners.

The notion of accessibility was mentioned by European, Maori and Pacific practitioners. While the definition of accessibility varied somewhat (cost, the client’s mobility, location of sessions, hours of service), all practitioners aimed to make their services as readily available as possible.

As evident from the above, practitioners fulfilled a large number of roles and functions. In order to perform these roles, each practitioner employed a wide variety of skills and many approached problem gambling in a holistic manner; most practitioners recognised the need to be flexible, adaptive, accessible, and culturally appropriate (Jackson et al., 2003; Jackson, Thomas & Blaszczynski, 2003). It is also interesting to note that the face-to-face practitioners (European) reported using a more diverse range of techniques. This section of findings is extremely consistent with the results of the Counsellors Tasks Analysis (see Subsection 2.3.9 results from CTA [PG]). An expansive range of theoretical/therapeutic orientations were also reported. While all practitioners were eclectic in their choice and use of theoretical models, the use of cognitive-behavioural techniques was particularly common and the importance of the client-practitioner relationship was consistently recognised. All but one practitioner considered the client-practitioner relationship to be extremely or very important in their service. However, while relationship aspects (building trust, developing a therapeutic relationship and building rapport) were reportedly used by a greater proportion of telephone practitioners,

108

XGAMBLE Report

face-to-face practitioners also reported that they proposed appropriate interventions and built client’s self-confidence.

Practitioners gathered information by using a variety of measurement tools. Regardless of ethnicity, almost all practitioners reported using a problem gambling screen (most frequently the SOGS), and a number also enquired about the amount of money spent on gambling (Petry, 2006). While a number of other health issues (e.g. alcohol use, depression) were assessed, there was a level of inconsistency as to which issues were assessed and the choice of measurement tools (for a comprehensive review on problem gambling and comorbidity, see Crockford & el-Guebaly, 1998; Stewart &. Kushner, 2003).

Difficulties around finances and relationships were the problems most frequently indicated by European, Pacific and Maori practitioners as causing clients to attend their services. In contrast, Asian practitioners identified settlement issues, the impact of gambling on other aspects of life (family, work, study), and financial problems as the main reasons that Asian clients utilise their service. These findings are consistent with other work in this area (e.g. Newscan, November 2002; Tse, Wong & Kim, 2004; for impacts of problem gambling on Asian from refugee background, see Petry, Armentano, Kuoch, Norinth & Smith, 2003). It is not surprising then, that most practitioners rated relationships and financial difficulties as two of the problem areas that take up most their time. However, it should be noted that financial difficulties rated in the top one to three issues for Maori, Pacific and Asian practitioners but only rated sixth as an issue that takes up the most time for Europeans. In line with previous research (Dyall, 2002), it is important to note that Maori and Pacific practitioners also referred to legal and judicial issues taking up a considerable portion of their time.

Practitioners tended to define successful client outcomes as a reduction in gambling (including gaining control of gambling). Practitioners from all ethnicities identified client-centred and culturally appropriate services as important to achieving these outcomes, and a lack of resources and funding (resulting in excessive workload, and poor training opportunities) as hindering positive client outcomes. A number of other factors were identified as negatively impacting on client outcomes. Maori, Pacific and Asian practitioners all commented on the impacts of client ambivalence: client outcomes are not as positive when clients are not prepared or committed to addressing problems. Consistent with an early New Zealand study (Tse et al., 2005), the wider context of gambling behaviour was also mentioned as a hindrance to positive client outcomes. European, Pacific and Asian practitioners discussed the roles of gambling politics (e.g. government legislation, gambling policies), controls in gambling venues (e.g. a lack of commitment to host responsibility in some venues), and advertising for gambling products, in relation to gambling harm. They felt that more could be done in these areas to enable clients to successfully address their gambling issues. While the defining factors of successful outcomes were somewhat consistent between the telephone practitioners and face-to-face practitioners, responses from the telephone practitioners were more diverse (i.e. each one was mentioned by a small proportion of practitioners): relationship

109

XGAMBLE Report

with the service, support networks, use of CBT, understanding self and what triggers them to gamble.

Overall, the gambling practitioners appeared to enjoy working in the problem gambling intervention field. Almost all practitioners indicated that their work was either extremely, or quite satisfying. Seeing people change and/or making a difference were cited by all ethnic groups as satisfying aspects of work in this field. Maori, Pacific and Asian practitioners also found working in a flexible way and using a wide range of skills, especially culturally appropriate skills/methods/models as satisfying features of their work.

A number of themes emerged in relation to unsatisfying work aspects. Practitioners from all ethnic groups found it unsatisfying to work with clients who lacked commitment to change/address gambling related issues. A lack of resources and training opportunities, and high workloads were also referred to by Maori, Pacific and Asian practitioners. Similarly, European, Maori and Pacific practitioners expressed frustration at the emphasis placed upon outcomes and funding structures by management and the funding body. Some differences were observed in relation to face-to-face and telephone practitioners: internal politics and a lack of genuine client-practitioner therapeutic relationships were mentioned by face-to-face practitioners, and negative experiences when referring to other agencies and an inability to follow up with clients were highlighted by telephone practitioners.

2.4.3 Summary: Clients and practitioners experience of problem gambling services To sum up the responses to one of the key research questions in Phase One of the present study, ‘what contributes to good problem gambling intervention outcomes?’, the following diagram (see Figure 5) was constructed. The ‘client’ is the centre point of the intervention service and their readiness and motivation to make changes were identified by practitioners in this survey as the key determinants of good outcomes. The other three important elements which contributed to the outcome of intervention services are: ‘Intervention’, ‘Therapeutic Relationship’ and ‘Social Support’. This study demonstrated that practitioners have to deal with both gambling problems and a wide array of other issues concerning the families and individuals presenting to problem gambling services. Use of specific therapeutic approaches such as cognitive behavioural therapy or narrative techniques were considered key factors which contributed to good outcomes. In the present survey no respondent explicitly mentioned that the selection of intervention approaches relied upon evidence-based practice. Service delivery (e.g. how often the client will be seen) was identified as another key variable. Undeniably, from both the practitioners and service user’s point of view, the therapeutic relationship has the strongest bearing to intervention outcomes. Based on the qualitative data, the last element related to social support (e.g. from immediate family). Based on the obtained data in the qualitative study, the overall effectiveness of a problem gambling intervention appears to depend upon the extent to which the service users’ cultural (defined broadly) and linguistic needs are met.

110

XGAMBLE Report

Figure 5: Diagrammatic representation of key factors that contribute to the outcomes of problem gambling interventions

Counsellors: 1) Personal qualities e.g. empathetic, genuine

2) Professional skills & knowledge 3) Interactions with clients e.g. non-judgmental, accepting

Problem gambling interventions: 1) Interventions - gambling & related concerns

2) Use of specific approaches 3) Services characteristics e.g. mobile services,

Easy access, flexible hours

Social and family support: 1) Support changes 2) Prevent relapses

Client Readiness & motivation to make changes

Clients’ cultural and linguistic needs are met

111

XGAMBLE Report

2.4.4 CTA discussion on comparison between New Zealand and Australian practitioners

The results presented below compare the findings of the New Zealand survey with comparative data derived form an analysis of practice of problem gambling practitioners (primarily psychologists and social workers) in Victoria, Australia. The comparisons should be treated with some caution, however, as although the Australian survey was conducted in a state-wide gambling counselling service (Gamblers’ Help), it was conducted seven years prior to the New Zealand survey, so some differences in task performance may be due to changes in practice emanating from, for example, greater awareness of aetiology and more attention being paid to comorbidities, than to differences in service design, service philosophy and cultural factors relating to clients and workers in the New Zealand context.

The correlation between “frequency of performing certain tasks” and “importance” rating was almost 1:1 as the Australian team found before, so it seemed useful to concentrate on frequency of task performance only. Table 8 compares the overall scale properties (i.e. scale mean, standard deviation (SD), alpha, standard error of measurement (SEM), corrected item-total correlations, and factor loadings) of the frequency subscales for the New Zealand sample and compares the same to the Australian data. It shows that the New Zealand problem gambling practitioners are on average performing all tasks related to clinical practice in problem gambling services, across all subscales, more frequently than their Australian counter parts.

The psychometric properties of the CTA (PG) based on an Australian cohort have been reported elsewhere (Jackson, Holt, Thomas & Crisp, 2003). These properties have been re-examined with the New Zealand sample of practitioners, in order to both confirm the established properties of the scale and to compare the New Zealand practitioners with their Australian counterparts.

Homogeneity of the items within each subscale was estimated statistically by measuring the internal consistency of the domain using Cronbach’s alpha (Cronbach, 1951). It was necessary to determine whether the different subscales had similar psychometric characteristics as compared to the Australian data and to contribute to our understanding of the reliability of the scale. The SEM ranged between .42 on the Treatment Goals Scale and a maximum of 1.97 on the Family Interventions scale for the New Zealand data. These results are similar to the Australian data.

The reliabilities of the scale ranged from a low of .56 on the Interventions for Related Problems and .64 on the Treatment Goals Subscales to .92 on the Family Interventions Subscales. The Assessment subscales of both the New Zealand and Australian data were similar. Overall the subscales demonstrate strong construct validity and reliability.

The nine subscales were also tested for their underlying factor structure using a principal axis factor analysis.

112

XGAMBLE Report

Tables 3 to11 (in the XGAMBLE Appendices Booklet) provide an overview of the factor analysis of the items in each scale. The items loaded as a single factor using the Principal Components methods of analysis. Results of the factor analyses provide evidence for the construct validity of the nine subscales for the New Zealand sample of problem gambling practitioners. In addition, Table 9 illustrates that each subscale explained a considerable proportion of the available factor space or variation ranging from 33% in the Assessment subscale to almost 65% in the Family Interventions subscale with strong eigen values across all scales.

113

XGAMBLE Report

Table 8: Means, Reliabilities, Standard Error, Corrected Item-Total Correlations of ‘How Often’ Subscales

Subscales – Number Of Mean SD SEM Alpha Factor Loading on Corrected Item-Frequency Tasks Across individual

subscales Across individual

subscales Across individual

subscales Across individual

subscales

based on the total CTA Scale

Total Correlation based on the total

CTA Scale

NZ Aus NZ Aus NZ Aus NZ Aus NZ Aus NZ Aus

Assessment 13 52.79 47.76 9.3 7.46 1.90 1.23 .79 .79 .15 .52 .13 .36

Treatment goals 5 22.70 22.00 2.2 2.22 .42 .34 .64 .58 .25 .46 .29 .38

General Interventions 9 36.11 34.30 5.0 4.41 .98 .71 .74 .72 .79 .59 .77 .48

Gambling Interventions 16 61.00 59.05 10.1 11.66 2.02 1.82 .91 .92 .85 .66 .79 .49

Family Interventions 9 28.71 30.20 9.7 5.88 1.97 .99 .92 .78 .89 .72 .81 .55

Interventions Related Problems 6 18.46 11.51 4.1 4.15 .84 .65 .56 .74 .80 .62 .67 ..47

Referral 4 13.61 12.61 3.7 2.53 .73 .39 .74 .63 .83 .65 .73 .53

Education 7 18.26 15.63 6.8 5.11 1.42 .80 .87 .80 .83 .56 .65 .42

Research/Policy 8 25.00 24.02 8.1 6.78 1.68 1.06 .89 .87 .78 .50 .64 .27

114

XGAMBLE Report

Table 9: Variance and Eigen Values of Frequency Subscales for NZ and Australia

Subscales Frequency Frequency

NZ Aus

Variance

Across individual subscales

Eigen Values Variance Eigen Values

Assessment 33.73 4.39 39.18 3.53

Treatment goals 42.57 2.13 15.36 1.38

General Interventions 38.24 3.41 12.04 1.08

Gambling Interventions 42.39 6.78 9.17 .83

Family Interventions 64.64 5.82 7.45 .67

Interventions Related Problems 40.21 2.41 7.01 .63

Referral 56.91 2.28 4.78 .43

Education 57.23 4.00 3.23 .29

Research/Policy 57.42 4.59 1.79 .16

Table 10 shows the frequency data correlation within the New Zealand data and then compares the same to the Australian data across all subscales of practice. From the New Zealand data ‘frequency’ point of view, Pearson correlations indicated significant relationships on a number of levels, especially between the establishment of treatment goals and the use of general therapeutic interventions. Significant correlations appeared between gambling interventions, family interventions, and interventions for related problems scales such as legal and financial to treat gambling problems, indicating a holistic approach to problem gambling clinical practice. However the Assessment subscale did not have significant correlations with any of the other subscales scales. That is, assessments were not related very well with other activities undertaken by practitioners. Inspection of the subscale items indicates that the explanation for this finding may be that assessment in some agencies could be a centralised function, not undertaken specifically by the practitioner who will undertake the interventions with a client.

Overall it could be stated that the New Zealand problem gambling practitioners’ practice is related to utilising interventions at a micro level that are highly specialised to address the issue of problem gambling. At a macro level they appear also to be highly involved in community education, outreach and advocating for clients on policy related problem gambling issues. This integrated micro/macro practice contrasts to some extent with the Australian benchmark data, which indicated a more either/or approach to micro and macro focussed practice.

Additionally the results of bivariate correlations computed between each of the background variables and the Frequency subscale scores for the New Zealand cohort showed that the New Zealand Gambling Intervention Subscale (r= .54* where p < .05) and New Zealand Family Intervention Subscale (r= .48* where p < .05) correlated significantly with the sex of the worker variable. This was similar to the

XGAMBLE Report

Australian data previously analysed. Furthermore it was also found that the methods of intervention significantly correlated with the New Zealand General Intervention Subscale (r= .52* where p < .05), Family Interventions (r= .63* where p < .05), and Research and Policy (r= .62* where p < .05) subscales respectively. Where New Zealand Problem Gambling practitioners engage in face-to-face counselling as their primary intervention mode they are more likely to use general and family interventions and to play a strong practice role in research and policy as well.

Table 10: Correlation of Subscales Relating to ‘HOW OFTEN’ in New Zealand

SUBSCALES NZ1 NZ2 NZ3 NZ4 NZ5 NZ6 NZ7 NZ8 NZ9

NZ Assessment 1 1.0

.38

.31

.29

.29

.17

.26

-.06

.12

1.0

.42*

.36

.18

.03

.12

-.15

-.03

1.0

.83**

.79**

.39

.49*

.47*

.53**

1.0

.74**

.57**

.59**

.53**

.52**

1.0

.68**

.71**

.71**

.62**

1.0

.78**

.77**

.56**

1.0

.75**

.76**

1.0

.69** 1.0

NZ Treatment 2

NZ General Inv 3

NZ Gambling Inv4

NZ Family Inv 5

NZ Inter Related Problems 6

NZ Referral 7

NZ Education 8

NZ Res/Policy 9

** Correlation is significant at the 0.01 level (two-tailed) *Correlation is significant at the 0.05 level (two-tailed)

Table 11 provides data for each pair of variables: correlation, average difference in means, and confidence interval for mean difference, standard deviation and standard error of the mean difference has been calculated.

Table 11: Correlation Statistics – New Zealand and Australian FREQUENCY Data

Mean N Std.

Deviation Std. Error

Mean Correlation Significance

Pair NZ Assessment 51.29 17 9.86 2.40 1 -.18 .49

Aus Assessment 50.00 17 7.28 1.80

Pair NZ Treatment Goal 22.50 22 2.22 .47 2 .04 .87

Aus Treatment Goal 22.00 22 2.33 .50

Pair 3

NZ General Interventions 35.75 20 4.73 1.06

.12 .62 Aus General Interventions 34.90 20 4.33 .97

Pair 4

NZ Gambling Interventions 59.95 20 9.67 2.17

.37 .11 Aus Gambling Interventions 60.70 20 11.52 2.58

Pair 5

NZ Family Interventions 29.07 15 8.53 2.22 .50 .05

116

XGAMBLE Report

Pair 6

Pair 7

Pair 8

Pair 9

Aus Family Interventions

NZ Inter Related Problems

Aus Inter Related Problems

NZ Referral

Aus Referral

NZ Education

Aus Education

NZ Research and Policy

Aus Research and Policy

31.87

18.44

10.83

14.10

12.20

19.00

16.55

25.11

22.55

15

18

18

20

20

18

18

18

18

6.58

4.45

4.78

3.58

3.00

6.88

5.67

8.62

7.55

1.70

1.05

1.12

.80

.67

1.62

1.34

2.03

1.78

.65

.30

.21

.22

.00

.20

.39

.40

This confirms that for most components of practitioner’s practice, the New Zealand practitioners tend to perform the tasks more frequently than their Australian counterparts.

This comparison also indicates that the strongest difference between Australian and New Zealand practitioners’ practice relates to their interventions in regard to inter-related problems, with New Zealand practitioners much more likely to address problems such as legal and financial issues than the Australian practitioners. This cluster of tasks is also ranked more highly in the New Zealand sample.

The significance level for family interventions may be an artefact of sample size.

In conclusion, this survey of New Zealand problem gambling practitioners has demonstrated:

• The present survey included both generic and population group specific services in Aotearoa, New Zealand. They were: Problem Gambling Foundation of New Zealand (including their Asian services), Gambling Helpline New Zealand, Oasis Centres/NZ Salvation Army, Pacific Mental Health, Alcohol, Drugs and Gambling Services – Tupu Team, Pacificare Trust, and Te Rangihaeata Oranga. This sample represented only a quarter of Ministry funded problem gambling intervention providers in New Zealand and most of practitioners were recruited from Auckland. As such, the results may not be generalisable to the total practitioner population.

• With regard to the survey tool itself, the CTA (PG) is a robust tool for the measurement of clinical practice tasks in this sample.

• The New Zealand practitioner results indicate a holistic practice in which problem gambling is placed within a framework encompassing family contexts and which draws on both macro and micro practice in shaping interventions.

117

XGAMBLE Report

CHAPTER 3 PHASE TWO STUDY

The pilot clinical trial phase of this research aimed to investigate the effectiveness of the same intervention model delivered across two different intervention modalities: face-to-face counselling and interventions over the phone. The comparability of the intervention model across two delivery modalities was ensured by the use of an intervention manual (based on the actual practices in New Zealand identified from Phase One findings) in the trial and the documented process to evaluate the integrity of the interventions (for details see p.116).

3.1 Research aims As this was the first time that an investigation of the effectiveness of problem gambling interventions in New Zealand was conducted, it was proposed that Phase Two of this project be kept at a small scale (pilot/vanguard study), thus enabling it to inform the conduction of a full-scale trial.

This research therefore aimed to obtain only general overall trends (due to the limited sample sizes) and exploratory information across ethnic subgroups. This research generated hypotheses about differences across ethnic groups, but was not able to draw definite conclusions for the observed differences.

The specific aims of this research were: 1. To investigate the effectiveness of these interventions in controlling gambling

behaviours and reducing harm caused by problem gambling 2. To investigate the effectiveness of these interventions, with regard to the

magnitude of effect 3. To provide preliminary information on the relative effectiveness of the

interventions for ethnic groups (NZ European, Maori, Pacific and Asian) 4. To develop and recommend processes for future studies to extend research in the

field of problem gambling.

3.2 Methodology The overall methodology employed in the Phase Two of the study was a pragmatic randomised trial.

3.2.1 Inclusion/exclusion criteria • Inclusion criteria: People were potentially eligible for the trial if:

o They were aged 16 years or above; o They were currently experiencing problem gambling (as measured by

DSM-IV: a score of 3 or more); o They were residing in one of the locations - Auckland, Hawkes Bay,

Nelson, and Christchurch, where face-to-face interventions were available;

o There was an appropriate practitioner available (choice of culture/gender);

o They were able to attend face-to-face intervention sessions and receive telephone calls;

o Their primary problem mode was pokie machines; and, o They provided informed consent.

118

XGAMBLE Report

• Exclusion criteria: People were ineligible for the trial if: o They were at immediate risk of harm to self or others; o They were not interested in participating in the research process

(refused consent); o They were not able to commit to face-to-face or telephone counselling

services; o They were ‘sensitive cases’ - such as individuals who were involved in

a court case against the gambling industry or treatment agencies; and, o They refused random allocation.

With regard to the inclusion criteria, relevant material are referred to from the literature review section in Chapter One: 1.2.3 methodological issues related to interventions studies for problem gambling. Jackson, Thomas and Blaszczynski (2003) reviewed over 60 intervention studies and suggested that one of the common methodological flaws in many intervention studies was poorly delineated selection criteria and procedures for the inclusion of individuals into intervention programmes. Therefore a clear step-by-step inclusion and exclusion criteria were set out at the beginning of the design for the present study. Another common concern was a lack of distinction between intervention effects in relation to different forms of gambling. In terms of specific forms of gambling activity, the present study recruited only individuals who had concerns about their gambling on pokie machines. This was decided on the basis that the majority of clients presenting to Ministry funded problem gambling intervention services reported that the activities concerning them the most was pokie machines. Participants were recruited from four specific locations (see above) where participants could receive either face-to-face or telephone interventions by the participating practitioners. These three locations were chosen for the following reasons: 1) they broadly represent the spread of participants from the North and South Island; 2) specialised Maori problem gambling services were located in one of the four areas; and, 3) they have a large ‘catchment’ area and population size to assist with participant recruitment. Lastly, it is worth pointing out that prospective participants were recruited specifically for the present study; in other words, they were not part of the existing clientele of Ministry funded intervention services.

3.2.2 Production of Manual

An interventions manual was produced, as informed by Phase One of this research. The primary aim of this manual was to ensure consistency of the approach within the present randomised study rather than to address competency issues in providing problem gambling interventions. It was expected that participating practitioners would follow their agencies policies, including procedures dealing with emergency situations, cultural safety, supervision and accountability. This manual was only concerned with ensuring a consistent, basic therapeutic outline/framework for problem gambling interventions. The manual was written to provide basic information to individual practitioners, to which they could then overlay their intervention approach(es), as deemed appropriate to working with clients allocated to the practitioners.

The writing of the manual was contributed to jointly, by a group of senior practitioners working in the field of problem gambling interventions. They were invited to join the group on the basis of: 1) practitioners’ experiences in the sector, 2) experiences in writing intervention manuals or protocols within their own

119

XGAMBLE Report

organisation or professional groups; and, 3) expertise and insights in working with specific population groups. A two-day workshop was held at the School of Population Health at Tamaki Campus, University of Auckland on 17th and 18th July 2006. The workshop process included: • Introduction to the background of the study - there was also opportunity for all

of the contributors to provide comments about the project itself; • Presentation of Phase One findings; • Deciding on the aims and scope of the manual; • Determining key headings or topics to be included in the manual; • Gathering ideas and material for each topic of interest; • Breaking into small writing groups on various topics; • Presenting back to the group – peer-review and comment; and, • Reviewing progress and making final remarks.

After the draft manual had been written up, it was circulated to all the contributors for further comments and additional information before completing the final document.

The Manual was utilised as the basic information for the one-day training (held on the 10th October 2006) provided to all participating practitioners involved in the Phase Two of the present study. Most of the practitioners participating in this study have experience with face-to-face and telephone counselling. Please note that the telephone counselling employed in this study does not represent any service that currently exists in New Zealand. In this study, telephone counselling involves therapeutic work between a practitioner and his/her client and the same practitioner over a period of time and a series of scheduled sessions. All the participating practitioners were asked to continue with their present supervisors and to inform the research project coordinator of any emerging issues that might arise.

The Manual includes helpful hints for telephone counselling and a list of common interventions used by most gambling practitioners (see below). Not every practitioner would use the same interventions for each client. • Referring on • Lifestyle balance • Tikanga protocols • Gambling education • Budget advice • Insight/Awareness-raising • Cognitive distortions/thinking • Normalizing/validating

errors • Instilling hope • Desensitization • Feeling identification • Bans and self-exclusions • Positive feedback • High-risk situations • Goal-setting • Early warning signs • Review of progress • Developing safety plan • Review of resources • Use of significant others • Coaching/Role-plays • Relationship counselling • Didactic interventions • Stress release

Upon close examination, the Project Team considered that the key points covered in the Intervention Manual strongly echoed the central themes of the Phase One findings. Using Figure 5 as an example, some of the key factors contributing to outcomes of problem gambling interventions relate to: interactions with clients, dealing with gambling related concerns, and strengthening social support to sustain the changes made. Within the manual, there was material describing the engagement phase,

120

XGAMBLE Report

outlining how to work with clients effectively over the telephone and how to conduct a user-friendly assessment. The Manual also had good discussions on supporting clients to deal with financial problems, stress, life-style issues, negative emotions and a sense of hopelessness. With regard to rebuilding social support, some sections within the Manual were devoted to topics such as relationship counselling and goal setting. For further details see the Manual attached separately

3.2.3 Schedule of intervention and measures 1. Informed consent was requested following an introduction to the research and

ascertaining that the potential participant had met ‘participant eligibility’. 2. Demographic and baseline data were collected. 3. Random allocation of the participant to one of the two groups (face-to-face or

telephone modality). All the participating practitioners, in both modalities, are qualified practitioners recruited from the Ministry funded problem gambling services. They have undergone training in the use of Intervention Manual for the present trial to make sure the practitioners are competent to provide the interventions in either face-to-face or telephone modality.

4. Allocation of the participant to the appropriate service and notification of the practitioner.

5. The practitioner contacted the participant and commenced the intervention. The intervention was considered completed for the purpose of the present trial after six sessions (as suggested by the participating practitioners). For those clients who required more than six sessions, they continued to receive intervention services, and the post-evaluation form was filled in by the responsible practitioner.

6. Following termination of the intervention, the Project Team were notified by the practitioner.

7. “Termination” was defined as one of the following situations in the present study: a) completion of planned sessions and achievement of the set goal; b) the clients no longer wished to attend counselling sessions; or c) the clients were not contactable despite three extra phone attempts made at different times of the day over a period of five days. For the actual distribution of participant termination status, see Table 40: Number of intervention sessions attended.

8. Assessments were undertaken at three points: a) at the commencement of intervention (i.e. baseline measure); b) the termination of the intervention (i.e. post-intervention measure); and c) at the subsequent follow up (i.e. six weeks after termination).

3.2.4 Study Design The overall design of Phase Two is known as a ‘pragmatic trial’ which aims to capture and compare the delivery of intervention services in the field. In comparison, an ‘explanatory trial’ tends to have very tight, laboratory-type control of various treatment conditions and its treatment effects are often diluted in the real world. The overall design of the study is best summarised in the following diagram (Figure 6).

121

XGAMBLE Report

Figure 6: Flowchart depicting Phase Two study design

Sample size Figure 7 explains the number of participants at registration and pre-randomisation; and at baseline, post-intervention and follow up.

122

XGAMBLE Report

Figure 7: Flowchart recording number of participants recruited in Phase Two study

Study Registration

(103)

Eligible to Randomise

(96)

“Face to Face” “Phone”

(47) (49)

Baseline measures Baseline measures (46) (46)

Did not attend any session- 12 (26%) Did not attend any session- 17 (37%) Attended between 1 & 5 sessions- 16 (35%) Attended between 1 & 5 sessions- 13 (28%) Attended 6 or more sessions- 18 (39%) Attended 6 or more sessions- 16 (35%)

Post-intervention measures Post-intervention measures

(14) (13)

Follow-up measures

(2)

Follow-up measures

(10)

For details on number of sessions attended by participants in the ‘face-to-face’ and ‘phone’ condition respectively, see Table 41.

123

XGAMBLE Report

Randomisation • Randomisation was achieved through a Central Telephone Service (research call

centre) via a web-based computer system. • The telephone service was staffed by research staff/associates, not practitioners. • Consenting participants were randomly assigned to one of the two intervention

arms, stratified by practitioners. • The method used to generate the random allocation sequence was computer-based

and called ‘stratified minimisation’. • Randomisation was stratified by practitioners to ensure that roughly equal

numbers of participants allocated to each practitioner were randomised to each of the two intervention arms. The distribution between groups was similar.

Blinding • Double-blinding was not possible in this design. • Assessor blinding relied on participants not disclosing which group they had been

allocated/referred to. • Data was collected by researchers, not the practitioners that provided the

intervention.

Interventions Practitioners from the four major population groups delivered the intervention (as based on the content in the intervention manual developed in Phase One). Each practitioner was a currently practicing clinician within a Ministry funded problem gambling service, with the relevant training and years of experiences in working with individuals affected by problem gambling and their significant others. In other words, they were competent practitioners that provided support and treatment for the clients that presented to this trial.

The study was described as “XGAMBLE” to all participants, who were aware that different groups would have different intervention modalities made available to them.

Participants commenced interventions from one of the two treatment arms: face-to-face and telephone. The primary intervention model or intervention procedures adopted in this trial were informed by the Intervention Manual as outlined above in Section 3.2.2 Production of Manual. All of the services provided were free, strictly confidential and delivered in a professional manner.

Face-to-face and Telephone services Intervention services were provided according to the Interventions Manual which outlined the assessments, major intervention approaches and activities/tasks to be performed by practitioners. These were identical for both face-to-face and telephone interventions.

There was no restriction on the frequency or duration of counselling sessions - that is, this trial evaluated the delivery of intervention services in a naturalistic setting. Similarly, if they were deemed to be appropriate, the services could offer any culturally-specific interventions.

The intensive part of the intervention lasted up to six sessions. Practitioners were asked to complete a form documenting the major elements and/or activities of the

124

XGAMBLE Report

services provided for each client involved in this study (see practitioner/call centre/ Project Team interface in the Section 3.2.6: Project Infrastructure below).

Outcome measures at the end of the intervention and six-week follow ups were collected by the call centre researchers. All participants were contacted for post-intervention and follow up measures regardless of the number of sessions or intervention completed.

Intervention integrity The evaluation of intervention integrity was performed by a multi-disciplinary team comprising of experts on the two problem gambling treatment modalities and experienced researchers, including those from relevant ethnic backgrounds. The monitoring and evaluation process took place in the following steps: • Description: Phase One of the overall study, compiled a description of the

problem gambling intervention services available in New Zealand. • Definition: The major treatment models that are utilised in intervention services

were defined (i.e. written in the Intervention Manual). • Formative evaluations: These were also conducted on the experiences of

participants (in terms of both clients and practitioners). • Evaluation: Evaluative stages also included documentation and feedback from the

participating practitioners upon the termination of intervention. Examples of information collected from the practitioners included: o Any reasons for termination (e.g. as planned, failure to attend, or “I am OK

now”). o A brief summary of what was covered over the course of the intervention. o Outcomes achieved. o Evaluation of efficacy of interventions:

� In respect to the goals set, which have been achieved? � What was most helpful about the intervention? � What did not work?

For further details on information collected from participating practitioners, see Section 3.2.7 Data collection/ measures.

3.2.5 Recruitment strategies The recruitment process began in June 2006, and was an ongoing process that tested which strategies worked the best and were the most cost effective. It ceased on 31 May 2007, in accordance with the completion of data collection for this final report.

Three main recruitment approaches were employed. They included: articles and advertising in major community newspapers (feature stories with pictures were the most successful); advertising in other media (general and ethnic specific mass media, including features or special items on printed materials, including newspapers and posters/flyers, and radio/TV/websites); and, ethnic researchers working with individuals/small groups of people and promoting the research generally within their own communities.

In order to promote the research and try to obtain further referrals, the Project Team also worked with the University of Auckland, service providers and community/social agencies (e.g. budgeting centres, CABs, Age Concern, probation centres, Marae and other cultural, church and sports groups).

125

XGAMBLE Report

The client participants were not part of the existing clientele of Ministry funded intervention services. In other words, they would not have received any problem gambling had this trial not been implemented and this has to be taken into account as one of the confounding factors for the present study. For instance, the participating clients might not necessarily represent the client population who would normally seek help from Ministry funded services.

See the XGAMBLE Appendices Booklet for a summary of the full recruitment strategy.

3.2.6 Project infrastructure In addition to the practitioners and the Project Team, two other units from within the University of Auckland were involved in this phase of the research. They are described in the following section.

Clinical Trials Research Unit The Clinical Trials Research Unit (CTRU), University of Auckland, is a multi-disciplinary group of over 50 health researchers that has forged an international reputation for studies into the causes, prevention and treatment of leading health problems. The Unit has recruited over 25,000 participants in clinical trials throughout Australasia, Asia and Europe and has developed a highly successful network of partners across these regions to co-ordinate studies and motivate people to participate and complete international trials. The unit is staffed by a broad range of experts including epidemiologists, biostatisticians, project managers, data mangers, IT specialists and administrative support.

The CTRU developed the web-based computer system which was utilised in this research to randomise participants into a treatment arm. The CTRU also assisted researchers (call centre interviewers) in the data collection phases and provided overall technical advice and support to the project. Their biostatisticians also conducted the statistical analysis.

The website, which included information on the research, the randomisation process and the data collection instruments, was developed intensively over a five month period between the Project Team and CTRU staff (including biostatisticians), and incorporated input from the Technical Advisory Group (including service providers/ practitioners/ethnic groups) and the call centre staff.

The web-based system went live on Monday 28th August 2006.

Call Centre The Survey Research Unit (SRU), University of Auckland, was sub-contracted to provide the call centre for this research. The call centre was operational from 28 August 2006 to 31 May 2007. They worked with the Project Team to set up the 0800 number (0800 XGAMBLE) and ran the overall call centre.

The call centre had initial hours of operation set for Thursdays, Fridays and Saturdays across the hours of 2 – 10pm, with a message service in operation for calls outside of these hours. A facility was set up to enable operators to divert / transfer crisis calls to

126

XGAMBLE Report

the Gambling Helpline. Experienced people were on hand to debrief call centre operators if necessary.

The SRU provided a team of experienced interviewers/researchers, who explained the research to participants, gained informed consent, registered the participants, and collected demographic and baseline data. They also conducted follow up calls in order to collect post-intervention and six week follow up data for the research.

A training session (involving CTRU staff, the Project Team, and gambling/telephone practitioners from TUPU and Lifeline) was held on August 7th 2006. The aim of this session was to introduce call centre staff to the research, train them on the use of the website (including registration, randomisation, data collection), discuss and provide information for their use (Participant Information Sheet, Consent Forms, Practitioner Contact Details, Verbal Scripts) and conduct role plays of calls in order for them to practise and learn how to work with this vulnerable population group.

The Project Team and the call centre ensured that cultural/ethnic diversity was catered for. This was accomplished through the provision of ethnic interviewers (European, Maori-Cook Island, Chinese, Korean, Samoan and Tongan interviewers). However, with the exception of Chinese and Korean participants, all interviews were conducted in English to ensure consistency.

The Project Team produced spreadsheets for the call centre staff to record practitioner/client records relating to progress checks and for any information that was required to be sent out and/or monitored. A spreadsheet of available practitioners, locations and cultural specific services was also produced. A record of client details was emailed to practitioners upon client registration, including information on whether or not consent had been granted for messages to be left on contact phones.

Finally, further training/enhancements were provided after pilot testing the recruitment strategies, call centre operation and data collection instruments. The ethnic recruiters in the field were also brought in to work with the interviewers: this was done to enhance the interviewers’ knowledge of the population group they were working with and to enable them to engage participants fully and gain their trust.

The Call Centre was ready to receive calls on Monday 28th August 2006.

Practitioners/Call Centre/Project Team Interface A protocol was developed which managed the interface between the call centre, the project manager (who monitored the progress of the study) and the practitioners involved. This is summarised below.

The Call Centre staff notified the project manager each time a client registered for the study. The project manager then sent the following details (via email) to the allocated practitioner:

• Registration number; • Client’s first name; • Contact phone number; • Time appropriate for the practitioner to call/able to leave message; and, • Mode of services that the client had been allocated to (i.e. face-to-face OR

phone-based intervention).

127

XGAMBLE Report

Unless advised that it was inconvenient to the client, the practitioner then made initial contact within 24 hours. This contact aimed to:

1) Establish rapport; and, 2) Arrange an appointment (if allocated to face-to-face based

intervention), OR, initiate a session (if allocated to phone based intervention) if convenient to client.

For those clients who could not be contacted as agreed or planned, the following actions were taken:

• At least three phone attempts were made at different times of the day spread over five days; and,

• Details of each attempt were recorded (i.e. the date and time calls are made).

3.2.7 Data collection/ measures The following is a summary of the data collection measures/instruments that were utilised in Phase Two of this research:

Data collection summary • Demographic information: age, sex, ethnicity, income level, job type, preferred

gambling mode etc. • Health and wellbeing. • Money (proportional to income) spent on gambling. • Time spent on gambling. • Gambling Attitudes and Beliefs Survey (GABS) (Breen & Zuckerman, 1999). • Gambling Readiness to Change Scale (GRTC) (Neighbors, Lostutter, Larimer &

Takushi, 2002). • Number of Diagnostic Statistical Manual-IV (American Psychiatric Association,

1994) criteria that were met. • Overall ratings on acceptability, satisfaction and accessibility of the intervention

as well as subjective ratings of the interventions’ success (Riley & Jean-Mary, 2004).

Baseline measures • The following demographic information was collected at baseline: age, gender,

ethnicity (including preferred), residence region, occupation, education level, marital status, dependents, recruitment method, alcohol & drug issues, mental health issues, previous gambling counselling, and income level.

• Two gambling questions (preferred gambling activity; self-excluded or banned from entering gambling venue in last two months) were asked at baseline only.

• Participants completed the DSM-IV (American Psychiatric Association, 1994) gambling questionnaire in order to measure their severity of gambling at baseline. This information was utilised to assess their eligibility in problem gambling and as stratification for randomisation.

Justifications: This instrument has been widely utilised and shown to be suitable in a New Zealand context, including in research by the lead investigator (Why people gamble). The DSM-IV is also considered a more stringent measure of problem/pathological gambling for both screening/diagnostics and evaluating intervention outcomes (e.g. Cox, Enns & Michaud, 2004; Stinchfield, Govoni & Frisch, 2001). In order to improve the understanding of what DSM-IV scores mean, Toce-Gerstein, Gerstein and Volberg (2003) analysed data drawn from two community-based samples

128

XGAMBLE Report

(total sample size, n= 2,417 and 530 respectively) from the United States. It was found that adult individuals who scored:

o 1-2 on DSM-IV criteria reported ‘chasing their losses’; o 3-4 criteria (so called sub-clinical level or problem gambling) reported

‘elevated rates of gambling-related fantasy, preoccupation, lying and gambling to escape;

o 5-7 criteria (pathological gambling) reported marked elevation of loss of control, withdrawal symptoms and tolerance (need to spend more money gambling in order to achieve desired level of excitement), risking social relationships and need to be bailed out financially; and,

o 8-10 criteria (highest level of pathological gambling) reported committing illegal acts to support gambling.

Another routinely used instrument in treatment outcome studies is the South Oaks gambling Screen (SOGS, Lesieur & Blume, 1987) - which has also previously been used to collect New Zealand problem gambling data. However, while an advantage of the SOGS would the ability to compare data collected in this trial with the existing national dataset and to estimate the generalisation to the wider problem gambling client population, the SOGS was neither developed nor validated for evaluating treatment outcomes. As such it was not considered appropriate for use in the present research.

• Information on participants gambling behaviour (money/time spent gambling, sense of control, income in last two weeks) was collected.

Justifications: These two types of data - money and time spent gambling have been collected as part of New Zealand’s national problem gambling data. They are also regarded as the most sensitive, direct and meaningful measures of any change in treatment outcomes studies (e.g. Toneatto & Ladouceur, 2003).

• The Gambling Attitudes and Beliefs Survey (GABS; (Breen & Zuckerman, 1999) was used as a baseline measure.

Justifications: Two major methodological flaws in problem gambling intervention studies are: 1) lack of clear articulation of intervention model(s) employed; 2) few attempts have been made to measure the process changes which are congruent to the intervention models and experiences of participants. Toneatto and Ladouceur (2003) have strongly recommended the use of the GABS as a way to measure participants’ changes in cognition while undergoing a cognitive-behavioural orientated intervention service. The GABS assesses general attitudes toward gambling and has been constructed to capture a wide range of cognitive biases, irrational beliefs, and positively valued attitudes toward gambling. It focuses on cognitive factors related to gambling and includes a 35-item questionnaire to which participants’ report the extent to which they agree, from “1” (strongly disagree) to “4” (strongly agree). Sample items include “Some people can bring bad luck to other people” and “if I lost my bets recently, my luck is bound to change”. The GABS is scored as the mean of all items. Higher scores indicate pro-gambling attitudes and beliefs. The GABS has demonstrated good internal consistency in a pilot sample of university students who were active gamblers (Cronbach’s alpha= .93). GABS scores reliably discriminated a non-problem from problem gambling sample. Factor analyses of both students and a community sample revealed that one overriding factor accounted for over 30% of the variance in scores. The GABS has also demonstrated predictive validity - higher scores on GABS predicted a willingness to gamble risking real money in a sample of university students.

129

XGAMBLE Report

• The Gambling Readiness to Change Questionnaire (GRTC; Neighbors, Lostutter, Larimer & Takushi, 2002) was also collected for each participant at baseline.

Justifications: The measure on participants’ level of motivation or readiness to change is deemed to be one of the critical factors related to intervention outcome. At a clinical or practice level, information gathered can be useful to aid both practitioners and clients to plan the intervention. The GRTC was modelled after the alcohol Readiness to Change questionnaire (Rollnick, Heather, Gold & Hall, 1992) which was based on the stages-of-change model (Prochaska & DiClemente, 1986). The GRTC has nine items with three items measuring each of three stages: Pre-contemplation, Contemplation and Action. Research participants indicate the extent to which they agree with the statement presented in each item, from “1” (strongly disagree) to “5” (strongly agree). Samples of items include “I enjoy my gambling but sometimes I gamble too much”, “There is no need for me to think about changing my gambling”. For the present study, results were interpreted by categorising individuals as “pre-contemplator”, “contemplator” or “in the action stage” according to the highest subscale score. Neighbors and colleagues (2002) found that the GRTC scale was only appropriate for measuring readiness to change among individuals who gamble at least moderately and thus might have reason to change their gambling behaviours. Principal component analysis revealed strong support for a three-factor configuration: factor 1 (contemplation items, explained variance 41.9%); factor 2 (action items, explained variance 14.3%); and, factor 3 (pre-contemplation items, explained variance 11.2%). The GRTC also demonstrated satisfactory reliability for each of the three subscales, with alphas of .64, .90 and .74 for pre-contemplation, contemplation and action respectively.

• Participants were also asked to provide a rating of their expectation of the effectiveness of the intervention they were about to receive (Riley & Jean-Mary, 2004).

Justifications: This set of simple outcome ratings by research participants who were using the services provided in this trial, was thought to provide very useful information to the Project Team and the Ministry in terms of future policy and services development. Research participants indicated the extent to which they agree with the statement presented in each item, from “1” (strongly disagree) to “5” (strongly agree). Sample statements include “I am satisfied with the XXX (this would be either face-to-face or telephone services) services provided to me in this study”, “The XXX services are easily accessible” or “I have been successful in controlling my gambling behaviours”.

(See XGAMBLE Appendices Booklet for Form A & B data collection instruments)

Outcome measures Post-intervention Measurement: • At the end of their intervention, all participants were again asked to complete the

Gambling Questionnaire (to assess money/time spent gambling, sense of control, income in last two weeks), the Gambling Attitudes and Beliefs Survey (GABS) and the Gambling Readiness to Change Questionnaire (GRTC).

• Each participant was also asked to provide an overall rating and explanation of effectiveness, satisfaction, accessibility, preferred option for help seeking, and if they received any other support during the intervention.

130

XGAMBLE Report

Post- follow up Measurement (another 6-weeks): • All participants were asked to complete the Gambling Questionnaire (to assess

money/time spent gambling, sense of control, income in last two weeks), the Gambling Attitudes and Beliefs Survey (GABS) and the Gambling Readiness to Change Questionnaire (GRTC) at the end of a 6-week follow up.

Primary outcomes: The primary endpoints of this study were the comparison of the two interventions in relation to: • Total money spent on gambling over the last four weeks; and, • Total time spent on gambling over the last four weeks. Both were measured repeatedly at baseline, post-intervention and end of 6-week follow up.

Secondary outcomes: The secondary endpoints of this study were the comparison of the two interventions in relation to: • Proportion of total money spent on gambling over the last four weeks to the total

income over the last four weeks; • Overall score on the Gambling Attitudes and Beliefs Survey (GABS); and, • Overall score on the Gambling Readiness to Change Questionnaire (GRTC). All of these measures were repeated at baseline, post-intervention and the end of 6-week follow up.

Other measures of interest: In addition, the following variables were measured to gain information on responses to interventions, variability in responses, and to assist planning for future studies in this client group: • Sense of control over gambling behaviour over the last four weeks; and • Post-intervention ratings and qualitative data on effectiveness, satisfaction,

accessibility, preferred option for help seeking, and if they received any other support during the intervention.

Collecting practitioners information Following the first session, the practitioner notified the project manager regarding the start, or otherwise, of counselling. Table 12 summarises the minimum documentation that practitioners were required to send to the project manager, subsequent to the third session. Note that the practitioner’s agency policy may require additional information.

Table 12: Minimum documentation to be completed by practitioners subsequent to the third counselling session

Intervention Modality

Face-to-Face Telephone • Date and time of session • Who attended the session • Presentation of client e.g. including mood &

risk (must be more explicit in area of risk) • Location of the session • Goals from the previous week# • Presenting issues and interventions put in

place#

• Date and time of session • Who participated in the session • Client’s location if appropriate (e.g. where

the call is received or made?) • Goals from the previous week • Presenting issues and interventions put in

place# • Presentation of client e.g. including mood &

risk (must be more explicit in area of risk).

131

XGAMBLE Report

• Goals of next session# • Signed and dated

Presentation will be partially reliant on oral/audio cues e.g. tone of voice, uneven breathing, pressured speech#

• Goals of next session# • Signed and dated

# denotes the areas where the practitioners’ compliance with the Intervention Manual were monitored and managed. For example, 39 out of 92 (42.4%) practitioners’ notes were reviewed and confirmed that the main interventions employed by practitioners in this project were around changing behaviours, increasing the level of knowledge about gambling and problem gambling and shifting from maladaptive to adaptive cognitions.

When the intervention was terminated, (or at the sixth session if the client was continuing with counselling) the following areas were also documented: • All information to be identified with the registration number (NO client

participant full name). • Date of termination (to be notified to project manager within 24 hours). • Any reasons for termination (e.g. as planned, or failure to attend, or “I am OK

now”). • A brief summary of what was covered over the course of the intervention. • Potential risk of relapse and general safety. • Recommendations for future contacts, including continuation of counselling. • Outcomes achieved. • Evaluation of efficacy of interventions:

- In respect to the goals set, which have been achieved? - What was most helpful about the intervention? - What did not work?

These data were to be sent to the project manager within 24 hours of completion/termination to allow data collection in a timely manner.

3.2.8 Follow up and retention In order to maximise follow up and reduce drop-out rates, three specific strategies were utilised: 1. The practitioners explained the importance of follow up measures to research

clients during the intense period of intervention (i.e. approximately the first 30 days);

2. A ten-dollar DVD/video hire voucher was given to individuals who took part in the follow up assessment; and,

3. The follow up assessment was conducted in a way convenient to the research clients (e.g. over the phone or via text-messaging, or web-based internet).

3.2.9 Data analysis and statistical methods Overall plan Procedures of the statistical analysis system SAS v9.1.3 (SAS Institute Inc. Cary NC) as well as SPLUS were utilised for all statistical analyses. All statistical tests were two-tailed and a 5% significance level was maintained throughout the analyses.

The main treatment evaluations were performed on the principle of ‘intention to treat’. However, data for those study participants who were randomised but did not have any of the Form B data collected (i.e. their outcome measures are totally missing

132

XGAMBLE Report

for all visits) could not be analyzed. The statistical analysis included all remaining participants who were randomised with at least baseline Form B completed. The randomisation was stratified by the practitioners, which was considered as one of the most important prognostic factors to the study outcomes (total money and time spent gambling). With 20 practitioners (i.e. strata) in total, the stratified minimisation was considered the most appropriate method. The weighting within each stratum was 50%/50% if the same number of participants have been allocated to each of the two arms, or 10%/90% if more allocated to one arm (say the former). This ensures that roughly equal number of participants allocated to each practitioner were randomised to each of the two intervention arms.

Data were analysed following a pre-specified analysis plan.

Baseline characteristics Demographic information (age, gender, ethnicity, region of residence, occupation, education, income level, A&D issues, mental health concerns, etc.), along with relevant assessments of their gambling behaviours (DSM-IV score, gambling activity, self-excluded or banned from entering gambling venue) were summarised for participants in each of the intervention arms. As differences between randomised groups at baseline can only have occurred by chance, no formal significance testing was conducted.

Summaries of continuous baseline variables with normal distribution were presented as means and standard deviations, or medians and inter-quartiles for skewed data, while categorical variables were described as frequencies and percentages.

Effectiveness analyses Primary analysis Summary statistics were provided for each group at each visit on the primary endpoints measured from all eligible participants. Their distributions were checked and the amount of missing values at each visit was noted.

Analysis of the primary endpoints was carried out using appropriate repeated measurement analyses, assuming that non-observed data were missing at random. If the data were essentially normally distributed, the Mixed model was used to compare the treatment effects between the two intervention groups at baseline, post-intervention and end of 6-week follow up, while adjusting for their ethnicities and other stratification factors (for randomisation) in the regression model. Other potential covariates such as whether or not they had received prior counselling and their health histories were adjusted in the regression model if they were statistically significant.

Preliminary estimates of the intervention effect, together with standard errors, were calculated in order to establish what clinically relevant differences would be desirable to detect in a full-scale (nation wide) trial.

Secondary analysis Similar repeated measurement analyses were carried out for the secondary endpoints using the collected information. Descriptive information was provided for additional variables of interest (sense of control; post-intervention measurements). Results were compared between treatment groups using standard statistical techniques.

Sensitivity analysis

133

XGAMBLE Report

Sensitivity analysis was carried out on the two primary endpoints. In particular, the following simple imputation strategy was utilised to fill in any missing values that may occur in later visits: • At visit 2 (post-intervention), impute any missing value with its baseline value

assuming no change from baseline; and, • At visit 3 (follow up), impute any missing value with the average of those

measured at baseline and post-intervention.

There should be no missing observations at the baseline visit.

Primary outcome measures after imputation were first summarised for each group at each visit. Same repeated measures analyses were next carried out using the imputed dataset. Results were compared with the primary analysis for the estimation of main treatment effect. With a proposed sample size of 200 (~100 recruited), the repeated measures analysis was considered using all available information collected at multiple phases. The regression model has adjusted for a list of important confounding factors, including: age, sex, ethnicity, city to live, previous counselling and medical history. With a very high drop out rate, however, there were certainly limitations on the scope of analysis that could be possibly carried out.

Data cleaning, analysis, interpretation and reporting were completed within a two month timeframe.

3.3 Results

3.3.1 Evaluation of intervention effects

Descriptive summary of demographics This section provides a descriptive summary on important demographics of the 96 randomised study participants. The summary covers a description of demographic features for the whole sample followed by a breakdown of both the face-to-face and telephone based intervention groups. Further description is provided for the four population groups based on participants’ preferred ethnicity, whenever it is appropriate (e.g. when there are considerable differences in the trends of participants’ responses).

Age The mean age of all study participants was 43 years old (SD1 12.26), with 44 participants (SD 12.16) in the face-to-face intervention group and 42 participants (SD 12.41) in the telephone intervention group. Further details are provided in Table 13.

Diagnostic Statistical Manual- IV (DSM-IV) scores DSM-IV score was calculated as the total number of affirmative responses to 10 questions. A higher score indicates greater severity of problems with gambling. In this study, participants must have a score of no less than 3 to be eligible.

The mean DSM-IV score of all study participants was 7.4 (SD 1.89), with 7 (SD 1.79) in the face-to-face intervention group, and 7.7 (SD 1.94) in the telephone intervention group.

1 SD – Standard Deviation.

134

XGAMBLE Report

Table 13: Distribution of participants’ age and DSM-IV scores (n=96)

Measure Face-to-face (n= 47)

Telephone (n= 49)

Total (n=96)

mean (range)

SD median mean (range)

SD median mean (range)

SD median

Age in years 44.51 12.16 47 42.51 12.41 44 43.49 12.26 45 (22-68) (19-67) (19-68)

DSM-IV score ( 0 – 10)

7.04 (3-10)

1.79 7 7.73 (3-10)

1.94 8 7.4 (3-10)

1.89 8

Table 14 shows that participants of non-European ethnicity were more likely to obtain a score of eight or higher (i.e. the median score of the present sample) on the DSM-IV.

Table 14: Distribution of participants scoring eight or more on the DSM-IV according to ethnicity

DSM score of 8 or more

Ethnicity NZ European

(n= 53) Maori

(n= 20) Pacific (n= 8)

Asian (Korean, Chinese &

South Asian) (n= 7)

Other (n= 8)

N 27 13 4 7 2 % 51 65 50 100 25

Gender There were 64 female participants (67%) and 32 male participants (33%) in the study, with slightly higher proportion of females in the face-to-face group (34 out of 47; 72%) compared to the telephone group (30 out of 49; 61%).

Participant’s Location Sixty-seven percent of participants (n=64) lived in the Auckland area. The remaining participants lived in Christchurch (n=26, 27%) or Hawkes Bay (n=6, 6%). The distribution is very similar between intervention groups – see Table 15.

Table 15: Distribution of participants’ location according to intervention modality (n=96)

Intervention Modality Total (n=96)Face-to-face

(n= 47) Telephone

(n= 49) n % n % n %

Auckland 31 66 33 67.3 64 66.7 Christchurch 13 27.7 13 26.5 26 27.1 Hawkes Bay 3 6.4 3 6.1 6 6.3

It is worth noting that casinos are located in both Auckland and Christchurch. The study was promoted and practitioners were recruited within the Nelson region, however, no participants were recruited from this area.

135

XGAMBLE Report

Preferred ethnicity Fifty-five percent of the study participants (n=53) self identified as New Zealand European: 29 in the face-to-face group and 24 in the telephone group. There were 20 Maori participants, eight Pacific participants, two Chinese participants and one Indian participant.

Twelve participants chose “Other” as their preferred ethnicity. Four stated “New Zealander”, three Korean, and the rest were American (1), Canadian (1), Lebanese (1), South African (1) and Sri Lankan (1). Being consistent with the term ‘Asian’ used by Statistics New Zealand, the Indian and Sri Lankan and Korean are grouped under ‘Asian’. Table 16 summarises the ethnic distribution of participants.

Table 16: Distribution of participants according to preferred ethnicity (n= 96)

Preferred Ethnicity N %

NZ European 53 55.3 NZ Maori 20 20.8 Pacific Islander

Samoan (3) Cook Island Maori (2) Tongan (2) Niuean (1)

8 8.3

Asian Korean (3) Chinese (2) Indian (1) Sri Lankan (1)

7 7.3

Other New Zealander (4) American (1) Canadian (1) Lebanese (1) South African (1)

8 8.3

Preferred language Ninety-five percent of participants (n=91) had chosen English as their preferred language for a counselling intervention. The distribution was similar between intervention groups. Three participants preferred Korean and two preferred Mandarin (Northern Chinese).

Occupation and current work situation There were more than 50 different occupations listed for participants. Forty-eight percent of participants were working full-time (n=46), 18% part-time (n=17), 7% home-makers (n=7), 5% retired (n=5) and 4% students (n=4). The remaining participants were unemployed or other beneficiaries – see Table 17.

Table 17: Distribution of participants current work situation according to intervention modality (n= 96)

Occupation Intervention Modality Total

136

XGAMBLE Report

Face-to-face (n= 47)

Telephone (n= 49)

(n=96)

n % n % n % Full-time 22 46.8 24 49.0 46 47.9 Part-time 8 17.0 9 18.4 17 17.7 Home-maker 5 10.6 4 8.2 9 9.4 Retired 3 6.4 2 4.1 5 5.2 Student 3 6.4 2 4.1 5 5.2 Unemployed 4 8.5 2 4.1 6 6.3 Beneficiary 2 4.3 6 12.1 8 8.3

Table 18 illustrates the distribution of participants’ employment status according to the five population groups.

Table 18: Distribution of participants’ employment status by preferred ethnicity (n=96)

Occupation Participant Ethnicity

NZ European

(n= 53) (%)

Maori (n= 20)

Pacific (n= 8)

Asian (n= 7)

Other (n= 8)

Full-time 26 (49%) 11 2 2 5 Part-time 10 (19%) 4 2 1 0 Home-maker 4 (7%) 3 1 0 1 Retired 3 (6%) 0 0 1 1 Student 2 (4%) 1 0 2 0 Unemployed 2 (4%) 0 2 1 1 Beneficiary 6 (11%) 1 1 0 0

Qualifications Thirty three percent of study participants did not have a formal qualification (n=32). Thirteen percent had NZ school certificate or national certificate L1 (n=12), and 23% had a university or polytechnic degree or diploma (n=22). Further details are provided in Table 19.

Table 19: Distribution of participants’ highest educational qualifications according to intervention modality (n=96)

Educational Qualification

Intervention Modality Total

(n=96)Face-to-face (n= 47)

Telephone (n= 49)

n % N % n % University or polytechnic degree or diploma 10 21.3 12 24.5 22 22.9

NZ School Certificate, or National Certificate Level 1 6 12.8 6 12.2 12 12.5

Trade Certificate, eg. trade or technician apprenticeship 6 12.8 3 6.1 9 9.4

NZ Sixth Form Certificate, or National Certificate Level 2 4 8.5 1 2 5 5.2

137

XGAMBLE Report

University Entrance qualification from NZ University Bursary 2 4.3 2 4.1 4 4.2

NZ Higher School Certificate, or Higher Leaving Certificate 3 6.4 0 0 3 3.1

NZ University Entrance before 1986 0 0 2 4.1 2 2.1

NZ A or B Bursary, Scholarship, or National Certificate Level 3 0 0 2 4.1 2 2.1

Overseas school qualification 2 4.3 0 0 2 2.1

None 12 25.5 20 40.8 32 33.3

Other 2 4.3 1 2 3 3.1

Missing 0 0 0 0 0 0

Marital Status and Children Thirty-two percent of study participants were married (n=31) and 34% single (n=33). The rest were living with a partner (n=13), separated (n=10), divorced (n=6) and widowed (n=3). A breakdown of marital status by intervention modality is provided in Table 20.

Fifty-nine percent of study participants did not have children and 23% have only one child (n=22).

Table 20: Distribution of participants’ marital status according to intervention modality (n=96)

Marital Status Intervention Modality Total

(n=96)Face-to-face (n= 47)

Telephone (n= 49)

n % n % n % Married 17 36.2 14 28.6 31 32.3 Civil Union 0 0 0 0 0 0 Living with a partner 4 8.5 9 18.4 13 13.5 Single 15 31.9 18 36.7 33 34.4 Separated 6 12.8 4 8.2 10 10.4 Divorced 3 6.4 3 6.1 6 6.3 Widowed 2 4.3 1 2 3 3.1 Did not know 0 0 0 0 0 0 Declined to comment 0 0 0 0 0 0 Missing 0 0 0 0 0 0

Initial knowledge of the Study Sixty participants initially heard about this study from newspapers or magazines. Ten participants initially heard from radio, eight participants from word of mouth, and six received word from friends/relatives. Fewer people heard about this study from flyers (3), problem gambling services (3), TV (3), internet/websites (2) and researchers (1). Details relating to the source of initial knowledge regarding the study are presented in Table 21 according to treatment modality.

138

XGAMBLE Report

Table 21: Source of participants’ initial knowledge of the study by treatment modality (n=96)

Source of Initial Knowledge Intervention Modality Total

(n=96)Face-to-face (n= 47)

Telephone (n= 49)

N % n % n % Flyers 1 2.1 2 4.1 3 3.1 Word of mouth 5 10.6 3 6.1 8 8.3 Friends/relatives 4 8.5 2 4.1 6 6.3 Problem gambling service 1 2.1 2 4.1 3 3.1 Radio 5 10.6 5 10.2 10 10.4 Newspaper/Magazine 29 61.7 31 63.3 60 62.5 TV 1 2.1 2 4.1 3 3.1 Internet/Websites 0 0 2 4.1 2 2.1 Direct contact with researcher 1 2.1 0 0 1 1 Other 0 0 0 0 0 0 Missing 0 0 0 0 0 0

Newspapers/magazines remained the most frequently cited media where participants first heard about this study across the five population groups. If the categories of ‘word of mouth’, ‘friends/ relatives’, and ‘direct contact with researcher’ were combined together to form a ‘personal relationships’ category, they accounted for six NZ European (11.3%) and nine people from the non-NZ European ethnic groups (20.9%). Within the combined ‘personal relationships’ category, 70.5% of participants were females. It is also noted that ‘radio’ did not recruit any non-NZ European participants for this study. See Table 21 for details.

Table 22: Distribution of participants’ initial knowledge of the study according to preferred ethnic group (n=96)

Source of Initial Knowledge

Participant Ethnicity NZ

European (n= 53)

Maori (n= 20)

Pacific (n= 8)

Asian (n= 7)

Other (n= 8)

Flyers 1 1 1 0 0 Word of mouth 4 1 2 0 1 Friends/relatives 2 2 0 2 0 Problem gambling service 1 0 0 2 0 Radio 9 0 0 0 1 Newspaper/Magazine 35 13 5 2 5 TV 1 1 0 0 1 Internet/Websites 0 1 0 1 0 Direct contact with researcher 0 1 0 0 0 Other 0 0 0 0 0 Missing 0 0 0 0 0

Treatment for Alcohol, Drug and Other Mental Health Issues

139

XGAMBLE Report

Ninety four percent of participants (n=90) did not receive treatment for any alcohol and/or drug issues during the study. Seventy eight percent of participants (n=75) did not receive treatment for any mental health issues during the study. Twenty-one participants received treatment for depression (11), anxiety (3), epilepsy/seizures (1), schizophrenia (1) and stress (1). Three participants did not report what treatment they have received. Table 23 provides further details on these variables.

Table 23: Distribution of participants receiving treatment for alcohol and/or drug and mental health issues

Other Treatment Intervention Modality Total

(n=96)Face-to-face (n= 47)

Telephone (n= 49)

N % n % n % Treatment for alcohol and/or drug issues

Yes 3 6.4 3 6.1 6 6.3 No 44 93.6 46 93.9 90 93.8 Missing 0 0 0 0 0 0

Treatment for mental health issues Yes 10 21.3 11 22.4 21 21.9 No 37 78.7 38 77.6 75 78.1 Missing 0 0 0 0 0 0

Specify if "Yes" Anxiety 1 10 2 18.2 3 14.3 Depression 6 60 5 45.5 11 52.4 Epilepsy/seizures 1 10 0 0 1 4.8 Not specified 2 20 2 18.2 4 19 Schizophrenia 0 0 1 9.1 1 4.8 Stress 0 0 1 9.1 1 4.8

With regard to mental health issues and ethnicity: 12 NZ European, five Maori, two Pacific and two Asian participants were receiving treatment for mental health issues. This distribution is outlined below in Table 24.

Table 24: Distribution of participants receiving treatment for mental health issues across five groups of preferred ethnicity (n=96)

Receiving treatment for mental health issues

Participant Ethnicity NZ

European n= 53 (%)

Maori n= 20 (%)

Pacific n= 8 (%)

Asian n= 7 (%)

Other n= 8(%)

Yes 12 (23%) 5 (25%) 2 (25%) 2 (29%) 0 No 41 (77%) 15 (75%) 6 (75%) 5 (71%) 8 (100%) Missing 0 0 0 0 0

Ninety six percent of participants (n=92) did not receive treatment for any serious head injury, one participant had received treatment for a car accident and another one

140

XGAMBLE Report

for medical mishap. Two participants did not specify what treatment they have received.

Problem gambling counselling Forty one percent of participants (n=39) had previously received counselling for their problem gambling. Three participants did not answer this question. The response distributions for responses to this item are provided in Table 25, including intervention modality.

Table 25: Distribution of participants receiving problem gambling counselling (n=96)

Have had prior counselling for problem gambling

Intervention Modality Total (n=96)Face-to-face

(n= 47) Telephone

(n= 49) n % n % n %

Yes 20 42.6 19 38.8 39 40.6 No 26 55.3 28 57.1 54 56.3 Missing 1 2.1 2 4.1 3 3.1

Table 26 shows that the number of participants who had received problem gambling counselling prior to this research was distributed evenly between NZ European and non-NZ European groups: 41.5% and 39.5% respectively. However Pacific People were less likely to have had problem gambling counselling prior to the present study.

Table 26: Distribution of participants receiving problem gambling counselling across five groups of preferred ethnicity (n=96)

Have had prior Participant Ethnicity counselling for problem gambling

NZ European n= 53 (%)

Maori n= 20 (%)

Pacific n= 8 (%)

Asian n= 7 (%)

Other n= 8( %)

Yes 22 (42%) 11 (55%) 1 (13%) 2 (28%) 3 (37%) No 30 (67%) 7 (35%) 7 (87%) 5 (72%) 5 (63%) Missing 1 (1%) 2 (10%) 0 0 0

Annual Household Income Sixty-one participants (64%) indicated that their annual household income (before tax) was $50,000 or less. Twenty-seven participants cited an annual household income of between $50,001 and $100,000, and six participants reported that their income was greater than $100,000. Two participants indicated ‘Don’t know’ for this question. A breakdown of annual household income (before tax) by intervention modality is provided in Table 27

Table 27: Distribution of participants’ annual household income according to intervention modality (n=96)

Total annual household income Intervention Modality Total (n=96)before tax (Q2.34) Face-to-face

(n= 47) Telephone

(n= 49) n % n % n %

141

XGAMBLE Report

$20,000 or less 13 27.7 8 16.3 21 21.9 $20,001 to $30,000 6 12.8 11 22.4 17 17.7 $30,001 to $40,000 11 23.4 6 12.2 17 17.7 $40,001 to $50,000 1 2.1 5 10.2 6 6.3 $50,001 to $70,000 9 19.1 8 16.3 17 17.7 $70,001 to $100,000 3 6.4 7 14.3 10 10.4 $100,001 to $120,000 2 4.3 1 2 3 3.1 $120,001 or more 1 2.1 2 4.1 3 3.1 Do not know 1 2.1 1 2 2 2.1 Declined to comment 0 0 0 0 0 0 Missing 0 0 0 0 0 0

Table 28 shows that there was a peak frequency for the NZ European’s level of earning at the “$20,000 or less” and “$50,000 to $70,000” range whereas this trend did not apply for the non-NZ European population groups. Also there were few people in the Maori, Pacific Islander and Asian group earning more than $70,000 household income.

Table 28: Distribution of participants’ annual household income across five groups of preferred ethnicity (n=96)

Total annual household income before tax (Q2.34)

Participant Ethnicity NZ

European (n= 53)

Maori (n= 20)

Pacific (n= 8)

Asian (n= 7)

Other (n= 8)

$20,000 or less 12 7 1 0 1 $20,001 to $30,000 7 3 2 2 3 $30,001 to $40,000 8 4 2 2 1 $40,001 to $50,000 2 2 1 1 0 $50,001 to $70,000 13 2 1 1 0 $70,001 to $100,000 7 1 0 0 2 $100,001 to $120,000 2 0 0 0 1 $120,001 or more 2 1 0 0 0 Do not know 0 0 1 1 0 Declined to comment Missing

Self-exclusions from Venues Eight participants (8%) stated that they have been self-excluded or banned from entering a gambling venue within the last two months. The table below (Table 29) provides a brief profile of these participants. With one exception, all self-excluded participants were female. They tended to reside in Auckland and the majority were NZ European. All of them were primarily concerned with their gambling in pokies: half in relation to pokies in pubs, and half in relation to casino pokies.

Table 29: Profile of self-excluded or banned participants

Participant Demographics Details Relating to Gambling

142

XGAMBLE Report

Location Gender (M/F)

Preferred ethnicity

Mode of primary concern

Total money spent in the last 4 weeks

Money spent in

proportion to income#

(before tax) Auckland M Asian pokie in casino $1,800 150% Auckland F NZ European pokie in pub 800 80% Auckland F Maori pokie in pub 2,400 80% Christchurch F NZ European pokie in pub 800 44% Auckland F Maori pokie in casino 2,100 35% Auckland F NZ European pokie in casino 2,400 55% Auckland F NZ European pokie in casino 2,250 23% Christchurch F NZ European pokie in pub 4,800 480%

# In Form A, participants were asked about “total annual household income before tax” in terms of cateogries (Q2.34, see Table 28). In Form B (baseline, post-intervention and 6-week follow up), participants were asked about their household income in dollar terms: “Could you please tell me what your total household income (before tax) from all sources was over the last four weeks?” (Q3.05)

Gambling Activity of Most Concern Sixty-eight percent (n=65) of participants indicated that pokie machines in a class 4 gambling venue (Under the Gambling Act 2003) concerned them the most. Thirty-two percent (n=31) of participants specified ‘casino pokie machines’ as concerning them the most. Table 30 portrays the distribution of participants’ primary mode of concern in relation to intervention modality.

Note that participants must have chosen either one of these three choices to be eligible in this study. Also note that the majority of participants resided in either the Auckland or Christchurch locations, where main casinos are located.

Table 30: Participants’ primary problem gambling mode by intervention modality (n=96)

Gambling activity most concerned by (Q2.36)

Intervention Modality Total (n=96)Face-to-face

(n= 47) Telephone

(n= 49) n % n % n %

Casino pokie machines 16 34 15 30.6 31 32.3 Pokie machines in bars/pubs 28 59.6 30 61.2 58 60.4 Pokie machines in clubs 3 6.4 4 8.2 7 7.3

Table 31 shows that proportionately more NZ European participants were concerned about their pokie gambling in bars/pubs and clubs than their non-NZ European counterparts. However, Maori tended to play pokies evenly between casinos and bars/pubs. Pacific tended to stay more in the pubs/bars, whereas Asians tended to play in the casinos slightly more.

Table 31: Distribution of participants primary problem gambling mode across five groups of preferred ethnicity (n=96)

143

XGAMBLE Report

NZ European (n= 53) (%)

Maori (n= 20)

Pacific Islander (n= 8)

Asian (Korean, Chinese & South Asian) (n= 7)

Other (n= 8)

Casino pokie machines (n=31) 14(26%) 9(45%) 1(12%) 5(71%) 2(25%) Pokie machines in bars/pubs (n=58) 34(64%) 10(50%) 7(88%) 2(29%) 5(63%) Pokie machines in clubs (n=7) 5(10%) 1(5%) 0 0 1(12%)

Table 32 shows that those participants who gambled in clubs, spent less money (median= $800 or 50% of a person’s household income before tax) in gambling compared to participants who gambled in casinos or bars/pubs. But it was still a significant amount of money in dollar term or in portion to average income level observed in this study. In Form A, participants were asked about “total annual household income before tax” (Q2.34). In Form B (baseline, post-intervention and 6-week follow up), participants were asked about their household income in dollar terms, “Could you please tell me what your total household income (before tax) from all sources was over the last four weeks?” (Q3.05). It also highlights that a higher proportion of women cited pokie machines in clubs as the mode of gambling that they were concerned about.

Table 32: Distribution of money and proportion of income spent gambling and gender according to gambling mode (n=96)

Mode of Gambling Money spent gambling Proportion of income spent gambling

Gender breakdown: number of women (%)mean median mean median

Casino pokie machines (n=31) $4,456 $2,325 181% 55% 23 (74.2%) Pokie machines in bars/pubs (n=58) $7,638 $1,500 142% 68% 35 (60.3%) Pokie machines in clubs (n=7) $1,060 $800 92% 50% 6 (85.7%)

Practitioners A total of 20 practitioners were available to provide services for this research. Six were New Zealand European (three each in Auckland and Christchurch), eight were Maori (three each in Auckland and Hawkes Bay, two in Nelson), three were Pacific (Auckland) and three were Asian providers (two in Auckland and one in Christchurch).

144

XGAMBLE Report

Descriptive summary of primary/secondary measures at baseline, post-intervention and follow up visits This section provides a descriptive summary of the five primary/secondary outcome measures.

Baseline measures of primary and secondary outcomes Ninety-two of the participants who were randomised to face-to-face or telephone interventions (46 in each group) have baseline information. The following findings relate to these participants. Table 33 provides a summary of findings from primary and secondary baseline measures in relation to intervention mode.

Total money ($) spent on gambling This outcome was derived by multiplying money spent on one gambling session (Q3.01 in Form B) with total number of sessions spent gambling (Q3.03) over the last four weeks. The values ranged between $0 and $120,000, with an overall mean of $4,495 and a median of $1,550.

For those in a face-to-face intervention group, their average total money spent in gambling over the last four weeks was $2,505 with a median of $960. The maximum value was $20,000.

For those in the telephone intervention group, their average total money spent in gambling over the last four weeks was $6,484 with a median of $2,050. The maximum value was $120,000. Note that the summary values were much higher than those from the face-to-face group.

Total time (hours) spent in gambling This outcome was derived by multiplying hours spent on one gambling session (Q3.02 in Form B) with total number of sessions spent gambling (Q3.03) over the last four weeks. The values ranged from 0 to 240 hours, with an overall mean of 30 hours and a median of 16 hours.

For those in the face-to-face intervention group, their average total hours spent in gambling over the last four weeks was 21 hours with a median of 14.5 hours. The maximum value was 84 hours in total.

For those in the telephone intervention group, their average total hours spent in gambling over the last four weeks was 39 hours with a median of 20.5 hours. The maximum value was 240 hours. Again, the values were higher than those in the face-to-face group.

Proportion of total money ($) spent gambling As this outcome depended upon the participants’ total household income before tax over the last four weeks, which may be “Declined” in the Form B, the value became missing when participants declined this information.

Although a proportion should normally lie between 0 to 100%, it is no longer applicable to this outcome as 25 participants stated a lower total household income than their total spending in gambling over the last four weeks at baseline. The values ranged from 0% to 2,400% (i.e. spending 24 times as much as their total income in gambling).

145

XGAMBLE Report

Similar to the two primary outcome measures, participants randomized to the face-to-face intervention had a much lower average percent of total money spent gambling over last four weeks at baseline compared with those in the telephone intervention (94% versus 207%). NB: a value of 207% explains that the gambler has spent 2 times the amount of money gambling than their income over the last four weeks.

Gambling Attitudes and Beliefs (GABS) score The GABS score was calculated from 35 related questions, each scaled from 1 to 4. The final score ranged from 35 to 140, with higher values indicating a higher inclination to gamble.

The GABS scores evaluated at baseline ranged from 66 to 129, with an overall mean of 92 and a median of 93. The average GABS scores at baseline were similar, regardless of intervention mode (92.0 for face-to-face versus 92.3 for telephone).

Gambling Readiness To Change (GRTC) score The GRTC score was calculated from 9 questions (each scaled from 1 to 5) separated into three sections: (1) Contemplation; (2) Action; and (3) Pre-contemplation. Each section had its own total score (see Table 33 for their summaries). The final GRTC score is a weighted sum of three sections ranged from -21 to 39. Higher GRTC scores indicate a greater inclination to change. The average total scores were 18.6 and 21.3 for face-to-face and telephone intervention groups at baseline respectively.

146

XGAMBLE Report

Table 33: Descriptive summary of primary and secondary measures - baseline (n=92)

Intervention Modality Total

(n=92)Face-to-face (n= 46)

Telephone (n= 46)

mean (range)

SD median mean (range)

SD median mean (range)

SD median

Primary Outcomes

Total $ spent gambling over the last four weeks

$2,505 ($0-$20,000)

3904.9 $960 $6,484.1 ($0-$120,000)

18,109.1 $2,050 $4,495 ($0-$120,000)

13,179.8 $1,550

Total hours spent gambling over the last four weeks

20.7 (0-84)

20.4 14.5 39.0 (0-240)

56.1 20.5 29.9 (0-240)

43.0 16

Secondary Outcomes

% of total $ spent gambling to total income over the last four weeks (n= 44 for 'Face-to-Face' ; n= 43 for 'Phone')

93.7% (0%-480%)

112.5 52 207.1% (0%-2400%)

475.8 75 149.7% (0%-2400%)

346.6 64%

GABS score (ranged from 35 to 140)*

92.0 (69-129)

11.2 92 92.3 (66-122)

12.8 93.5 92.2 (66-129)

11.9 93

GRTC score (ranged from -21 to 39)**

18.6 (5-31)

5.8 18 21.3 (3-33)

7.4 22.5 19.9 (3-33)

6.7 20

• Contemplation (3 to 15) (Qs 1, 2 & 6)

13.2 (9-15)

1.6 14 13.6 (8-15)

1.7 14 13.4 (8-15)

1.7 14

• Action (3 to 15) (Qs 8, 4 & 5)

10.4 (4-15)

2.4 10 11.3 (6-15)

2.1 12 10.9 (4-15)

2.3 10.5

• Pre-Contemplation (3 to 15) (Qs 3, 7 & 9)

7.7 (3-11)

1.9 8 7.4 (3-15)

2.4 7 8.0 (3-15)

2.1 8

Note: * GABS - higher score indicating higher inclination to gamble; ** GRTC - higher score indicating more ready to change

It should be noted that correlation analyses were performed on between a number of demographic and characteristic variables, for example, previous experience of receiving counselling for gambling, employment status, hours and money spent gambling and gambling attitudes and behaviours were performed. However, as none

147

XGAMBLE Report

of these analyses reached a level of statistical significance they have not been reported in this report.

Post-intervention measure of primary and secondary outcomes Only 27 randomised participants completed their post-intervention forms; resulting in a retention rate of 29.4%. Therefore the following results have to be interpreted with caution. Table 34 provides a summary of findings from primary and secondary post-intervention measures in relation to intervention mode.

Total money ($) spent on gambling The money spent on gambling over the last four weeks ranged from $0 to $16,000, with an overall mean of $1,271 and a median of $20.

For those in the face-to-face intervention group (n=14), the average total money spent on gambling over the last four weeks was reduced to $2,332 with a median of $140. The maximum value was $16,000. For those in the telephone intervention group (n=13), their average total money spent on gambling over the last four weeks was $128 with a median of $0. The maximum value was $800.

It is noted that the baseline summary for the telephone group was much higher than that of the face-to-face group. At the post-intervention visit, however, the trend was reversed for these participants based on their observed information.

Total time (hour) spent gambling The time (hour) spent over the last four weeks ranged from 0 to 50 hours, with an overall mean of 6.3 hours and a median of 0.2 hours.

For those in the face-to-face intervention group (n=14), their average total hours spent gambling over the last four weeks was reduced to10 hours with a median of 1 hour. The maximum value was 50 hours. For those in the telephone intervention group (n=13), their average total hours spent gambling over the last four weeks was 2 hours with a median of 0 hour. The maximum value was 16 hours.

Consistent with the previous primary outcome of interest – ‘Money spent in gambling over the last four weeks’, the baseline summary for the telephone group was much higher than that for the face-to-face group, however the trend was reversed for these participants at the post-intervention visit.

Proportion of total money ($) spent gambling The average proportions were reduced to 20% and 3% at the post-intervention visit for face-to-face and telephone groups respectively. Again, the trend was reversed between the two groups compared to their baseline summaries.

Gambling Attitudes and Beliefs (GABS) score The average total scores for face-to-face and telephone intervention groups decreased to 82 and 79.8 respectively at post-intervention. The higher scores indicate a higher inclination to gamble.

148

XGAMBLE Report

Gambling Readiness To Change (GRTC) score The average total scores for face-to-face and telephone intervention groups increased to 21.6 and 22.5 respectively at post-intervention. The higher scores indicate a higher level of readiness to change.

Table 34: Descriptive summary of primary and secondary measures - post intervention (n=27)

Intervention Modality Total (n=27)Face-to-face

(n= 14) Telephone

(n= 13) mean

(range) SD median mean

(range) SD median mean

(range) SD median

Primary Outcomes

Total $ spent gambling over the last four weeks

$2,332.1 ($0-$16,000)

5052.1 $140 $128.5 ($0-$800)

236.0 $0 $1,271 ($0-$16,000)

3,747.9 $20.0

Total hours spent gambling over the last four weeks

10.3 (0-50)

17.0 1 2.0 (0-16)

4.5 0 6.3 (0-50)

13.1 0.2

Secondary Outcomes

% of total $ spent gambling to total income over the last four weeks (n= 12 for 'Face-to-Face' ; n= 11 for 'Phone')

19.8% (0%-80%)

27.4 5 3.0% (0%-12.5%)

4.5 0 11.8% (0%-80%)

21.4 4%

GABS score (ranged from 35 to 140)

82.0 (39-114)

18.6 84.0 79.8 (44-105)

15.2 79.0 80.9 (39-114)

16.8 81.0

GRTC score (ranged from -21 to 39)

21.6 (9-38)

7.6 19.5 22.5 (13-37)

7.8 20.0 22.1 (9-38)

7.6 20.0

• Contemplation (3 to 15) (Qs 1, 2 & 6)

11.6 (3-15)

3.1 13.0 10.1 (4-15)

3.4 9.o 10.9 (3-15)

3.3 12.0

• Action (3 to 15) (Qs 8, 4 & 5)

12.6 (10-15)

1.4 12.0 13.2 (10-15)

1.8 13.0 12.9 (10-15)

1.6 13.0

• Pre-Contemplation (3 to 15) (Qs 3, 7 & 9)

7.6 (3-12)

2.5 8.0 6.9 (3-10)

2.1 7.0 7.3 (3-12)

2.3 8.0

149

XGAMBLE Report

Follow up measure of primary and secondary outcomes Only 12 randomised participants returned their follow up forms. Only two participants in the face-to-face group returned their forms while ten in the telephone group returned their Form B at the follow up visit.

This has made interpretation almost impossible for the follow up measure. Therefore the following results have to be interpreted with extreme caution and further analyses on the follow up data were not performed. The following Table 35 records the descriptive summary of the follow up data.

Table 35: Descriptive summary of primary and secondary measures - follow up (n=12)

Intervention Modality Total (n=12)Face-to-face

(n= 2) Telephone

(n= 10) mean

(range) SD median mean

(range) SD median mean

(range) SD median

Primary Outcomes

Total $ spent gambling over the last four weeks

NA ($0-$4,000)

NA NA $71.1 ($0-$400)

132.0 $0.5 $392.6 ($0-$4,000)

1,142.5 $0.5

Total hours spent gambling over the last four weeks

NA (0-20)

NA NA 0.8 (0-3)

1.1 0.1 2.3 (0-20)

5.7 0.1

Secondary Outcomes

% of total $ spent gambling to total income over the last four weeks (n= 12 for 'Face-to-Face' ; n= 11 for 'Phone')

NA (0%-27%)

NA NA 3.8% (0%-25%)

8.2 0 5.5% (0%-27%)

10.2 0%

GABS score (ranged from 35 to 140)

NA (70-87)

NA NA 61.5 (42-79)

12.8 64.0 64.3 (42-87)

13.8 66.5

GRTC score (ranged from -21 to 39)

NA (10-24)

NA NA 16.1 (3-32)

9.5 17.0 16.3 (3-32)

9.1 17.0

• Contemplation (3 to 15) (Qs 1, 2 & 6)

NA (12-12)

NA NA 7.1 (3-14)

4.0 6.5 7.9 (3-14)

4.1 8.5

• Action (3 to 15) (Qs 8, 4 & 5)

NA (7-13)

NA NA 12.0 (3-15)

3.7 12.0 11.7 (3-15)

3.7 12.0

• Pre-Contemplation (3 to 15) (Qs 3, 7 & 9)

NA (7-8)

NA NA 7.5 (3-15)

3.5 7.5 7.5 (3-15)

3.2 7.5

150

XGAMBLE Report

Descriptive summary of other relevant measures at baseline and post-intervention visits

Sense of control Participants randomly allocated to the face-to-face and telephone intervention groups gave similar levels of ratings on their sense of control over their gambling behaviours. At the post-intervention visits, both of the intervention groups demonstrated a dramatic increase in the percentage of participants who provided a self rating of either ‘completely in control’ or ‘mostly in control’. See Table 36 for details.

No further report on the follow up visits is provided here due to the very low response rate.

Table 36: Sense of control of gambling at baseline and post-intervention by intervention modality

Intervention Modality Baseline measure

(n=46) Post-intervention

(n=14) n % N %

Face-to- face

completely in control

1 2 2 14

mostly in control

8 18 9 64

mostly out of control

19 41 2 15

completely out of control

18 39 1 7

missing 0 0 0 0

Baseline measure (n=46)

Post-intervention (n=13)

n % N %

Telephone completely in control

2 4 4 31

mostly in control

8 18 7 54

mostly out of control

22 48 2 15

completely out of control

14 30 0 0

missing 0 0 0 0

151

XGAMBLE Report

Effectiveness of intervention (anticipated and experienced) Table 37 illustrates that participants randomly allocated to either the face-to-face or telephone intervention group provided similar levels of rating on anticipated effectiveness of intervention at baseline. Most participants anticipated the effectiveness of the intervention to be ‘somewhat effective’, ‘effective’, or ‘very effective’.

Table 37: Anticipated effectiveness of intervention at baseline measure by intervention modality

Baseline measure: anticipated effectiveness

Intervention Modality

Face-to-face (n=46)

Telephone (n=46)

n % n % NOT be effective 2 4 1 2 be somewhat effective 9 20 12 26 effective 18 39 16 35 be very effective 12 26 10 22 be extremely effective 5 11 7 15 missing 0 0 0 0

Similarly, Table 38 illustrates that there were no major differences in ratings in relation to the experienced effectiveness of intervention between the face-to-face or the telephone intervention group at post-intervention visits. One minor difference was that slightly more participants from the telephone group regarded the services they received as ‘extremely effective’.

Table 38: Experienced effectiveness of intervention at post-intervention measure by intervention modality

Post-intervention measure: experienced effectiveness

Intervention Modality

Face-to- face (n=14)

Telephone (n=13)

N n NOT effective 2 1 somewhat effective 3 2 effective 0 3 very effective 6 2 extremely effective 3 5 missing 0 0

Satisfaction of intervention services provided Overall, participants provided similar ratings for level of satisfaction of the interventions they received, regardless of intervention group: face-to-face or telephone. One difference however was that more participants from the face-to-face group indicated that they were ‘extremely satisfied’ with their intervention services. The response distributions for this item are displayed in Table 38.

152

XGAMBLE Report

Table 39: Level of satisfaction with provided intervention service by intervention modality

Level of satisfaction Intervention Modality

with provided intervention

Face-to- face (n=14)

Telephone (n=13)

n n NOT satisfied 2 1 somewhat satisfied 1 0 satisfied 2 2 very satisfied 2 7 extremely satisfied 7 3 missing 0 0

Accessibility to services As illustrated in Table 40, participants provided similar ratings for accessing the interventions they received, regardless of intervention group: face-to-face or telephone.

Table 40: Accessibility to services by intervention modality

Accessibility to services

Intervention Modality

Face-to- face (n=14)

Telephone (n=13)

n n yes 13 11 no 1 2 missing 0 0

Number of sessions attended Thirty-eight percent of participants attended six or more sessions during the study period. Thirty-one percent of participants either failed to turn up or could not be contacted for any of the face-to-face or telephone sessions. Further information was sought in order to explain why these participants did not turn up for counselling sessions. It was found that, for example, the participant’s phone number had changed or they initially provided us with the wrong contact phone number.

153

XGAMBLE Report

Table 41: Number of intervention sessions attended

Number of sessions attended

Number of ‘face-to-face’ clients (n= 46)

% Number

of ‘phone’ clients (n=46)

%

Total percentage

(n=92)

1 8 18 8 18 18 2 1 2 1 2 2 3 3 6 2 4 5 4 2 4 2 4 4 5 2 4 0 0 2 6 14 31 14 31 31 >6# 4 9 2 4 7 did not attend any sessions and/or was non-contactable

12 26 17 37 31

# attended between seven and eight sessions during study period

It is worth noting that participants who thought their intervention would be ‘extremely effective’ (i.e. anticipated effectiveness) attended more counselling sessions than their counterparts (see Table 42 for details). However, no such trends were observed for GRTC or GABS scores.

Table 42: Number of intervention sessions attended according to anticipated effectiveness of intervention

Anticipated effectiveness of intervention

Number of participants

Average number of sessions

not effective 3 0 somewhat effective 21 2.6 effective 34 2.1 very effective 22 2.7 extremely effective 12 3.7

Correlations among primary and secondary outcome measures Table 43 shows there were strong to moderate positive Pearson correlations among the ‘money spent gambling’, ‘time spent gambling’ and ‘proportion of money spent gambling’. GABS has moderate to low positive correlations with money and time spent gambling; whereas GRTC did not show any stable correlations with each of the four outcome measures.

Table 43: Correlations among primary and secondary outcome measures

money time Percentage GABS GRTC money 1.0 time 0.61*** 1.0 percentage0.24* 0.36*** 1.0 GABS 0.19* 0.26** 0.14 1.0 GRTC -0.15 -0.02 -0.12 0.12 1.0

154

XGAMBLE Report

Noted: *< .1; **< .01; ***< .001

Descriptive summary of practitioners’ notes at midway review and termination of counselling In order to gain a better understanding of what might be contributing to the outcomes of problem gambling interventions practitioners were asked to provide notes about their sessions. Table 44 shows there was a good match between the problems presented in midway reports and the outcomes achieved over the course of interventions. The techniques utilised by practitioners could be broadly categorised into two domains: 1) is related to provision of information, increasing clients’ level of understanding about problem gambling, monitoring own progress and changing one’s cognition; and, 2) is related to changing behavioural patterns and preventing relapses.

Table 44: Problems (and outcomes perceived by practitioners) and intervention techniques used

Third session or midway review (n=39) Review upon termination of counselling (n=39)

Items Ranking of endorsement (number. of practitioners endorsed that item)

Items Ranking of endorsement (number of practitioners endorsed that item)

Presenting problems perceived by practitioners

Positive outcomes perceived by practitioners

Gambling 1st (26) Gambling 1st (25) Finance 2nd (18) Finance 2nd (23) Self-esteem 3rd (13) Self-esteem 3rd (22) Emotional health 4th (12) Hope 4th (21) Relationship/ partners

5th (11) Emotional health 5th (15)

Interventions used by practitioners Interventions used by practitioners

Positive feedback 1st (25) Life style 1st (24) Gambling education 2nd (22) Gambling education 1st (24) Review progress 3rd (17) High risk 1st (24) Safety plan 4th (16) Positive feedback 2nd (23) Life style 5th (15) Goal setting 2nd (23) Instil hope 5th (15) Review Progress 3rd (21) Cognitive distortion 4th (19) Early warning 4th (19) Safety plan 4th (19)

Insight 5th (18)

155

XGAMBLE Report

Direct comparison between groups In the following section, we use scatter plots with raw means and 95% confidence intervals for direct between-group comparisons.

For the purpose of presentation and part of data cleaning, extreme high values (outliers) are not presented in these figures as they distort distributions, making comparisons difficult. The presence of skewed data or outliers in a dataset may also seriously distort the interpretation. Therefore extreme values (outliers) were removed before analysis to reduce its potential effect on the distribution, which is a key assumption in parametric modelling. It is also important to bear in mind that: 1) there was a large amount of information missing at post-intervention and follow up visits; and 2) The wide confidence intervals at follow up visits for the face-to-face group are due to the small number of forms collected from participants (n=2 only).

Total money ($) spent on gambling The scatter plot with raw means and 95% confidence intervals in Figure 8 clearly shows the reversed trend between groups (telephone versus face-to-face intervention) from baseline to post-intervention and follow up visits. While participants in the telephone intervention group had higher mean scores (of money spent gambling) at baseline, these dropped dramatically at post-intervention and follow up compared with those observed in the face-to-face group.

Figure 8: Total money spent on gambling at baseline, post-intervention, and follow up by intervention modality

0 50

00

1000

0 15

000

Mean and 95% C.I. using Observed Data

Period

Tota

l $ s

pent

gam

blin

g ov

er th

e la

st fo

ur w

eeks

Baseline Post-intervention Follow-up

Face-to-face Phone

Similar figures are provided below (Figure 9 - Figure 12) for the rest of the primary and second outcome measures. Large confidence intervals can make the change in means observed over time difficult to interpret; however the overlap of the mean scores between the face-to-face and telephone groups is still apparent.

156

XGAMBLE Report

Figure 9: Total time (hour) spent on gambling at baseline, post-intervention, and follow up by intervention modality

-20

0 20

40

60

80

10

0

Mean and 95% C.I. using Observed Data

Period

Tota

l hou

rs s

pent

gam

blin

g ov

er th

e la

st fo

ur w

eeks

Baseline Post-intervention Follow-up

Face-to-face Phone

Figure 10: Proportion of total money spent on gambling at baseline, post-intervention, and follow up by intervention modality

-50

0 50

10

0 15

0 20

0

Mean and 95% C.I. using Observed Data

Period

% o

f tot

al $

spe

nt g

ambl

ing

over

the

last

four

wee

ks

Baseline Post-intervention Follow-up

Face-to-face Phone

157

XGAMBLE Report

Figure 11: Gambling Attitudes and Beliefs (GABS) score at baseline, post-intervention, and follow up by intervention modality

0 50

10

0 15

0

Mean and 95% C.I. using Observed Data

Period

GA

BS

Sco

re

Baseline Post-intervention Follow-up

Face-to-face Phone

Figure 12: Gambling Readiness To Change (GRTC) score at baseline, post-intervention, and follow up by intervention modality

0 1

0 20

30 40

50

Mean and 95% C.I. using Observed Data

Period

GR

TC S

core

Baseline Post-intervention Follow-up

Face-to-face Phone

158

XGAMBLE Report

Repeated measures analysis Extreme values (outliers) have been removed before analysis to reduce their potential effect on the distribution, which is a key assumption in parametric modelling.

We first carried out a repeated measures analysis using all observed information at each visit. Due to the limited amount of information collected at the follow up visit, we have not focused on this set of data in the analysis. Both analyses with and without the interaction term between GROUP and PERIOD were considered.

Comparison of treatment effect between interventions There was no evidence of a significant difference between the two intervention groups in all outcome measures using the observed information. Differences between intervention groups at each visit were also statistically non-significant. This is indicated by large p-values (> 0.05) associated with the PERIOD and GROUP interaction.

Period effect for each intervention For both primary and secondary outcomes, we found a significant overall PERIOD effect between visits. This is indicated by an associated p-value of < 0.05 for the term PERIOD (for details, see XGAMBLE Appendices Booklet). The PERIOD effect between baseline and post-intervention visits was found for all outcome measures except GRTC by performing a series of simple Student’s t-test supported by Wilcoxon non-parametric tests, with combined groups without imputation and with outliers removed. For example, mean (+ std dev) of proportion of total money spent gambling were 101.5%+141.1 at baseline; decreased to 11.8% (+21.4) at post-intervention visits. Table 45 provides these details.

Table 45: Period effects between baseline and post-intervention visits by outcome measure

Outcome measure

Baseline measure Post-intervention Period effect (level of significance based on p-value)

mean (n)

std dev (range)

mean (n)

std dev (range)

% of total money gambling to income

101.5 (85)

141.1 (0-937.5)

11.8 (23)

21.4 (0-80) **

total hrs spent (4 wks)

23.3 (89)

23.7 (0-108)

6.3 (27)

13.0 (0-50) ***

GABS 92.2 (92)

11.9 (66-129)

80.9 (27)

16.8 (39-114) ***

total money spent (4 weeks)

3225.7 (91)

5076.6 (0-30,000)

1271.1 (27)

3747.9 (0-16000) *

GRTC 19.9 (92)

6.7 (3-33)

22.1 (27)

7.6 (9-38)

NS

Note: * p< .05; ** p< .005; *** p< .0005. NS- non-significant

Although with each of the interventions, our observed data analysis indicates a trend of reduced money, time spent gambling gambling-related cognition over the last four

159

XGAMBLE Report

weeks with reduced inclination to gamble during the course of study (for example, see Figures 13 and 14), it is difficult to ascertain whether or not such a change in behaviour was due to the effectiveness of the intervention itself without a control group for comparison. It should be noted that the overall findings need to be viewed with caution given the large amount of missing data in the present study.

Figure 13: Total time (hour) spent on gambling, model-adjusted means and their 95% confidence intervals from the regression for each group at each visit by intervention modality

Model Adjusted Means and 95% C.I. using Observed Data

Tota

l hou

rs s

pent

gam

blin

g ov

er th

e la

st fo

ur w

eeks

0 1

0 20

30

Face-to-face Phone

Baseline Post-intervention Follow-up

Period

Figure 14: GAB scores, model-adjusted means and their 95% confidence intervals from the regression for each group at each visit by intervention modality

Model Adjusted Means and 95% C.I. using Observed Data

GA

BS

Sco

re

50

60

70

80

90

100

Face-to-face Phone

Baseline Post-intervention Follow-up

Period

160

XGAMBLE Report

Repeated measures analysis (after imputation) A sensitivity analysis was proposed earlier in the Statistics Analysis Plan to carry out the same repeated measures analysis using the coded data. For each participant, the post-intervention measure was coded by the baseline value if it was missing. The follow up measure was further computed from the average of the previous two measures if it was also missing.

As the loss-to- follow up rates were 70% and 96% at post-intervention and follow up visits for the face-to-face intervention group, and 72% and 78% for the telephone intervention group, there is very limited value to proceeding with the sensitivity analysis.

Study power The original sample size proposed was 200 participants in total, with 100 randomised to each intervention group.

Due to the difficulty of recruitment, the final sample size was 92 eligible participants, with 46 in each of intervention groups for final analysis.

The loss to follow up rate was very high in this study, with 70% and 96% at post-intervention and follow up visits for the face-to-face intervention group, and 72% and 78% for the telephone intervention group. This loss has implications for the reliability of the findings.

The limited amount of observed information also affected the implementation of the simple analysis we proposed in the Statistics Analysis Plan. Most of the baseline values were ‘copied’ into the subsequent measures and therefore artificially reduced the variability of data. Other statistical analysis will also be limited as any alternative analysis depends upon the observed information that has been collected.

Due to the limited quality of data, re-evaluation of the study power is no longer appropriate or practical for this study.

Non-parametric test As some of the outcome measures were not normally distributed, with several potential outliers, we also carried out simple non-parametric tests on change from baseline to post-intervention measures. We first evaluated the change between baseline and post-intervention for each intervention group without and with the imputation. Note that, without imputation, there are only 14 observations available for ‘Face-to-Face’ intervention group and 13 for ‘Phone’ intervention group. After imputation, the sample size increases to 46 per group as measured at baseline, with simply more zeros being imputed (no change from baseline). The following shows one example of analysis.

Change at post-intervention measures Face-to-Face: The median change at post-intervention in total money spent over last four weeks is -$900 with values ranging from $0 to -$6000. Non-parametric testing shows a significant reduction (p-value 0.0002) in money spent at post-intervention compared with the baseline.

After imputation, the median change at post-intervention in total money spent over last four weeks increases to $0 with the same range of values. Non-parametric testing

161

XGAMBLE Report

shows the same results as that without imputation. The reason is that we have simply added more zeros into the data.

Phone: The median change at post-intervention in total money spent over last four weeks is -$800 with values ranged from $0 to -$30000. Non-parametric testing shows a significant reduction (p-value 0.001) in money spent at post-intervention compared with the baseline.

After imputation, the median change at post-intervention in total money spent over last four weeks increases to $0 with the same range of values. Non-parametric testing shows the same results as that without imputation.

Similar statistically significant results from the non-parametric testing were also observed for ‘change in total hours over last four weeks’, ‘change in percentage of money spent over last four weeks’, and ‘change in GABS score’, This conclusion is consistent with that from the repeated measures analysis.

Between groups differences (face-to-face and telephone services) We then compared the difference in change from baseline at post-intervention between the two groups, for the five primary/secondary outcomes. None of the non-parametric tests showed a significant difference between the two groups in change from baseline measures, without or with the imputation. This conclusion is consistent with that from the repeated measures analysis.

162

XGAMBLE Report

3.3.2 Participants’ responses to open-ended questions in Form A and B

This research utilised open-ended questions to help elicit some rich qualitative data that will supplement and further qualify/explain some of the quantitative data.

One qualitative question was posed in the Baseline measurement, its aim was to elicit participant expectations of the effectiveness of the intervention about to be received. Five qualitative questions were asked at the Post Intervention measurement stage. These questions addressed: effectiveness, satisfaction, accessibility, preferred treatment option, and other support received.

Findings relating to each of these questions are detailed in the following section.

Baseline Question - Expected Effectiveness of Intervention All participants, registered and randomised (92), were asked prior to their intervention to rate the ‘expected’ effectiveness of the counselling they were about to receive and to provide explanations for their rating. The following is a summary of the themes for both of the intervention modalities (face-to-face and telephone).

Face-to-Face Interventions (n = 46) Only two participants (both NZ European) in the face-to-face intervention thought that the counselling would not be effective, and based these expectations on previous counselling experiences:

“I’ve had counselling before, so I don’t think it will work” (#10039)

The remaining face-to-face participants expressed positive expectations. These can be grouped into eight general themes and are listed below in order of priority (highest frequency of responses):

1. The majority of the participants expressed that they had the motivation and desire to change now:

“I just feel I’m at the right stage to do something” (#10022)

“Because it is up to the person, you have to be willing to change yourself and I am” (#10106)

“Because I am going to be more truthful this time. I am much more motivated this time as well” (#10088)

2. The next most common theme was hope for a positive outcome to stop/control gambling:

“can help me, I can stop going to the casino” (#10008)

“want to stop gambling” (#10104)

3. Participants then mentioned the need for external support and someone to talk to:

“Just to have someone else to talk to about it” (#10011)

“Because I can’t do it successfully on my own” (#10081)

XGAMBLE Report

4. The next theme expressed from a few participants was having no expectations of the intervention they were about to receive:

“Don’t have an expectation really” (#10024)

“Just don’t know what to expect” (#10037)

5. The fifth theme was general scepticism of the overall effectiveness of the intervention to be provided:

“I don’t know because its different to other things, this is more of an addiction” (#10055)

“I would like it to be effective but I do not know until I start. The last counsellor was not that good last time” (#10077)

6. A few participants mentioned previous counselling that had worked for them so they had positive expectations:

“I’ve done counselling before and I have a good feeling about it” (#10034)

7. A couple of participants mentioned desperation and financial pressure:

“I’m desperate to get out of this situation. I have tried stopping myself from gambling. I lie to get out of problems” (#10041)

“Because I need the money that I am gambling to buy things for the house and everyday things” (#10048)

8. The final theme mentioned was about learning new techniques/skills to bring about change:

“…hopefully I will learn new skills in how to beat this thing” (#10102)

Also, for Maori participants, there seemed to be the desire to stop gambling with an acknowledgement of the difficulty in doing this and a need to get support/help:

“It has been a big step for me to ask for help on my gambling but I feel I need to do this” (#10067)

Telephone Interventions (n = 46) Only one telephone participant (NZ European) did not expect the intervention to be effective at all. They cited similar reasons to the face-to-face participants – a general lack of confidence in counselling due to previous experiences:

“I’ve had counselling for other things before and it didn’t work” (#10098)

On the positive side, there were also participants in this intervention mode who had received previous counselling and were positive and/or hopeful about the future effectiveness:

“Because I have been to counselling before so I know what it is like” (#10047)

“I have found so far that face-to-face with X, I have got a lot from her, this should help further” (#10076)

164

XGAMBLE Report

Overall, there appears to be little difference in the expected effectiveness ratings of telephone counselling compared to face-to-face counselling. However, a greater proportion of participants were sceptical in their explanations, despite still being hopeful:

“just a little bit sceptical of the phone counselling, but I will give it a go” (#10032)

“hoping it will be extremely effective, but I am not about phone intervention” (#10065)

“because it is not face-to-face and body language is excluded” (#10095)

In comparison, two participants commented that they thought the telephone intervention would be more effective:

“To make it effective I will need to keep appointments, it’s easier with phone counselling” (#10056)

“Just the fact you just don’t see the person face-to-face, it can be tempting to lie. In some ways I’ll find it effective” (#10075)

All those participants who expected the intervention to be ineffective did not participate in any intervention. In fact one participant supplied the call centre with an incorrect contact number.

Again, as in the face-to-face mode, the majority of the participants were hopeful and had high expectations relating to current motivation, wanting help to stop/control their gambling, and wanting support/someone to talk to:

“just because I really want to stop, expecting my self to work with it” (#10029)

“Pretty motivated to give up. I have already started myself” (#10043)

“I believe counselling will help” (#10073)

“I recognize I have a problem and I don’t want to spend money on it anymore” (10080)

“Because, when you are talking about it, it seems to help” (#10016)

“because I have been trying myself and with a bit of outside help I can get on top of it” (#10086)

There were also some telephone participants who had no expectations or did not know what to expect:

“Not too sure what the outcome is going to be” (#10010)

“I don’t know what will transpire” (#10096)

Two main themes emerged from the telephone intervention participants. Compared to responses from the face-to-face participants, more telephone participants expressed desperation and the need for skills to control/manage their behaviour:

“Desperate for help” (#10103)

“I wanna quit, I really want to stop” (#10112)

“no more gambling, stop gambling” (#10042)

165

XGAMBLE Report

“I’m about to get a lump sum of money, and I’m scared I’m going to spend it all. I need to do something about it” (#10044)

“Well um your out there to help people to stop gambling aren’t you and I really want to stop gambling” (#10105)

A theme that was only identified in the telephone intervention was the expected professional expertise of the practitioner’s themselves:

“Because they are a professional person. Any advice they give me will help” (#10027)

“I guess the counsellor will be trained to help me” (#10094)

As with the face-to-face interventions, Maori participants emphasised the need for support/someone to talk, actual acknowledgement of their gambling problem, and the need to do something about it:

“It was good to openly talk to someone and even that would be helpful” (#10009)

“In my case I would want someone face-to-face” (#10092)

“I’m motivated to change. Because I do have concerns about my own gambling issues. And I suppose for me I’m serious”

The Asian participants expected to get control of their gambling:

“get control” (#10014)

“no more gambling, stop gambling” (#10042)

“help me cope with gambling addiction” (#10045)

The Pacific participants were more unsure/ cautious in their expectations:

“Not too sure what the outcome is going to be” (#10010)

“hoping it will help” (#10097)

This concludes the qualitative data in the baseline measurement. The next section relates to the qualitative questions that were posed after the intervention was received (i.e. post-intervention).

Post intervention question one - actual effectiveness of intervention Participants who completed their intervention (n=27) were asked to rate the actual effectiveness of it and to provide an explanation for their rating. Responses to this item are summarised according to intervention mode below.

Face-to-Face Interventions (n = 14) Two participants stated that the intervention was not effective:

“No it didn't work for me at all, if I wanted to talk about myself I could have talked to my family it just wasn’t really what I expected... I felt as though I was on a psychiatric chair.” (#10025)

“Not at all. I saw it (intervention) as a compulsory requirement for re-entering the casino” (#10062)

166

XGAMBLE Report

Two other participants rated it only as somewhat effective because it did not stop the gambling completely, or do what they expected:

“It helps me to apply for self-exclusion but unfortunately I still go to gambling venues very often” (#10015)

“but as I have a limited amount to spend anyway I don’t gamble as much as others do, I don’t think the gentleman that came to see me saw that there was no need for me to have counselling he just saw that I needed someone to take control of my money” (#10048)

Two main themes emerged from those participants who rated the interventions as effective to extremely effective. The first related to learning the necessary specific skills/ knowledge to understand and control their gambling, and the second centred on having a good therapeutic relationship (i.e. a practitioner who is understanding, caring and non-judgmental):

“I think it was that they don’t judge and listen and explain where you are on the cycle, counsellor was lovely, having someone there who I respected to talk to, to check in with every week she was wonderful” (#10012)

“just probably talking to someone who knows about what I am going through” (#10026)

“He was patient, kept falling over every week, finally had a break through in the last 4 weeks. Looked forward to going, someone to talk to about it. Non-judgmental approach was fantastic” (#10082)

“He got into the psychology of my gambling - given me great perspective and tools to help me. Understanding the thought process. X was fantastic, I've used the skills he's given me to quit smoking” (#10051)

“We worked on access to money and cutting that off so now there's no way of gambling” (#10115)

The next section is a summary of responses concerning the effectiveness of the telephone intervention.

Telephone Interventions (n = 13) Only one participant rated the telephone intervention as not effective, they attributed this rating to an unsatisfactory connection with the practitioner:

“Not effective at the time, counsellor…(was not available) Only had one session which was face-to-face, meant to be Phone sessions and then never heard from counsellor again” (#10028)

Two participants rated the phone interventions as somewhat effective. One indicated that they would have preferred a face-to-face intervention, and the other would have preferred the counselling to continue for longer:

“When I took up this trial, I was hoping for face-to-face, but I realised that f2f vs phone was random. I remain skeptical about how effective phone can be. Some great discussions and delved into the cause, but f2f has body language and tone” (#10032)

“Wish the counselling to have gone on for longer then perhaps I wouldn't have lost $900 recently” (#10043)

As with the face-to-face clients, participants who experienced effective telephone interventions attributed their success to having the necessary skills/ tools to control their gambling and the qualities of the practitioner/ therapeutic relationship:

167

XGAMBLE Report

“I liked the guy who was phoning me, he has a gentle manner, he is not judgmental. And I could open up to him and be totally honest” (#10046)

“um well it has stopped me from gambling so far I haven’t gambled, it was someone to talk to who didn’t judge” (#10065)

“made myself monitor what I was doing every week” (#10010)

Ethnic Comparisons of Intervention Effectiveness As only one Maori and one Pacific participant fully completed an intervention, and thus responded to the follow up calls for data collection, it is not possible to analyse themes for these two ethnic groups. In comparison, all of the Asian participants completed an intervention and provided follow up data. Two of the Asian participants did not find the face-to-face counselling particularly effective, except as helping to satisfy the conditions for self exclusion and ‘re-entering the casino’ (as noted above). The other three Asian participants thought the interventions (both modalities) were effective in helping them to gain control of their gambling:

“by analysing my gambling from a psychological perspective, helping me find ways to overcome my problems” (#10042)

“It help me to control my mind, encourage me to focus on my business rather than gambling. Because of my job I have to go casino with clients, the counsellor helped me to do my business, help me to avoid gambling with my clients” (#10045)

Post intervention question two - satisfaction with intervention The second post intervention question for participants related to their level of satisfaction with the treatment services provided. This section summarises responses to this item. In general, the responses follow the same pattern as those concerning intervention effectiveness (as detailed above).

Face-to-Face Interventions (n = 14) The two participants that did not see the intervention as effective also were not satisfied for the same reasons (see above) and added:

“I don't think any service can help gamblers, unless they can control themselves” (#10062)

The remaining participants were ‘satisfied’ to ‘extremely satisfied’ with their intervention, mostly because of the way they were treated by the practitioners:

“For the same reasons as above, I knew where I was and she helped me and she knew exactly how I was feeling she also makes you feel very proud of yourself” (#10012)

“Treated like a person, treated with respect, counselling was specific to my needs” (#10051)

“Having someone there who was patient, non-judgmental, checking in. Great having someone to look out for you” (#10082)

Two other responses are also noteworthy in that their experiences were ‘extremely satisfying’; one because it highlights that dealing with other issues in life is important, and, the other is the use of couple counselling:

“I think that it went further than actually looking at the gambling... we looked at the surrounding things going on in my life” (#10063)

168

XGAMBLE Report

“Because we looked at me personally then us as a couple, worked well for both of us, non-judgmental and very relevant” (#10115)

Next is the summary of the telephone intervention satisfaction.

Telephone Interventions (n = 13) Two participants from the telephone intervention group mentioned that they required the intervention to be longer in order for it to be more satisfying. Responses from both of these participants were detailed in the section above (i.e. one did not find the intervention effective as they only had one session and the other participant had wanted the counselling to go on for longer in order to control their spending).

As with the face-to-face participants who were satisfied, the phone intervention participants mostly talked about positive treatment from their practitioners and the gaining of skills:

“Wasn't obtrusive but he was there, and very easy to talk to, appeared to be very intelligent, and uh, very respectful and didn't make me feel like an ass, even though I sometimes make myself feel like one. He was just perfect for the job” (#10030)

“The counsellor had good working attitude, showed her good understanding of my gambling problems, and then provided step by step guide, to help me recognise my own problems and overcome them” (#10042)

Two other responses that made the phone interventions ‘very satisfying’ were; discussing the impact of gambling on others, and, the use of culture/ language specific practitioners:

“it was good to talk to someone about the gambling learning about how I was effecting everyone we also looked at how to cut down until i stopped also how to monitor the days that i wanted to” (#10010)

“Korean counsellor understands what I think, and I could pass my concern to my counsellor without interpreter” (#10045)

Post intervention question three - intervention accessibility This question addressed the accessibility of services and interventions for participants.

Face-to-Face Interventions (n = 14) Only one face-to-face participant indicated problems with the accessibility of the intervention (this person also indicated that the intervention was not effective or satisfying (see above quotes):

“I only attended two sessions of face-to-face counselling. I was busy with study. The counsellor called me on a few occasions, but I felt no hope with this type of psychological treatment” (#10062)

Other face-to-face participants commented that the services and practitioners were conveniently located, accessible at all times, and flexible:

“well because it was easy to get to the building, it was on my terms and my time when I had the time, it was very flexible” (#10012)

“It is easy to contact my counsellor and parking is very convenient for me and counselling time is very flexible” (#10015)

“Because it was set in evenings, so convenient to attend after work” (#10055)

169

XGAMBLE Report

“Contact times, no problem with access” (#10081)

The remaining participants attributed accessibility for face-to-face interventions to the practitioner visiting them at a convenient location:

“He came out to the area where I worked in my lunch time which was quite good” (#10026)

“Because he came around” (#10048)

Next is the summary of the telephone intervention accessibility.

Telephone Interventions (n = 13) Two participants commented that the telephone intervention was not easily accessible; one was a Korean participant who appeared to be hesitant initially:

“I hesitated to call service provider but later I was very happy with the service” (10014)

The other participant attended one session only and could not access the practitioner:

“Yes and no. Just didn't really have enough sessions to know” (#10028)

A similar theme emerged with the telephone participants; the flexibility of the services appeared to be important:

“They called on certain dates that I was free on... it was flexible for me” (#10010)

“Very busy and phone is often pretty hard. Certainly the service made itself flexible to my needs” (#10032)

“Flexible time suits me. After each counselling session, I told the counsellor to contact me at a certain time that suited me” (#10042)

The ease of telephone counselling was also highlighted:

“It was just a matter of phoning, which was a lot easier than having to face up to someone at that time” (#10021)

“I had phone counselling I didn’t have to make a major effort to be somewhere, I just had to be home at a certain time” (#10065)

There were no apparent ethnic differences here as all population groups indicated happiness with the flexibility of the services.

Post intervention question four - preferred intervention option The fourth post intervention question asked participants if the service they were allocated to (and thus used) was their preferred option for seeking help with their gambling problems.

Face-to-Face Interventions (n = 14) Other than participant #10062, who did not think any intervention would help, one participant replied that it was not their preferred option:

“Well I thought that this was the only type I could get” (#10025)

170

XGAMBLE Report

A couple of participants noted that they were either unaware of any other options or did not want other counselling interventions. For example:

“I don’t know of any other options, and I did not want to be involved in group counselling” (#10012)

Other participants commented that seeing someone face-to-face makes you ‘front up’, and was therefore a good option for them:

“Was good having to go in and having to face my problems” (#10055)

“I felt if it was on the phone would have been easy to lie whereas face-to-face I had to be more upfront” (#10115)

The remaining participants preferred the face-to-face intervention because of its convenience and flexibility, and because it worked:

“Because I like he came around personally and on the phone wouldn’t have been any good” (#10048)

“Accommodating around schedules and expectation” (#10081)

“It's the personal approach, I wasn't judged for gambling and the counselling really worked for me” (#10051)

The next section provides a summary of the preferred option status for participants who undertook a telephone intervention.

Telephone Interventions (n = 13) Two participants indicated that phone counselling was not their preferred option:

“Would have preferred face-to-face counselling” (10043)

As with the face-to-face participants, other phone participants attributed the convenience and ease of phone counselling with making it their preferred option:

“It was easier to do it by the phone” (#10010)

“Because I'm about to lose my car so transportation is hard. That’s why phone counselling was good” (#10046)

“It was easier than face-to-face, as I may not of been able to make appointments” (#10065)

“Because it didn't take over my life, and didn't make the rest of my life inconvenient, but it did make me more aware and give me a chance to get it straight in my own head” (#10030)

The only other comments mentioned by both telephone and face-to-face participants were the preference for specific cultural interventions:

“Since I speak Korean, I prefer a Korean counsellor, so I can easily understand what my problem is and how I deal with it” (#10014)

“I preferred Korean counsellor because I believe Korean counsellor can understand my unique circumstance through cultural understanding” (#10045)

171

XGAMBLE Report

Post intervention question five - other support received This is the final post intervention qualitative question. Participants were asked if they obtained any other support while receiving the treatment provided, and if so, from whom.

Face-to-Face Interventions (n = 14) Four face-to-face participants received support from their family, and one also received support from their employer and friends:

“Family support, they were very proud of me” (#10012)

“Got a support person, my sister - she was great, developed a much closer relationship” (#10055)

“Colleagues, friends and family also employer - some of it was actually helping me with my finances also helping me to offload what I was feeling also having supportive friends” (#10063)

“Let family members know of my problem to make sure I couldn't access any cash from them” (#10115)

The remaining ten face-to-face participants said they did not receive any other support; however, one participant did provide the following comment:

“Wouldn’t invite me to their pokie sessions but my employer was very helpful letting me go to counselling sessions” (10081)

Telephone Interventions (n = 13) Three telephone participants received further support from either seeing another practitioner, an internet website or family and friends:

“Was seeing another counsellor, seeing them with other issues” (#10016)

“Chat site with other gamblers wanting to quit” (#10043)

“Talking about knowing that when I wanted to talk about it I could ring a friend and know they would be around straight away” (#10047)

It is interesting to note that only NZ European participants cited receiving additional support.

This concludes the qualitative data section of the Phase Two pilot/vanguard trial.

172

XGAMBLE Report

3.4 Discussion

3.4.1 Intervention effectiveness and measures

Following the structure of the results chapter, this subsection of the discussion chapter is divided into four parts. The first part discusses the significant aspects of participants’ demographic features gathered in this study in relation to literature. The second part focuses on the five primary and secondary outcome measures used in the present study with some reflection on the usefulness and appropriateness of those measurements. The third part of the discussion chapter examines the pattern of changes in outcome measure scores between the baseline and post-intervention visits and how the findings of the present study are compared to closely related published studies.

Demographic characteristics of the research sample Age The mean age of all study participants was 43 years old and only one person (or 1%) recruited in this study was under the age of 20. This corresponds to the national service statistics, which reported that 2% of new clients presenting to either face-to-face intervention services or Gambling Helpline Ltd were younger than twenty-years old (Ministry of Health, 2006, see pp 33-34). It highlights the ongoing challenge of how problem gambling service providers can promote the availability of services for young people and remove barriers for youth in accessing the services.

DSM-IV scores In this study, the mean DSM-IV score of all study participants was 7.4 (SD 1.89), which seems to be consistent with other similar community-based studies (e.g. Crisp, Thomas, Jackson, Smith, Borrell, Ho, Holt & Thomason, 2004; Hodgins & el-Guebaly, 2004). In the present study, the relatively higher percentage of non-European participants (e.g. Maori [65%], Asian [100%]) scoring 8 or above could be explained by at least two reasons. Firstly, given the DSM-IV was originally developed with a sample strongly grounded in white European culture, it might not be culturally sensitive or appropriate in diagnosing problem or pathological gambling results (or over-diagnoses) among members of other ethnic populations. Secondly, the higher DSM-IV scores, or more serious level of gambling symptoms, may merely reflect the fact that indigenous or migrant people tend to seek help later or wait until mental health problems become completely out of control.

Gender Sixty-seven percent of the present sample were women. In the United States, the percentage of women who have ever gambled rose between 1975 and 1998 by 22%; from 60% to 82%. Recent New Zealand service users statistics report that more women (n=593) than men (n=556) stated non-casino gambling machines as their primary mode of gambling (Ministry of Health, 2006, see p. 42). One possible explanation for this gender difference is that it may be fewer barriers for female problem gamblers in admitting their difficulties and seeking assistance to overcome their problems. In the context of the present study, this trend may be attributed to one of the key eligibility criteria. In relation to their gambling behaviour, all prospective participants’ had to be primarily concerned about pokie machine gambling (electronic

XGAMBLE Report

gambling machines). Electronic gambling machine venues were described by Legge (1996, p. 103) as “more sanitised, even family friendly, entertainment-style venues” which has greater appeal to women. It is suggested in the literature (e.g. Crisp & associates, 2004; Korn & Shaffer, 2004) that female gamblers favour games such as slot machines and bingo that are not skilled based; also they prefer to gamble in casino and bingo halls that are perceived to be safe. Compared to men, women tend to adopt a so-called ‘escape gambling mode’ that is, women gamble more to reduce boredom, escape from responsibility and relieve loneliness, compared with men who gamble for excitement, money or pleasure (Korn & Shaffer, 2004). Crisp and her colleagues (2004) assert that practitioners and researchers need to learn more about women’s experiences of gambling, problem gambling, and specific needs for problem gambling counselling. The specific needs documented in Crisp and associates (2004), and Korn and Shaffer (2004) include: 1) women prefer community-based to residential treatment as they do not need to worry about care for dependent children while they are receiving professional help; 2) women want counselling services to be housed in existing agencies with which women may already have a relationship e.g. social services, primary healthcare facilities; 3) women may prefer group formats to individual sessions for counselling and support; 4) women prefer treatment settings in which females are not the minority group; and, 5) women who are concerned about being stigmatised for their gambling problems prefer agencies that provide discreet venues for counselling.

Ethnicity One of the strengths of this study was that up to 45% of the participants were non-Europeans; in particular, Maori, Pacific Peoples and Asians. This contrasts with the majority of problem gambling studies conducted in English speaking countries; generally, between 2% to 25% of participants are indigenous people or born overseas (Petry, 2006). Unfortunately, given that the total sample size of the present study was small, the analysis of ethnic responses could only suggest a trend instead of drawing any conclusive results of the investigation.

Education level and employment status In terms of the education level and occupational status of participants in the present study, it simply confirms that problem gambling affects individuals from every walk of life – both the well educated and those who self-reported not having any education.

“Myth: Gambling does not affect ‘nice people’, ‘religious people’ ‘educated people’…it is basically a problem that weak people acquire through selfish behaviour”

“Fact: Problem and pathological gambling can affect anyone who wagers as well as the individual’s family, friends, and members of social institutions who do not wager” (adapted from McCown & Howatt, 2008, p. 10, italic added).

Marital status Noticeably, up to 68% of the present sample described themselves as not married or in a partner relationship. Members of the Project Team suggested that stable and committed relationships could have three potential positive impacts on someone affected by problem gambling. Firstly, it is often the spouse or family members of the gambler who first seek help from professional organisations for problem gambling. In New Zealand approximately 30% of clients presenting to either face-to-face or Gambling Helpline Ltd services were partners or significant others of gamblers rather

174

XGAMBLE Report

than gamblers themselves (Ministry of Health, 2006). Secondly, having a family or a stable relationship may prompt individuals to go forward to seek help for their problem gambling and family members will also often provide support. Hodgins and his colleagues (2002) recruited 101 research participants who had recently quit gambling to investigate reasons for stopping gambling. One of the major findings was that participants in Hodgin’s study sought help because they had concerns about family/children. Long-resolved gamblers were likely to endorse “social support” as the reason for stopping gambling. Hodgins et al. (2002) also divided the reasons to cease gambling into internal and external reasons (see Table 46). Similarly, a smoking cessation study found that intrinsic motivations to quit are more predictive of successful cessation than external motivations.

Table 46: Why people seek help for problem gambling (Hodgins et al., 2002)

Five most frequently endorsed reasons for stopping gambling

Reason %

endorsed reason

Intrinsic or extrinsic motivation

financial concerns 96 intrinsic emotional factors 92 intrinsic concern about family/ children

69 intrinsic

hit ‘rock bottom’ 61 intrinsic evaluated pros & cons 54 intrinsic

Thirdly, the level of social support that individuals can receive helps sustain the changes in their gambling behaviour in the medium or long term. However this study had a very high drop out rate in both the post-intervention and follow up visits that did not allow the researchers to gauge the relationship between social support and the longer term effects of the intervention provided.

First time heard about the study In the present study, the majority of participants recruited through the ‘personal relationships’ category (i.e. combining the categories of ‘word of mouth’, ‘friends/ relatives’, ‘direct contact with researcher’) were non-NZ Europeans and women. Recruitment was a major challenge in this research. Based on the United States experience, it is estimated that approximately only 8% of individuals with problem gambling receive counselling for gambling and almost no at-risk individuals seek treatment (National Research Council, 1999). Over 80% of the participants in the study by Hodgin and el-Guebaly (2000) cited that they did not seek treatment as they wished or believed they could handle the problem on their own. The next five commonly endorsed reasons for not seeking professional assistance for problem gambling were: 1) embarrassment or pride, 2) not considering their gambling to be serious enough to necessitate treatment, 3) being unaware of treatment available, 4) inability to share their problems with others, and, 5) concern about stigma associated with problem gambling. Based on the observed data from this study, for native people, migrants and women, having a personal relationship with someone who knows about a problem gambling service substantially removes most of the anxieties and worries surrounding the actions to seek help for gambling problems.

175

XGAMBLE Report

Comorbidity One strength of the present study is that the research team did not use psychiatric comorbidity to screen out potential participants, thus creating a research sample with individuals receiving treatment for mental health (21%), alcohol and drug substance abuse (6.3%) and having a history of head injury (2%). These profiles of comorbidity are reflective of routine clinical practice and thus increase the generalisation of the findings.

Prior problem gambling counselling experience There were a relatively high percentage of Pacific participants who did not have any prior experiences of receiving problem gambling counselling. The post-study Fono with Pacific practitioners in Auckland and service providers suggested this could be due to four major reasons: 1) normalisation of gambling in Pacific culture and everyday activities, 2) lack of understanding about counselling and the potential benefits, 3) shame and stigma associated with problem gambling, and 4) concerns about confidentiality in some small close Pacific communities.

Annual household income Lastly, the household income results support concern about the gambling related harms caused in low socio economic samples. There were few people in the Maori, Pacific and Asian groups earning more than $70,000 in annual household income. Up to 57% of participants in this study had an annual household income of less than $40,000. This coupled with the facts that the majority of the participants were women (possibly with dependent children), scoring in the pathological gambling range, and not in stable relationships, lends itself to great concern.

The primary and secondary outcome measures The following discussion focuses on the primary and secondary outcome measures, the usefulness and appropriateness of those instruments used in this study and to what extent the present study has added to the present body of knowledge in the field. Researchers need to use reliable and valid instruments that can gauge changes in gambling behaviours and other gambling-related problems within a given time or over time.

The selection of the primary and secondary outcome measures employed in this study were guided by two major considerations. Firstly, money wagered in gambling and time spent gambling are commonly used measures in research on problem gambling interventions. Secondly the use of GABS and GRTC in this study was based on the understanding that a common element of interventions provided in this trial is related to improving clients’ understanding about their own gambling (e.g. education on problem gambling, faulty cognitions associated with problem gambling) and changing the clients’ gambling behaviours (e.g. action plan to reduce or stop gambling, manage risky situations, prevent relapse). Although no specific intervention strategies or models were isolated or tested for application intensity or specificity, review of 39 out of 92 (42.4%) practitioners’ notes confirmed the main interventions employed by practitioners in this project were around changing behaviours, increasing the level of knowledge about gambling and problem gambling and shifting from maladaptive to adaptive cognitions. Along a similar vein, it is hypothesised that “cognitive changes predict success in reducing gambling behaviours” (Breen, Kruedelbach & Walker, 2001, p. 247; for details of a recent study on maladaptive thinking and coping, see Parke, Griffiths & Parke, 2007).

176

XGAMBLE Report

This research also followed some of the best practices in problem gambling intervention studies. Participants were asked to report the specific frequency (number of times per month) and intensity (amount waged per month) and gambling behaviours (e.g. diagnostic symptoms, motivation to change, substance abuse, and concomitant psychopathology) (Lopez Viets, & Miller, 1997). Undeniably there are still potential problems about the way researchers’ measure someone’s spending on gambling.

“In terms of intensity or amounts spent gambling wealthy individuals may bet hundreds of dollars a month without experiencing gambling problems, whereas people with lower income may suffer difficulties with substantially lower overall rates of wagering. No assessment instruments available at present adequately take such differences into account in assessing gambling and gambling problems” (Petry, 2006, p.280).

The issue is further complicated because of the many different forms and different levels of availability of gambling opportunities that exist. This is why only prospective participants whose concerned gambling activity was pokie machines were included in the present study, whereas most other studies did not specify or restrict the mode of gambling activity.

Also, as with all self-report measures, there is the risk of under-reporting problem behaviours. We attempted to address this by emphasising to participants that all responses were completely anonymous and that no one would ever know how they personally responded to any of the measures used in this study. However there may also be concerns about the accuracy of participants’ recall of their gambling behaviours. The present study asked participants to recall in average how many sessions they spent gambling over the last four weeks and how much money they spent on gambling in each session, rather than simply asking participants how much money they spent in total. The rationale is: asking participants to retrieve specific details surroundings the events (e.g. how many sessions, how much spent per session) may help improve the accuracy of the memories. Also it was found that gamblers’ self-reported frequency of gambling and amounts wagered yielded an acceptable level of reliability and validity (Hodgins & Makarchuk, 2003).

However, there is an additional measure or technique proposed in the recent literature to further improve the accuracy of participants’ recall of events or gambling behaviours (Petry, Litt, Kadden & Ledgerwood, 2007; published after the design phase of the Phase Two study therefore the idea was not incorporated in the present study). The Timeline Follow-Back (TLFB) uses calendar prompts to elicit frequency and intensity of past behaviours (i.e. number of days gambled and amount of money wagered daily) for the past 30 days at each visit or assessment. The gambling self-reports are then verified by collaterals nominated by participants - collaterals are asked over the phone “how often did (participant) gamble on average in the past month?” and “On days when (participant) gambled, how much money do you think s/he spent on average” (Petry et al., 2007, p.1282).

It is generally impossible to establish the real impact of the money lost on an individual’s financial situation. This study asked participants to report their total household income (before tax) so that we could calculate the individual’s proportion or percentage of money spent gambling. It is worth mentioning that this remains a very conservative estimate of the impact of money lost on individuals who gamble

177

XGAMBLE Report

and their families, given that a significant portion of the income goes to expenses such as tax or accommodation (either rental or mortgage, food and other basic necessities). Neighbors, Lostutter, Larimer and Takushi (2002) went a few steps further in their study. They asked research participants to report how much money they lost and won and their disposable income in a given time. These additional data remove the ambiguity of whether the participants’ expenditure includes only money lost that was intended to be spent or whether money won that was subsequently lost was counted as expenditure. Also arguably for example, a $75,000 annual household income provides a comfortable lifestyle for a single individual, will be inadequate to support a family of seven (e.g. a couple, three dependent children and two ageing parents), and that disposable income is the more relevant figure to measure the impact of money lost in gambling. The challenge of this is it will take precious data collection time to explain to participants what ‘disposable income’ means.

Furthermore, Petry et al. (2007) cautioned that the above three indices (in the present study: frequency - days gambled; intensity - dollars gambled or lost and related subjective problems - DSM-IV) may assess somewhat different aspects of gambling. They also argued that the days or time wagered may be the least sensitive index to change, however, the present findings did not seem to be consistent with this postulation.

The other two secondary measures namely GABS and GRTC are discussed in the next section, as the focus is on improvement in scores between visits.

Improvements in overall scores of primary and secondary measures between ‘Face-to-Face’ and ‘Phone’ based interventions The present study is exploratory by nature. Therefore it is irresponsible or even dangerous to make any changes in policy or funding decisions based on this present study. Based on this small sample study, there were no statistically significant differences in terms of the improvements in overall scores of measures between ‘Face-to-Face’ and ‘Phone’ based interventions.

However there was a significant overall PERIOD effect for both ‘Face-to-Face’ and ‘Phone’ based interventions between visits for both primary and secondary outcomes. The observed primary data had a significant reduction in the amount of dollars spent and time spent gambling between baseline and post intervention. The PERIOD effect between baseline and post-intervention visits was found for all outcome measures except Gambling Readiness to Change scale (GRTC). However, without a control group (and with the small sample size), it is not possible to attribute the change to positive intervention effects – a full scale trial with control groups would be required to infer that.

Gambling Attitudes and Beliefs (GABS) scores were found to be a useful measure in this study. There were moderate (r= .26, p<.01) to low (r= .19, p<.1) levels of positive correlations between GABS and self-reported money and time spent in gambling (see Table 40). It yields partial evidence to support the hypothesis of Breen et al. (2001) that changes in cognition and attitude towards gambling are associated with changes in gambling behaviours. One example in changing cognition to change gambling behaviour is the work by Coman, Evans and Burrows (2005); they used a step-by-step approach to use cognitive strategies to assist individuals reducing gambling related harms. The steps involve the use of role plays, games, white board, and drawing or diagrams (e.g. ‘factory roof line graph’) to illustrate how gambling has negatively

178

XGAMBLE Report

impacted on one’s financial wealth over time and the course of remedial actions that should be taken to reduce the damage.

The present findings have shown that there were not any significant relationships between GRTC and the above mentioned indices of measures on money and time spent on gambling. One possible explanation is that the GRTC scale is only appropriate for measuring readiness to change among individuals who gamble at least moderately, and thus might have reasons to change their gambling behaviours (Neighbors et al., 2002). However participants in this study would have very different concept of what counts ‘gambling at moderate level’ before they consider changing their gambling behaviours.

The qualitative data obtained from both clients and practitioners helped to provide some understanding of those factors contributing to positive changes between the base-line and post-intervention visits. Details are available in the section headed ‘Descriptive summary of practitioners’ and the entire Subsection 3.3.2 ‘Participants’ responses to opened questions in Form A and B’. The diagrammatic representation (see Figure 5) of key factors that contribute to the outcomes of problem gambling interventions also provides further insight of what might constitute effective interventions for problem gambling. To sum up, based on the obtained qualitative data, the main contributing factors to successful outcomes for individuals with problem gambling are: 1) clients’ readiness to change; 2) information, knowledge and skills imparted to the client; 3) the genuine warmth and acceptance communicated from practitioners to clients; and, 4) the level of social support available for clients to make and sustain the changes in their gambling behaviours and attitudes.

Griffins and Cooper (2003) compared the two modes of service delivery and concluded that telephone-based counselling for people with problem gambling has the following advantages: • It is convenient; • Cost and time-effective for clients; • Overcomes barriers that otherwise may prevent people from seeking face-to-

face help; • Helps overcome social stigma associated with problem gambling; and, • Reaches people nationwide, in rural areas and those without transport.

Meanwhile, they identify the advantages of face-to-face counselling in supporting individuals with problem gambling as the following scenarios: • The nature of the problem can be very complex and have occurred for

prolonged periods of time thus being difficult to manage over the telephone only;

• An individual suffers from severe or chronic problems (e.g. depression, trauma) body language and facial cues can be crucial for the practitioner to function effectively;

• An individual may not like talking to someone or sharing personal problems over the telephone without establishing trust in a personal face-to-face relationship first;

• An individual who wants to show his/her commitment to change by taking the effort to attend face-to-face counselling sessions; and,

• An individual may prefer having the privacy to talk openly with practitioners.

179

XGAMBLE Report

See Subsection 3.3.2 for qualitative results on the strengths and limitations of face-to-face and phone interventions in this present research as perceived by participants.

Regardless of the mode of service delivery, McCown and Howatt (2007) in their latest book on problem gambling described seven principles of effective interventions for problem gambling (pp.144-145): • No single intervention is appropriate for all individuals; • Intervention needs to be readily available; • Effective intervention attends to the multiple needs of the individual, not just

his or her gambling problems; • An individual’s intervention and service plan must be assessed continually and

modified as necessary to ensure that the plan meets the person’s changing needs;

• Remaining in interventions for an adequate period of time is critical for intervention effectiveness (particularly for medium and long-term recovery);

• Counselling (individual and/or group) and other behavioural changes are critical components of effective intervention for problem gambling; and,

• Intervention does not need to be voluntary to be effective.

3.4.2 Effectiveness of recruitment strategies Increasingly low rates of service uptake (as mentioned previously, it is estimated that approximately only 8% of individuals with problem gambling receive counselling for gambling and almost no at-risk individuals seek treatment (National Research Council, 1999) and the decreasing number of clients entering problem gambling intervention services, resulted in greater than expected recruitment challenges for the Phase Two pilot/vanguard clinical trial. However, the recruitment/retention rates observed in this research are consistent with New Zealand’s current problem gambling service statistics in terms of engaging/maintaining clients and the gender and population group distributions.

Print Media The most cost effective advertising strategy utilised in this research were major articles (with stories and photos) in prominent positions in the NZ Herald (resulting in the recruitment of 17 people in three days) and local community newspapers. In each study location, the Project Team negotiated free editorials/stories in the free local newspapers along with paid smaller ‘classified’ advertisements. Recruitment through this method was found to be slow but consistent, unless other major advertising relating to problem gambling and help-seeking were also running. It appeared that the Health Sponsorship Council’s problem gambling social marketing campaign, which ran from the end of March through April 2007, hindered rather than helped this study: the number advertised on TV and radio directed a large number of new ‘clients’ to the Gambling Helpline. This was unfortunate timing for this piece of research.

Television The Project Team attempted, but had no success, with negotiating television airtime. The main current event programmes were unwilling to schedule any more gambling stories despite being approached at various times. The only programme that did agree to air an item would not allow client anonymity; as clients did not feel safe to go on air under these circumstances, an item could not be produced. Note that local newspapers did allow client anonymity for their articles.

180

XGAMBLE Report

Radio The Project Team investigated the utilisation of advertising the study via the radio. In addition to the high cost, the team were discouraged from this avenue as service providers advised that radio advertising had not resulted in clients utilising their services. However, it was found that free community notice board advertising on radios (especially 1ZB Sports Programme) did generate interest and participants during the small period it was available. Furthermore, the Project Team purchased some radio interview time with NiuFM over March and April 2007 in an attempt to increase the low recruitment rate in Pacific populations. While this was not successful for the study (most likely because of the limited amount of air-time at specific times, and only specific population groups that listen to these talk back sessions), it did generate discussion and may have been effective given more time and resources. The Project Team also secured some time on Asian radio and was interviewed about the research and Asian gambling. The Project Team were unable to gain time on Maori TV or radio during the recruitment period, despite numerous attempts at contact by various people.

Community Based Advertising and Recruitment Other general recruitment strategies included advertising on websites, billboards, newsletters, through posters and flyers in various organisations and locations and having community recruiters out in the community promoting the research through flyers and word of mouth. Overall, this took considerable time and resources and did not gain many participants. However, it seemed to help in promoting general awareness of problem gambling in communities (as reported by the recruiters) and may have contributed to the process of de-stigmatisation and education of the public about the availability of services for problem gambling. Feedback from the community recruiters also indicates a very large presence of gambling harm in communities that is not currently being addressed. For example, many shopkeepers requested flyers as they knew of many local people with gambling problems that were not accessing any help.

Further to the radio and TV advertising aimed at ethnic populations, Pacific, Maori & Asian community recruiters were present at local ethnic festivals/forums (to distribute flyers) and in their specific local ethnic communities (including hui, community, church and sports meetings, ethnic shops, health and education forums etc). Again, this strategy consumed much time and resources with little results. Earlier on in the recruitment process, the research was advertised in various ethnic newspapers and websites. Apart from the Asian population which had a small response (particularly via the websites), very few participants were recruited via this method. The PGF Asian team was also very supportive of our Asian recruiter and often allowed her to promote the project at their various meetings, events and networks.

Despite a number of ongoing relationships and networks between the Project Team and Maori researchers and health providers, the Project Team has found it difficult to recruit and retain Maori for this research. The Project Team believe that one of the challenges facing the problem gambling community of New Zealand is the strengthening of relationships with Maori.

Probation Services The Project Team was successful in collaborating with probation services both nationally and regionally. After ethical permission was gained nationally, recruiters gave presentations at local probation services with positive responses and probation officers indicated they have clients that they could refer through to the project.

181

XGAMBLE Report

Although no specific referrals were received before recruitment closed it was thought that this was an important collaboration to foster. The Citizens Advice Bureaus (CABs) and Budgeting Services were also supportive of this research and took flyers and posters for their offices. The recruiters did have some difficulty with some local GP’s as they did not consider gambling a significant health issue.

Members of Gambling Industries While various members of the gambling industries were approached for their assistance with this research, most were unsupportive. ClubsNZ were the only gambling operator that supported the research and helped with recruitment by displaying advertising posters in their clubs. The research team spent several months pursuing a collaborative recruitment approach with a major gambling venue (i.e. the inclusion of recruitment material in host responsibility/self-exclusion packs). However, this relationship was terminated by the person in-charge of the discussion, following a statement in the media by a senior member of the research team on a topic unrelated to the research project. Another member who represented a major player in the gambling sector refused any support to this study from the outset, and attributed their decision to objections with the utilised methodology, particularly the recruitment of participants from the general population.

Summary of Recruitment Issues This project encountered significant difficulties in recruiting participants. The reasons for these difficulties as summarised below: • The period of time required to recruit participants for intensive research such

as this (e.g. multiple points of data collection and participation in therapy), is greater than in less intensive research (e.g. a single data collection point or interview appointment). Issues such as the building of trust take extended time (especially if no koha/incentives are provided).

• Limited support from gambling operators with recruitment options. • Closure of intervention services over the Christmas holiday period. • The prohibitive cost of major advertising (large articles in NZ Herald and

major papers, TV and Radio advertising) over an extended period of time for recruitment of a sample from the general population; the Project Team was required to reallocate additional funding for advertising.

• The existing barriers for this population group in relation to accessing treatment services. Six common barriers include: 1) a desire to handle the problem on their own; 2) embarrassment/pride; 3) the gambling behaviour not being considered problematic enough to necessitate treatment; 4) lack of awareness of available services; 5) an inability to share their problems with others; and 6) concern about stigma (Hodgins & el-Guebaly, 2000).

• Language and cultural barriers to research and services. • Fluctuations in participant motivation/commitment to acquiring and

maintaining treatment. • The required ongoing support and maintenance of participants, by recruiters

and practitioners (where the project has very limited control). • Limited service access for participants, particularly in relation to after-hours

service. • Difficulty in recruiting ethnic populations, despite strategies involving ethnic

researchers, ethnic specific newspapers, websites and radio programmes. • The encountering of technical difficulties with the website interface (the

interface that was utilised to register participants and complete baseline data collection).

182

XGAMBLE Report

• The co-occurrence of a social marketing campaign, which resulted in the direction of potential clients to alternative services, thus resulting in the depletion of an already limited client pool.

• The high level of energy and motivation required by practitioners to continue engaging with this group of research participants and the Project Team (e.g. being required to notify the Project Team of termination in timely manner).

In summary, the majority of participants (including ethnic population groups) were recruited via generic newspaper articles and advertising. Ethnic specific recruitment strategies did not result in participants per se but did generate interest and result in increased levels of general awareness. Increased numbers of participants (i.e. more clients seeking help) may result if future recruitment strategies: • were allocated more time and resources; • the level of support received from government, industry, social/community

agencies and ethnic communities was increased; and, • incorporated comprehensive consultation and relationship building.

The above challenges were noted throughout the recruitment period and several efforts were made to rectify these issues where possible (see recruitment reports in Appendix). Many of the issues remain significant and will continue to be a challenge for future research in this area.

3.4.3 Effectiveness of follow up methods Following initial recruitment, the monitoring of client/practitioner progress and ensuring the collection of data at the appropriate periods was a challenging and time intensive activity. The project manager maintained constant contact with the call centre and practitioners/services, and monitored progress via spreadsheets. This ensured a smooth and timely interface between clients, practitioners and the call centre.

The practitioners found it difficult to make initial and ongoing contact with many clients following their registration and referral. It is a time-consuming exercise to build and maintain relationships with this vulnerable and changeable client group. The services’ ‘rule of thumb’ (and the research protocol) was to attempt contact with a client/participant on three occasions, over different times of the day (mostly during working hours). If this was not successful the client/participant was then passed back to the Project Team, whereby the call centre attempted to re-establish contact and see if the client/participant still wanted to participate in the research. If contact was made and the client/participant was still willing to participate, the call centre facilitated the client-practitioner link once more. The call centre had some success with 3 clients re-engaging in this additional process, despite the practitioners being hesitant about this approach.

Towards the end of the recruitment/intervention period, the call centre also began to encounter difficulties in contacting and following up with participants (for data collection), despite the fact that the call centre could make numerous attempts at different times of the day (operational evenings and weekends). This process was complicated by three factors: 1) for follow up calls, we were also trying to reach people who did not participate in the intervention in the first place; 2) there were time delays in contacting the participants for data collection as the Project Team were not always notified in a timely manner of intervention termination; and, 3) some contact

183

XGAMBLE Report

details for clients, primarily mobile phones, were no longer valid. All of these factors contributed to the low retention rate of this research. However, analysis of service user statistics and feedback from practitioners and services indicate that this is consistent with normal practice. Future research in this area will need to work intensely at, and devote considerable resources, to client engagement/retention.

3.4.4 Learning from participants’ experiences in participating in the trial Qualitative data was collected at baseline and post intervention data collection points in Phase Two. This enables the gathering of some explanations/views from the participants at the time of their intervention (opposed to retrospective data which is normally obtained) and can be compared with the information gained via Phase One interviews.

It was interesting to note that those few participants who did not anticipate the intervention as being effective did not begin the intervention at all. They all expressed a lack of confidence in counselling, due to negative past experiences of it. However, the opposite of this was also true for some participants; due to previous positive experiences with counselling, they believed the current intervention would be effective. A small number of participants did not know what to expect and some expressed desperation for help, mainly due to financial pressures or being tired of their dishonest behaviours, for example, lying to family.

There were a wide variety of reasons why participants expected their intervention (both face-to-face and telephone) to have positive outcomes. The most common themes being: having motivation and the timing being right, hope and desire to stop or control their gambling, the need for external support and someone to talk to, and wanting to gain the knowledge and skills to change their behaviours. These themes were also identified in the Phase One service user interviews (when participants were asked to identify what contributes to a successful intervention).

Following the completion of interventions, a focusing of the themes was evident. Most participants rated the intervention received as effective due to: 1) gaining the required knowledge and skills to understand their problem gambling behaviour and be able to control it; and, 2) having a good therapeutic relationship i.e. being listened to, respected and not judged, and the practitioner being patient.

Only three participants thought the interventions were not effective. The following reasons were cited: they did not enjoy the counselling experience, they did not receive the service they wanted, treatment was seen as a compulsory requirement for re-entering the casino. This pattern was repeated for the question relating to satisfaction ratings (below).

When participants were asked how satisfied they were with the interventions they had received, there was again a clear theme: satisfaction was due to the patient, respectful and non-judgmental ways in which practitioners treated their clients. A couple of other themes contributed to interventions being ‘extremely satisfying’: dealing with other life issues, use of other counselling methods (e.g. couple counselling), the use of cultural/language specific interventions and discussing the impact of gambling on others. Again, all these themes were highlighted in the Phase One interviews.

184

XGAMBLE Report

Overall, participants who completed their interventions thought that services (both the face-to-face and telephone) were conveniently located, accessible at all times and flexible – especially when the practitioners came to participants and were flexible with appointment times. The ease of telephone counselling was highlighted.

As far as preferred option of help seeking is concerned, participants enjoyed the convenience and flexibility of both modalities of counselling. Face-to-face participants highlighted the importance of having to ‘front up’ to counselling while the telephone participants enjoyed the ease of only having to make a phone call. Being able to choose cultural/language services was also mentioned here.

Relatively few participants accessed other support while receiving their intervention, and this most commonly came from family members and friends, while one participant was seeing another practitioner (for other issues), and another was utilising internet chat sites for gamblers.

In summary, there was little difference in the qualitative feedback between face-to-face and telephone counselling, and the small number of ethnic participants that completed the interventions limited the interpretation of those results. Overall, general themes were evident which were similar to the Phase One results/discussion. These themes centred on the quality of the client-practitioner relationship.

3.4.5 Reflections on working with multiple stakeholders This research project was undertaken over an 18 month period and involved working with many key stakeholders. The following are the Project Teams brief reflections on some of the main working relationships, including learnings from this process and the contextualisation of findings.

Problem Gambling Service Providers Representatives from all of the service providers involved in Phase Two of this research were included in the Technical Advisory Group to provide their specific knowledge/expertise to the project and to help promote and support it. This was very useful to the Project Team and the TAG provided invaluable input and helped contextualise many of the results.

The Project Team consulted with each service provider prior to the commencement of this phase and provided a full days training for each of the 20 practitioners who were involved in delivering interventions. The project manager also visited each individual service after the beginning of the trial to discuss the project and try to enlist recruiters in each location. Once recruitment and the interventions where under way, the project manager maintained email contact with all the services in order to monitor progress, update services about upcoming advertising in their local regions and to respond to any queries/comments received. The Project Team was able to visit some of the services during the trial. This was appreciated and provided a forum for the services and practitioners to provide further feedback and receive updates while the trial was still in progress. After the interventions where completed the Project Team endeavoured to personally thank services and practitioners and provide an opportunity for feedback and reflection.

185

XGAMBLE Report

Communication and referrals via email seemed to work well for all parties involved as it was quick, easy and cost effective. Feedback from service providers indicates the importance of active and open two-way communication channels.

When local managers of services were directly involved with the research, the working relationship between practitioners and the Project Team appeared to be particularly effective. For example, one service manager developed an evaluation/ feedback template based on the research feedback requirements that significantly reduced the amount of time and paperwork required from practitioners. It is advisable that future research involves and keeps all local area mangers up to date throughout the research period.

As service providers have stated in various meetings and correspondence, it is imperative that they are involved from the beginning of any research project. This enables them to provide their expertise and ensure that timing and resource allocation is available and accessible. The Project Team believes that the support and ‘buy-in’ of local management in services is important. For future research in this area, it will be imperative to consult with service providers as early as possible to ensure there are no clashes with resource availability (e.g. practitioners and clients). Unfortunately there are still times when unforeseen circumstances will cause disruption between service provision and research requirements.

Practitioners At the beginning of the research, services were asked to provide 20 practitioners that would be available to conduct interventions during the clinical trial period. The Project Team indicated that it required a range of experienced practitioners, covering the four locations and ethnic populations, which were able to deliver interventions via face-to-face and phone settings. The clients referred to services by the research team were to be part of each services Ministry of Health quota, and therefore treated in the same way. Any additional time given to research procedures would be paid for from the research projects budget.

Of the practitioners secured at the beginning of the research, three were eventually lost or out of action for sometime, as they either left their employment or personal circumstances hindered their continued participation. This resulted in extra workloads for those practitioners remaining in the project. Moreover, at least four of the practitioners were part-time staff who found the extra effort of contacting and maintaining the research participants a strain on their already tight schedules.

The longer than anticipated timeframe of the recruitment and intervention period of this research, combined with other competing research projects and the HSC social marketing campaign appeared to decrease the motivation, time, and energy of some of the practitioners. It was noted that the engagement of clients dropped off significantly during the final two months of recruitment, and the numbers of participants completing interventions also reduced significantly. Some practitioners found the engagement and maintenance of some research participants frustrating and did terminate their participation with services, noting that the participants were not motivated to undertake the intervention.

Significant feedback was obtained from practitioners in relation to their perspectives of the effectiveness of telephone and face-to-face counselling. Some felt that telephone counselling was not as effective as face-to-face counselling due to: the difficulty of obtaining informed consent and engaging and maintaining clients, a

186

XGAMBLE Report

perceived lack of motivation from telephone participants, an inability to use other non verbal cues in counselling, and participant distractions while on the phone (note that this perspective comes from face-to-face practitioners who may lack experience with phone based interventions). This is somewhat contradictory to the perspectives given by telephone participants’ who were satisfied with the interventions they received.

Taking the complexity of the project and participant group into account, the work between the services/practitioners and Project Team proceeded relatively well. However, improvements can always be made. In particular, this would include involving service providers earlier in the research planning, getting local management buy-in, and finding ways to reduce the time and energy input required from practitioners (and participants).

Specific ethnic populations It is acknowledged that optimal research outcomes would result from Maori and other ethnic specific groups undertaking their own research. In the context of pilot/vanguard research, that is not initially a feasible option.

Within a limited timeframe and budget the Project Team incorporated ethnic specific researchers, services/practitioners, and community people/organisations. This was achieved at multiple levels, including: the technical advisory group, as part of the Project Team, in the Call Centre, and as service providers of interventions (to obtain some cultural representation and matching). Although best efforts were made by the research team, it is acknowledged that further ethnic representation is required for future clinical research.

As well as the ethnic data from Phase One interviews and the descriptive results from the Phase Two pilot/vanguard trial, feedback from ethnic focus groups was obtained to verify our results and ensure that they were given appropriate contextualisation.

The general and Maori specific findings were presented to a Te Herenga Waka O Te Ora Whanau (National Maori Gambling Reference Group) meeting and received positive feedback. It was noted that Maori represent a significant proportion of the new problem gambling service user population (approximately 30%) and therefore will need to be over-sampled in any future clinical research in order to reach that target representation. Maori in this study had a low recruitment and retention rate. It is thought there are many reasons for this and the feedback received highlighted that Maori often present in crisis and/or are highly motivated by other circumstances and are not generally ready to change. It is a challenge to capture people while ‘in crisis’ and to keep them motivated after the initial crisis has passed. The Maori recruiter in Auckland commented on the need for strong support and trust building to get Maori engaged with this research. It was interesting that the few Maori participants that completed the interventions in this research acknowledged they had a ‘problem’ and were now ‘ready’ to get help. It was also highlighted that Maori can tend to lose interest in counselling as there still remains a huge stigma attached to seeking help. Suggestions to improve Maori recruitment and participation included the need for significant Maori engagement in the full research process, including partnering with strong Maori organisations/iwi providers that can help refer (and not compete for) Maori participants and then support those participants through the entire process. Engagement, building relationships and trust, and providing on-going support remains the biggest challenges for gaining better Maori outcomes.

187

XGAMBLE Report

A Pacific Fono was also held to gain Pacific feedback and contextualisation of results. Again, the low numbers of Pacific participants in Phase Two of this research did not allow for analysis (as for Maori). However, the qualitative data from both phases was well received by Pacific representatives. The most important focus for Pacific peoples is “family”. Although practitioners are professional and maintain confidentiality with clients they acknowledge that engaging with whole families for education, awareness-raising (especially at social gatherings) and counselling is extremely important for the Pacific community. This has implications for the Pacific gambling workforce as they need skills in family engagement, assessment and counselling. It also was suggested that working with family and significant others could be a way to enhance Pacific engagement. Another focus for Pacific people was the importance of being/having a ‘good practitioner’ – professional manner and dress, respectful, empathetic, knowledge of Pacific culture and gambling behaviours were required for successful outcomes. It was noted that ‘free counselling’ in the Pacific context includes transportation and food costs and that Pacific practitioners often go to their clients. It was also said that having a holistic approach is important because Pacific people do not usually present for gambling issues, but often seek/need support for other issues. The gambling practitioners commented on the amount of time, energy and resources that are required to successfully work with a Pacific person, considering the crisis and other life issues that are affecting them as well. The other theme that was highlighted was the need to support Pacific people on the whole journey/counselling process, from initial engagement to follow up, and that this will take time and resources. As with Maori, the Pacific Fono contingent spoke of the importance of engaging the Pacific community right from the outset of any research through to the conclusion; to build trust, gain access, support and gather specific Pacific data and the need to support and build a Pacific workforce (including youth workers and researchers) that can provide these services.

The Asian focus group thought the results/findings were encouraging and provided the research team with some great explanations and contextualisation. They explained the isolation (from family, community and new country) that Asian problem gamblers often impose on themselves - to the point that they don’t even trust themselves. Therefore the Asian practitioners (and their clients) find real life examples of success stories to be particularly helpful and encouraging, as they instil a sense of hope. These success stories are used both as a counselling intervention and as a recruitment tool. Also, the counselling sessions are often utilised as an alternative social activity because clients are often bored in their new country and it is important for them to find other social and recreational activities to fill up their time. The practitioners believe that successful outcomes include three elements: 1) motivation to change is most important, 2) support from families and friends, and 3) professional support. Both of these sources of support provide the knowledge and help for clients to ‘stay strong’. The importance of having motivation to change is an interesting point. For example, an Asian participant in the Phase Two trial rated the intervention received as ‘not effective or satisfying’, as they were only there to ‘re-enter the casino’. It was also acknowledged that the Asian population had some differences to other population groups in this study – a larger proportion of clients were referred as part of the self-banning process at the casino or a court order, different recruitment methods were more effective (websites and word of mouth compared to generic newspapers), and the participants had a specific need for language and culture specific interventions. It was also noted that Asian problem gamblers were more likely to be casino table players (as opposed to pokie players who were required for this research) that are fewer in number and often have higher expenditure compared to other problem gamblers. This has implications for the relevance of future clinical trials to the Asian

188

XGAMBLE Report

population. The final feedback received from the Asian representatives was similar to the other ethnic population groups: there is a need to engage fully with organisations/services and community to improve Asian research outcomes, especially with regard to gaining credibility and trust in accessing participants.

Gambling operators As previously mentioned, only ClubsNZ supported this research and helped with recruitment. It is difficult to postulate how the non-participation of other gambling operators has impacted on the recruitment rate in this research.

Time, effort and expense were spent in trying to establish a relationship with a Casino. However, this did not result in a positive outcome and was very disappointing to the Project Team.

Similarly, the Charity Gaming Association (CGA) immediately declined an invitation to support this research. However, when one of the recruiters visited some bars and pubs to ask if posters could be displayed, there were mixed responses – some put the posters up and some declined. The project manager then received a message from the CEO of the CGA, that recruiters should seek permission from each individual Trust for each of the venues approached.

The Project Team believes that relationships between research organisations and the gambling operators are important and should also have some intermediary (e.g. Ministry of Health) to safeguard research integrity.

Social/ Community services As mentioned previously, several relationships have been developed with social and community services that are worth cultivating for future research opportunities. These include (but are not confined to) Probation Services, CABs, and Budgeting Services. All of these services were supportive and thought that over time, they would be able to provide referrals. National and/or regional approval is often required from their parent organisations but it seemed to be a worthwhile task, even though no (specific) referrals were received prior to the cut-off date for recruitment.

There are also other organisations that could be of assistance in recruiting and identifying problem gamblers e.g. Work and Income, Housing NZ, NZ Police etc but effort needs to be put into developing relationships and highlighting the importance of gambling with these sectors at a national level first.

Other University of Auckland sections The call centre (Survey Research Unit) has been invaluable in their support to this research. They have worked closely with the Project Team and were responsive to the changing needs of the research. They were flexible with the hours available to ensure an appropriate service was provided during the intensive advertising periods. They also trained and updated staff in new requirements, and assisted with the close monitoring of the participants through to the completion of data collection. In reflection, the only modification here would be to increase the hours of operation of a call centre and have wider ethnic representation for future research.

The Clinical Trial Research Unit also was vital to the successful completion of this project. They provided a wealth of knowledge/expertise on clinical trial implementation, developed and maintained the website utilised for data collection and analysed the majority of the quantitative data. This relationship was useful to both

189

XGAMBLE Report

parties as our separate knowledge bases provided useful experience and information in the development and utilisation of the data collection instruments/tools for this unique and vulnerable gambling population group.

Funding body The recent restructuring of the Ministry of Health in the Problem Gambling area will hopefully be beneficial to research organisations in the future as it will allow more focus and support to researchers. There was a significant portion of time during this project period where there was little communication between the funder and the Project Team. Their support and input at this time could have been valuable. This situation was rectified towards the midpoint of the project and it was noted that greater communication, planning and integration was forthcoming. Due to the importance of relationships and resource allocation, it is hoped that any future research requests for proposals will have sufficient time periods for full consultation with the multiple stakeholders that are required for the design, planning and implementation of complex research projects to high ethical and professional standards.

190

XGAMBLE Report

CHAPTER 4 SUMMARY AND CONCLUSIONS Problem gambling research has to be implemented in a developmental or step-wise fashion. This project is exploratory by nature yet it represents the achievement of a significant milestone in the development of the Ministry of Health’s gambling research programme. This project is the beginning phase of clinical trial or intervention research into problem gambling in New Zealand Aotearoa. This chapter summarises the major activities undertaken in this project and the outcomes achieved by the Project Team. It also provides an understanding of the issues and logistics of conducting this type of research in the context of New Zealand. It is followed by a discussion of the strengths and limitations of the present project. Finally, the chapter is concluded by examining the implications of the study, recommendations for future studies and a brief outline of a dissemination plan for the findings that emerged from the Phase One and Phase Two studies.

4.1 Achievements The achievements of the project are recorded in the following two columns (see Table 47). The ‘Activities’ column records the actions implemented in the course of the study period. The ‘Outcomes’ column summarises the outputs.

Table 47: Summary of the activities and outcomes of the project

Activities Outcomes

Preparation phase

Finalised the design and methodology; this included further discussion with the Ministry of Health and key stakeholders

The research design, time line and key milestones were finalised; the research contract was signed off

Reviewed literature and finalised the concept of the project

Part of the outcomes of the literature review were included in the final report of the Phase One study

Invited practitioners to join the TAG was established to provide both technical advice/ Technical Advisory Group (TAG) experience and all the necessary links to problem gambling

services providers; altogether six meetings were held throughout the course of this research project

Phase One of the study: Qualitative study involving clients (gambling clients, family members/significant others) and practitioners

Prepared and submitted ethical application for Phase One of the study

Ethical approval for Phase One was obtained (Ref: 2006 / 036)

Drafted semi-structured interview guidelines and questionnaires used for the study

A set of final interview guidelines were developed

Piloted the instruments and considered participants’ and interviewers feedback Recruited and trained ethnic interviewers from Auckland and areas external to Auckland, including Hawkes Bay

Four ethnic researchers were trained for Phase One of the project; two of them were studying towards their Doctoral or Masters degree; one was a practitioner

Recruited participants for Phase One from Auckland and Hawkes Bay

78 participants were recruited including 35 gambling clients, 16 family members and 27 practitioners. Participants covered four ethnic groups

Conducted interviews, most interviews were face-to-face and conducted at locations convenient to the participants

• Gained better understanding of client experiences and help-seeking behaviours across four population groups, in receiving problem gambling interventions (both Face-to-Face and Phone contacts; individuals receiving Monitored the progress of Phase One

XGAMBLE Report

interviews to ensure that recruited participants matched the sampling plan

1-2 sessions and multiple sessions), covering various pertinent topics e.g. how and when they identified their gambling as problematic, why they sought help, what they considered as helpful interventions, what the important elements contributing to helpful interventions were, and what practitioners could do differently to further enhance the interventions

Analysed the qualitative interview data involving. This process involved both members of the Project Team and ethnic researchers Presented preliminary data and findings in the TAG meetings, problem • Identified the needs and gained better understanding of gambling conference, and met with the the experience of whanau, family members or Ministry of Health to verify the significant others whose loved one was affected by trustworthiness of findings problem gambling; the topic areas covered were similar

to above • Gained better understanding of the New Zealand

problem gambling workforce covering various pertinent topics e.g. types of problems presented to practitioners, what they actually did during the counselling sessions, how often those tasks were performed, practice models/techniques used, how to meet clients’ cultural and linguistic needs

• Produced a concept diagram to provide an account on what makes successful problem gambling counselling interventions

• All the objectives of Phase One study were achieved

Discussed and completed write-up of the findings after taking into account the feedback received

Disseminated the Phase One study to the general public

A few items were published in the news media to disseminate the findings and promote the Phase Two study

Maintained regular communication Regular HealthPac and Ministry of Health progress reports with the Ministry of Health regarding were submitted the progress and design of the research and problems that arose

Phase Two of the study: Randomised trial on ‘Face-to-Face’ versus ‘Phone’ based interventions for individuals with problem gambling

Reviewed key findings from Phase One to inform the planning and implementation of Phase Two

The Intervention Manual reflects the findings from Phase One; it also informed the training of practitioners recruited for this study

Prepared and submitted ethical application for Phase Two of the study

Ethical approval for Phase Two was obtained (Ref 2006/203)

Drafted a set of questions to use for the study

• A set of intervention study measurements were developed

• An entire website system was established and used for data collection and real-time data entry

Held discussions with the Clinical Trial Research Unit (CTRU) in relation to measurement logistics (the measuring instruments, how the instruments would be used for data collection via the website) Recruited and trained researchers working in the Call Centre from the Survey Research Unit (SRU); this included role plays and piloting the use of questionnaires

A team of multi-ethnic and lingual researchers were formed to work with prospective participants

Recruited and worked together with a • A practice manual on problem gambling counselling small group of experienced was published practitioners to produce the ‘Trial • The Project Team ‘manualised/formalised what the Practice Manual’ interventions entail’- thus satisfying one of the best

practice standards for clinical trials Recruited a group of problem gambling A team of multi-ethnic and lingual problem gambling practitioners to provide counselling for practitioners were formed to work with participants enrolled clients recruited for this study; a one- for this study day training workshop was held in Auckland Recruited and trained a small group of ethnic researchers to assist with

A team of multi-ethnic and lingual recruiters were formed to recruit prospective participants

192

XGAMBLE Report

recruitment Met with multiple groups (government agencies, NGOs, members of gambling industries), media and news reporters (general and ethnic specific) to seek their support for recruitment

• 111 people contacted the Call Centre • 96 people were successfully randomised for the trial • Data were collected from both clients and practitioners

at baseline, post intervention and six week follow up points

Constantly reviewed recruitment progress and budget allocation; and adjusted the course of actions aimed at achieving the targeted outcomes Monitored the recruitment progress and ensured smooth running of the CTRU recruitment website and Call Centre at SRU Maintained regular contact with practitioners and ensured smooth seamless interfaces among: staff running the website, Call Centre staff and practitioners Contacted practitioners to collect data about the type of interventions they provided and clients progress Communicated with CTRU staff regularly to respond to their queries and to ensure the integrity of the data in preparation for data analysis

Evidence was gathered to respond to the objectives of the study: Objective 1) Compare the outcomes of ‘Face-to-Face’ and ‘Phone’ based intervention, Objective 2) Magnitude of the effects, if different

Analysed the data both quantitatively and qualitatively (e.g. open questions in the questionnaires for clients and practitioners) Held meetings with relevant staff to finalise the analyses plan and the final results Kept regular communication with the • Regular HealthPac and Ministry of Health progress Ministry of Health regarding progress reports were submitted of the research, its design and any • Request for time extension was submitted (and was arising problems approved)

Putting together: Gathering feedback and preparing the Final Report

Integrated Phase One and Two data / findings against the overall set of objectives in the research project

• The feasibly of conducting a problem gambling clinical trial study in New Zealand was evaluated

• Key aspects of the research design were reviewed e.g. usefulness of the measurements, recruitment effectiveness

• Limitations of the research were reviewed • Objective 3) Making recommendations for the next

phase of clinical study for problem gambling was achieved

• Dissemination plan was devised • The final report was prepared

Engaged ethnic researchers in analysing the data and writing (and proof-reading) the draft final report Presented the findings to the Maori, Pacific and Asian stakeholder groups (three separate meetings were organised in early October, 2007) to verify the interpretation of findings and gather feedback on their experience of being involved in the study Presented the findings to the TAG to gather their feedback and further analyse the data while considering the comments made Presented the findings to the service providers/practitioners to debrief, gather feedback, and further analyse the data Compared the present findings against

193

XGAMBLE Report

existing body of knowledge - locally and internationally Reflected on the learning and experience gained from the study Discussed with the funder about the outline and key foci of the final report Formed a core group of researchers to draft the final report

Another pleasing (and in some ways unintended) ‘achievement’ of the Phase Two study was the fact that the majority of individuals who went through the interventions and were re-assessed showed major reductions in the amount of time and money spent on gambling, not to mention improvements on psychological measures. It is worth mentioning two very ‘alarming’ cases: • An individual self-reported spending $160,000 over four weeks on

gambling, this reduced to: $12,000 and $4,000 at the post-intervention and follow up visits respectively.

• Another person self-reported spending nine times (or 900%) their household income (before tax) on gambling, this reduced to: none at post-intervention visit, then very slightly shifted back to only 7% of the person’s income at follow up.

4.2 Strengths of the study Mixed designs: Quantitative and qualitative methodology This project was divided into two phases. Phase One was primarily a qualitative study to capture the wealth of experience of individuals using, and practitioners providing, problem gambling services. Phase Two was a pilot/vanguard randomised trial, to compare and contrast the outcomes achieved by ‘Face-to-Face’ and ‘Phone’ based modalities of interventions. The qualitative design in Phase One provided a rich description of all the relevant contextual issues and considerations that had to be taken into account for the Phase Two study. The Phase Two study began to examine or verify some of the observations or experiences gained in the Phase One study.

The application of quantitative methodologies in this project was well supported by the expertise (e.g. biostatisticians, data manager) provided by the Clinical Trial Research Unit - the only clinical trial unit based in this country. The qualitative methodology employed by the Project Team was equally well supported by the Social and Community Health Section, which has a track record of conducting qualitative studies, within the School of Population. It is also worth mentioning that the Project Team’s location within the School provided them with easy access to other relevant expertise in relation to addiction and population studies.

Strong infra-structure to support the study In addition to the topic or methodological expertises availed to the Project Team, we had the privilege of working closely with the Clinical Trial Research Unit and Survey Research Unit. These two research centres were able to provide a strong infra-structure to support the implementation of the present project, for example, development of a research website data tool, internet based clinical trial management system, and computer-assisted telephone interview system (CATI).

194

XGAMBLE Report

Use of triangulation technique One of the positive features of this project was the ‘triangulation’. The practice of triangulation can be divided into three layers: 1) triangulation of data sources - collecting data from users, whanua/significant others and practitioners; 2) triangulation of data analysis - research members’ interpretation of data were verified or triangulated by checking with key stakeholders in the field; and, 3) triangulation of data which emerged from mixed research designs (i.e. qualitative and quantitative methodologies) over the two-year period of the study.

Ethnic populations and ethnic researchers Members of this project were able to recruit significant numbers of participants from the Maori, Pacific and Asian communities – something which is not commonly achieved in problem gambling research locally and internationally. In the present project there were 66.7% of participants in the Phase One study (Table 2) and 45% in the Phase Two study (Table 13), that could be grouped as ‘non-European’. A sample of this nature could broadly represent the New Zealand population landscape. Furthermore this Project Team consisted of members who did not only bring topic expertise but also cultural expertise to the project. The involvement of ethnic researchers serves three important functions: 1) providing strong links to the specific population groups; 2) ensuring the data were collected and analysed in a culturally sensitive manner; and, 3) enhancing the dissemination of data and knowledge transfer to the communities concerned.

Help-seeking experience: Retrospective views and current experience Considering the project as a whole, it became apparent that the two phases of the study could capture the retrospective views of individuals’ experiences of using problem gambling services (evident in Phase One) and individuals’ current counselling experiences as they unfolded over the course of the Phase Two of the study. The caveat relating to this is that only limited participants remained in the study, and thus provided comments about their current counselling experience, at the post-intervention and follow up visits respectively.

Study across two universities Working with the international named investigator (AJ), this project had the advantage of comparing the New Zealand data against the Australian work and experience, regarding the Counsellors Tasks Analysis (CTA). AJ is one of the developers of the CTA and having him on board allowed some benchmarking of the findings that emerged from New Zealand against the overseas experience.

High external validity evident in the Phase Two study External and internal validity are two fundamental challenges in clinical trials. While the Phase Two study suffered a major problem in terms of heavy drop out rates (with only about two-thirds of the total sample attending between 1-8 sessions), the project also had strong evidence of external validity, that is, a good reflection of what is happening in the real world. Unfortunately the high drop out rate posed a serious threat to the internal validity of the project findings.

Problem gambling research workforce development Upon reflection, this project provided a unique opportunity for problem gambling research workforce development. Altogether, two doctoral candidates and three masters’ students in health sciences or public health were involved in the present project in various capacities. The challenge remains though, how many of these bright

195

XGAMBLE Report

young scholars can be kept and supported by a rewarding and secure career structure in the gambling field in New Zealand.

The design of the present study followed a pragmatic, effectiveness trial model rather than a more highly controlled efficacy model. Effectiveness research assumes that the intervention resembles ‘usual practice’ as opposed to ‘best practice’ (Wells, 1999). It also assumes that the research participants sample closely resembles the ‘usual’ clients encountered in clinical practice. This study filled a small but significant gap in the literature on problem gambling interventions in New Zealand. This is the first experimental study of positive improvement in an array of measures on gambling intensity, frequency, associated symptoms, behaviours and attitudes following face-to-face and telephone interventions for problem gambling.

4.3 Limitations of the study Like most research studies, the present project is subject to several limitations and qualifications. There are three major limitations for this project: large drop out rates at post-intervention and follow up, small sample size and the absence of a control group factored into the design.

Drop Outs The present study suffered from a large drop out rate (71%) which is often considered a serious challenge in problem gambling intervention studies (e.g. Ledgerwood & Petry, 2005; Toneatto & Ladouceur, 2003). The participants, who chose to stay or were still contactable in the post-intervention (and follow up) visits, were probably susceptible to self-selection biases and demand characteristics, which could account for the improvements found in the present study. Although a closer examination of the number of sessions attended by participants in this study suggests it was rather comparable to the ‘normal’ behaviour of clients presenting to problem gambling services in New Zealand (see Table 41). Thirty-eight percent of participants attended six or more sessions in this study compared to national users statistics where 37.3% of users had between three and ten hours of counselling, or 34.6% of users had treatment between one and six months (Ministry of Health, 2006, p.15).

Milton, Crino, Hunt and Prosser (2002) made eight specific recommendations on how to enhance treatment adherence: • clients are positively reinforced and praised, a confirmation letter should be

sent after the appointment is made; • towards the end of the first assessment session, clients are given a positive

prognosis and are again praised for making the attempt to change their gambling behaviours;

• in the first assessment session, clients are informed that they will be contacted if they miss one session, this will be done on one occasion;

• after the first session, clients will be sent a short letter highlighting the positive prognosis, praise and encouragement;

• assessment results are discussed with the clients in graphical and numerical ways;

• clients are asked to fill in an evaluation sheet between every session; • clients are supported in identifying and removing barriers that hinder changes;

and, • the clients sense of self-efficacy is reinforced throughout the course of

interventions.

196

XGAMBLE Report

Upon closer examination of those who adhered to the post-intervention visits and those who did not, there were a few major trends (see Table 48). All Chinese and Koreans (n=5) enrolled in this study completed the post-intervention visits, but this did not apply to other population groups. Generally, there were more men in the completing group compared to those who dropped out from the study. And relatively speaking, those who remained in the study gambled at less intense levels, compared to their counterparts in the ‘drop out’ group. Those who felt the intervention would be ‘extremely effective’ were much more likely to stay in the trial.

Table 48: Compare those who adhered to post-intervention visits and those who left the study

Demographics and measurements at baseline Those who adhered to post-intervention and follow ups

(n=27)

Those who were neither contactable nor attending any post-intervention visits (n=

65)

Preferred ethnicity

9 or 33.3% were non-NZ European

3 Koreans, 2 Chinese, 2 Other, 1 Maori, 1 Pacific

28 or 43.1% were non-NZ European

17 Maori, 7 Pacific & 4 Other

Sex 11 men (40.7%) 19 men (29.2%)

Age mean= 43 median= 45

mean= 43.3 median= 44

Money spent mean= $4,169 median= $1,000

mean= $24,000 median= $1,750

Time spent mean= 18.4 hours median= 12 hours

mean= 34.6 hours median= 21 hours

Percentage money spent mean= 137.9% median= 66.7%

mean= 154.5% median= 56.4%

GABS mean= 93.1 median= 94

mean= 92.2 median= 91

GRTC mean= 20 median= 20

mean= 19.9 median= 20

Number of sessions attended

mean-= 5.1 median= 6.0

mean= 4.4 median= 5

Regarded the would-be interventions as ‘extremely effective’

n=12 or 44.4% n=6 or 9%

High levels of drop outs or attrition rates have significant impacts on the outcome of intervention studies. It does not only limit the internal validity of the findings but can also over-estimate the treatment size effect (Westphal, 2006). Recommended options are: • Applying ‘compliance-improving’ interventions (Milton et al., 2002). • Increasing the sample size calculation to absorb the expectedly high drop out

rates without taxing the sample base. • Over-recruiting the number of people in the experimental group (for details,

see Echeburua, Baez & Fernandez-Montalvo, 1996; Petry et al., 2007). • Rewarding participants (e.g. petrol voucher) who stayed on the trial. • Assigning a specific research team member to contact and encourage

participants to stay on the project. • Build in multiple-contact points throughout the course of study.

197

XGAMBLE Report

Having discussed all the problems surrounding drop outs, there is an alternative view to interpreting this ‘problem’. To date, addiction research has shown three trends: 1) the drop out rates are very high in the addiction and problem gambling field; 2) effort to reinforce treatment compliance has only partial success; and, 3) people said they went for more sessions than they actually did and yet people benefited enormously from the brief sessions (e.g. one or two sessions) they attended (e.g. Pulford, Adams & Sheridan, 2007; Pulford et al., 2006). The question is what service model best caters for this presentation reality - which is short term (in this project the average number of sessions attended was 2.5). We may be working from an idealised service model which makes the assumption that retention in treatment is a desirable end. The traditional addiction services tend to fit people into that model or ‘box’, as opposed to thinking the other way round, such as, how we can re-organise problem gambling services to engage service users in a way that best maximises the brief episodes of client involvement.

However, this study is the first to demonstrate constructive change in a measure of attitudes and beliefs with concomitant reductions in gambling behaviours within a simple pre to post-intervention.

The next step is to determine whether the improvements achieved can be sustained over time and employ a realistic and effective research design to ascertain the cause-effect relationships between interventions and outcome measures. In a comprehensive review of intervention effectiveness, it was concluded that two-thirds of treated cases were reported as abstinent or controlled and that such behaviour change was often accompanied by more general improvement in psychosocial functioning (Lopez Viets & Miller, 1997). Without a full-scale research design one can never dispute the so-called ‘critical period’ proposed by Litt, Kadden, Cooney and Kabela (2003). They hypothesised that individuals presenting for treatment may already have been committed to the idea of decreasing their problem behaviours and the current treatment programme gave them the opportunity (or excuse) to do it.

Small sample size The Project Team were not able to recruit the target of two hundred participants in order to achieve the statistical power that the study originally sought. Furthermore, achieving this larger number would have reduced the negative impact of the unexpectedly high drop out rates. Within the Phase One study, one criticism could be the relatively small number of users and family members (total was 51) across ethnicities that were engaged in the project. However it is of the Project Team’s opinion that the themes and data that emerged from Phase One had become repetitive, thereby indicating that the range of views had reached saturation. This was supported by key stakeholders’ feedback that no new categories or themes had to be added to the existing data set.

Lack of a control group This study was specifically designed to investigate the potentially different outcomes delivered by two modes of service delivery, without using a third arm of a ‘control group’. Undeniably this was one of the weak points in the design of the project. It did not allow any causal inference about the likely impacts of the ‘Face-to-Face’ and ‘Phone’ based interventions on participants. However, this apparent weakness has to be interpreted within the local context; this project is the first effectiveness study on problem gambling interventions to be conducted in New Zealand. Therefore the

198

XGAMBLE Report

learning gained from this study perhaps outweighs, to some extent, the limitations posed by the lack of control group in the design.

The naturalistic, pre and post design of the present study greatly limits the inferences that can be drawn. Therefore we cannot say that the interventions were responsible for reducing the money wagered or changing participants’ gambling attitudes, beliefs and behaviours; only that the data were consistent with such a formulation. Despite this, the research was able to shed light on the research questions and develop a range of interpretations based on the observed data. The next phase of development of intervention studies will have to improve on several areas, especially in managing drop out rates and difficulties in participant recruitment. Specific suggestions can be found in the Recommendation Section of this chapter.

4.3 Implications It is very important to emphasise again that the findings and the implications from this project have to be interpreted within the constraints of the present project. It is flawed or even dangerous to make any substantial changes to services based on the present study. What would be useful though is to begin exploring what the findings mean to the following areas: namely service delivery and research workforce development in the problem gambling field.

Service delivery • There are strong parallels between the Phase Two study and establishing a ‘new

problem gambling service’. It took the research team at least one year to recruit 111 people onto the register for the clinical trial. It is evident to the research team, that newly established services, both general and culturally specific, do take time (perhaps between one and half and two years) to establish their name and credibility in the community.

• It was evident in both the Phase One and Two studies that the quality of the therapeutic relationship between the practitioners and clients is critical. In particular, it is very important for practitioners to be empathetic, non-judgemental and respectful. It is paramount, both at the level of service engagement and at the level of encounters between practitioners and clients, that this kind of warmth and engaging atmosphere is conveyed clearly to the service users, regardless of the seriousness of the problems or of the clients’ socio-cultural, linguistic background.

• The client-service engagement issues (e.g. time taken, and in some cases the extra effort required – such as ethnic researchers accompanying prospective participants when enrolling in the study) revealed in this project, suggest that more resources are needed to ensure that new clients are comfortable and confident about the problem gambling services.

• One clear message from this project was the value of giving clients options or different modes of problem gambling services. The observed data and the trend of findings provide emerging evidence to support the use of both ‘face-to-face’ and ‘telephone’ based interventions for problem gambling. It is also important to note that this study only recruited people who were affected by pokie machines in casinos, pubs/bars and clubs. It is highly likely that different populations (age and gender groups, people affected by different forms of gambling activities) and interacting with individuals’ stage of gambling problems, might benefit from the option of alternative forms of services. For example, texting on mobile phones, accessing web-based counselling, group (vs. individual), attending couple counselling, working through a self-paced workbook, attending a residential

199

XGAMBLE Report

programme (versus community), and accessing culturally specific and language based (versus generic) treatment facilities.

• When comparing the responses from the New Zealand practitioners with their Australian counterparts on the Counsellors Tasks Analysis, it was apparent that the New Zealand practitioners tended to adopt an ‘integrated approach’ in their work; that is, they tend to combine counselling together with health promotion and community work. This seems to be a distinctive feature of the New Zealand problem gambling services.

Research Workforce • The majority of the agencies and individual practitioners were immensely

supportive of the present research project. In some cases their inability to help was more related to the substantial time pressures or work loads they were experiencing, particularly for those who worked only part time in problem gambling services and who were involved in multiple research projects at the same time. It is advisable to: 1) continue to promote the value and importance of research and/or evidence-informed practice in the sector; 2) allow counselling or clinical practitioners time and acknowledgement for their involvement in research projects; 3) support staff exchange programme whereby practitioners are seconded to a gambling research centre, or university-based researchers work at a problem gambling service provider for a defined period of time; and, 4) support practitioners who are interested in pursuing research as their career, to undertake further training as they are in the ideal position to bridge the divide between practice and principles or theories.

4.4 Recommendations for future studies

4.4.1 Contextual issues This section takes into consideration the relevant contextual factors when planning other problem gambling intervention studies in New Zealand. The following recommendations are based on both the research team’s reflections or learning gained from conducting this project and specific comments gathered while the Project Team met with key stakeholders towards the end of the project.

Coordination of research projects There is a need for commissioned research within the gambling area to be carefully coordinated. In particular, there is a danger of overburdening the small communities relevant to this field (e.g. people with gambling problems and their family/whanau, providers/practitioners). As such, the timing, scope and concurrent scheduling of future research projects will require careful consideration.

Sufficiency of allocated time for research The establishment of trust and communication channels between researchers and key stakeholders (including prospective participants, problem gambling service providers/practitioners, ethnic/cultural groups, communities, gambling industry) takes substantial time, particularly within the challenging field of problem gambling. The forming of these relationships are critical to the conduct of successful research and their importance cannot be underestimated. As such, the amount of time allocated to future research will require careful consideration.

Partnerships with providers /practitioners

200

XGAMBLE Report

The input and contributions of providers/practitioners to research are extremely important. Adequate resources are required to ensure that their participation in research is fully resourced and funded. There is potential for research projects which have been developed in partnership with providers/practitioners to benefit substantially from their endorsement (e.g. project media material includes endorsement wordings from CEO of service providers or well-known community leaders).

Partnerships with ethnic/population groups Apparently the word ‘research’ is “…one the dirtiest words in the indigenous world’s vocabulary…It has been a time of transition from ‘Maori as the researched’ to ‘Maori as the researcher’” (Herd & Richards, 2004, p. 180). Productive partnerships with ethnic/population groups require time and resources. Developing ethnic researchers and building partnerships between researchers and ethnic/ population groups are particularly important to ensure: 1) key ethnic/population groups are involved in the design and planning of the project at an early stage; 2) the research sample reflects the New Zealand population composition in the 21st century; 3) sufficient number of participants are recruited for the study; and, 4) an effective dissemination plan is employed to transfer the knowledge and feedback the findings to the communities concerned. It is important that the development and maintenance of such partnerships are resourced in an adequate and meaningful manner.

Partnerships with members of the gambling industry There is a need for an intermediary body (e.g. the Ministry of Health) to aid the negotiation of these partnerships and ensure the credibility and integrity of members of the research community and the research project itself (for details on challenges associated with forming partnerships between gambling industry and researchers, see Adams, 2007)

4.4.2 Specific methodological considerations Using control group The design logic and what constitutes a meaningful ‘control group’ requires careful consideration. Common options are: • Cross over designs • ‘Waiting list’ as control group • ‘Usual care’ as control group • ‘GA and workbook’ as control group

The decision about which design logic and ‘control group’ to be used in future research will be determined by not only the research question(s) under study but also the reality in practice and the specific New Zealand context, for example, the limited population size and availability of potential participants.

Selecting appropriate and sensitive outcome measure(s) Given our experience in the present study, we consider that the primary and secondary outcome measures were appropriate and sensitive enough to detect changes in participants’ gambling behaviours. Some minor recommended changes to the measures include: • Collecting data on participants’ amount of disposable income (versus

household income before tax). • Using the Timeline Follow-Back (TLFB) method and working with

‘collaterals’ (Petry et al., 2007).

201

XGAMBLE Report

• Re-measuring the DSM-IV during post-intervention and follow up visits. • Making sure participants fully understand the definitions used during data

collection, for example the wording of ‘average’, ‘disposable income’. • In addition to the decision on which instruments to use in future studies, it is

equally important to plan how and when the data will be collected. This needs to be conducted in ways that minimise the possibility that research personnel and practitioners ask similar questions twice in their interviewing.

• Mobile telephones were found to be a useful way of collecting data around gambling behaviours and changes in mood state in real time (Gee, Coventry & Birkenhead, 2005).

• Lastly, the selection of outcome measures has to be determined by the theoretical assumptions underpinning the design of the trial.

Description of the interventions In a trial such as this, it is important to ensure that procedures are standardised and that the ‘therapy’ under investigation is manualised accordingly in order for it to provide sufficient training to practitioners. In this study, the manual was produced and one-full day workshop was conducted to cover the procedures with practitioners.

Training of practitioners We view the level of training provided as sufficient for the purpose of the present study. However, when evaluating intervention techniques in the future, particularly with less experienced gambling practitioners, it would be desirable to provide more intensive training and individual coaching or monitoring. The manualisation of the interventions and training of practitioners seeks to minimise inter-practitioner variations and to provide exactness about the elements of the therapy and what contributes to the positive outcomes if found.

Recording number of interventions sessions It is strongly recommended to keep extremely clear records of the number of sessions attended by problem gambling clients. In the problem gambling research field, whether it is the length of treatment, or number of sessions that has bearing on intervention outcomes remains a contentious issue. People who adopt the spontaneous recovery model assert that high investment in professional interventions may not be cost effective. Our critique of this position argues that when comparing outcomes, spontaneously remitted clients are less able to deal with gambling triggers and manage relapse risks than treated clients.

Assessing intervention effectiveness It is strongly recommended that a reasonable length of time for follow up assessment upon the completion of intervention is allowed for. This enables a true reflection of the intervention effects. Recommended options are: • Allow for a usual follow up time for of 12-18 months for behavioural

interventions and 6-12 months for cognitive therapies (Petry, 2006). • Measure both the problem gambling symptoms and associated behaviours. • Assess the broader mental health and wellbeing functions e.g. employment,

relationships (see the Discussion in the Phase One qualitative study: Section 2.4.1).

• Attempt to follow up and report status of all clients entered into the trial, including dropouts.

• Verify client self-report by interviewing knowledgeable collaterals.

202

XGAMBLE Report

4.4.3 Priorities for future studies • Compare different types of interventions: simple support or specific treatment

techniques e.g. cognitive, cognitive-behavioural, or motivational enhancement techniques.

• Evaluate outcomes of different modes of problem gambling service deliveries: individual, small groups, workbook, web-based, phone contacts, text-based mobile phone, self-help group, couple and family therapies.

• Study practitioners’ characteristics: future research should attend to this potentially important determinant of gambling outcomes.

• Improve understanding of the mechanism of intervention efficacy: the mechanism of action of various techniques may generate further hypotheses regarding the etiology of the condition and may lead to further improvement of therapies, determine which elements have had specific efficacy, only behavioural treatments have been tested sufficiently to be judged as “effective”.

• Improve the matching between interventions provided by services and the needs and presentation patterns of potential clients.

• Study factors that contribute to treatment failures and relapse in clients with problem gambling.

• Study various sub-groups of the population: o Youth; o Older adults; o Women; o Families affected by problem gambling; o Indigenous/ native people; o Ethno-cultural minorities; o Refugees; o Patients with comorbid mental disorders; and, o Inmates. o Specific occupational groups (e.g. food industry, shift workers)

(Korn & Shaffer, 2004; Ledgerwood & Petry, 2005; Lopez Viets & Miller, 1997; Najavits & Weiss, 1994; Tan & Tam, 2003).

4.5 Dissemination plan The dissemination of research findings is planned within the regulations of the University of Auckland Human Participants Ethics Committee. The results of this project will be detailed in a comprehensive report that will be made available to the Ministry of Health, other key stakeholder groups, and relevant social and counselling services.

Problem gambling service providers are interested in successful ways to reduce gambling related harms in various population groups, including youth and older people. Therefore, the results of this trial will be provided for the organisations to allow incorporation into their intervention initiatives for individuals affected by problem gambling. The appropriate and relevant findings will also be disseminated to the community through various media format e.g. newspaper items, and radio. Furthermore, the researchers will prepare articles both on the overall project and/or specific aspects for publication in international, peer-reviewed journals such as Journal of Gambling Studies, International Gambling Studies, International Journal of Mental Health and Addiction, New Zealand Journal of Psychology, Electronic

203

XGAMBLE Report

Journal of Gambling Issues: eGambling, British Journal of Guidance and Counselling, and Addiction.

To sum up, on completion of this project (including the submission of final report), members of the Project Team will disseminate research results: • to key stakeholders (e.g. government departments, problem gambling service

providers) through meetings and reports; • to the wider community of social and counselling agencies through seminars and

conference presentations; • to the wider circle of researchers or research students and communities of interest

via published papers, meetings, conferences and other academic forums; and, • to the New Zealand mainstream and ethnic/cultural communities (regarding

relevant findings) using a variety of media (e.g. hui, fono, TV and radio programmes, and ethnic community newspapers).

204

XGAMBLE Report

CHAPTER 5 REFERENCES

Abbott, M.W. (2001). Problem and non-problem gambling in New Zealand: a report on phase two of the 1999 National Prevalence Study. Wellington: Department of Internal Affairs.

Abbott, M. W. & Volberg, R. A. (2000). Taking the pulse on gambling and problem gambling in New Zealand: Phase One of the 1999 National Prevalence Survey. Report number three of the New Zealand Gaming Survey. Wellington: Department of Internal Affairs.

Adams, P. (2004). The history of gambling in New Zealand. Journal of Gambling Issues. Retrieved October 27, 2007 from http://www.camh.net/egambling/issue12/jgi_12_adams.html#overview

Adams, P. J. (2007). Assessing whether to receive funding support from tobacco, alcohol, gambling and other dangerous consumption industries. Addiction, 102, 1027-1033.

American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

Bell, C., & Lyall, J. (2002). One night out gambling. In B. Curtis (Ed.), Gambling in New Zealand. (pp. 231-244). Palmerston North: The Dunmore.

Bellringer, M., Perese, L. M., Abbott, M. W., Williams, M. M. (2006). Gambling among Pacific mothers living in New Zealand. International Gambling Studies, 6(2), 217-235.

Blaszczynski, A., Huynh, S., Dumlao, V. J., Farrell, E. (1998). Problem gambling within a Chinese speaking community. Journal of Gambling Studies, 14(4), 359-380.

Blaszczynski, A. & McConaghy, N. (1989). The medical model of pathological gambling: Current shortcomings. Journal of Gambling Behavior, 5, 42-52.

Bramley, D., Riddell, T., Whittaker, R., Corbett, T., Lin, R., Wills, M., Jones, M., & Rodgers, A. (2005). Smoking cessation using mobile phone text messaging is as effective in Maori as non-Maori. New Zealand Medical Journal, 118 (1216), 1-10.

Breen, R. B., & Zuckerman, M. (1999). Chasing in gambling behaviour: Personality and cognitive determinants. Personality and Individual Differences, 27, 1097-1111.

Breen, R. B., Kruedelbach, N. G., & Walker, H. I. (2001). Cognitive changes in pathological gamblers following a 27-day inpatient program. Psychology of Addictive Behaviors, 15(3), 246-248.

Brown, R., & Raeburn, J. (2001). Gambling, harm and health. Two perspectives on ways to minimise harm and maximise health regard to gambling in New Zealand. Auckland: The Problem Gambling Committee of New Zealand and the Gambling Studies Institute of New Zealand.

205

XGAMBLE Report

Cheers, B. (2000). Social support in remote areas of Australia. Rural Social Work, 6, 4-16.

Chevalier, S., & Griffiths, M. (2004). Why don't adolescents turn up for gambling treatment (revisited)? Electronic Journal of Gambling Issues: eGambling, 11.

Chinese Family Life Services of Metro Toronto, (1995). Working with gambling problems in the Chinese community. Toronto: A United Way Member Agency.

Ciarrocchi, J. W., & Reinert, D. F. (1993). Family environment and length of recovery for married male members of Gamblers Anonymous and female members of GamAnon. Journal of Gambling Studies, 9, 341–351.

Clarke, D., Abbott, M., DeSouza, R., Bellringer, M. (2007). An overview of help seeking by problem gamblers and their families including barriers to and relevance of services. International Journal of Mental Health and Addiction. Retrieved October 27, 2007 from http://www.ijma-journal.com/content/proc/abstracts_presentations/5/00017

Clarke, D., Tse, S., Abbott, M., Townsend, S., Kingi, P., & Manaia, W. (2007). Reasons for starting and continuing gambling in a mixed ethnic community sample of pathological and non-problem gamblers. International Gambling Studies, 7(3), 299-313.

Coman, G.J., Burrows, G. D. & Evans, B. J. (2001). Telephone counselling in Australia: applicatons and considerations for use. British Journal of Guidance and Counselling, 29(2), 247-258.

Coman, G. J., Evans, B. J., & Burrows, G. D. (2005). An innovative cognitive strategy to assist problem gamblers. British Journal of Guidance and Counselling, 33(1), 129-140.

Coman, G.J., Evans, B.J. & Burrows, G.D. (1996). Problem gambling: treatment strategies and rationale for the use of hypnosis as a treatment adjunct. Australian Journal of Clinical and Experimental Hypnosis, 24 (2), 73-91.

Costello, T. & Millar, R. (2000). Wanna bet? Winners and losers in gambling’s luck myth. St. Leonards, N.S.W.: Allen & Unwin.

Cox, B., Enns, M. W., & Michaud, V. (2004). Comparison between the South Oaks Gambling Screen and a DSM-IV-Based Interview in a community survey of problem gambling. Canadian Journal of Psychiatry, 49(4), 258-264.

Creswell, J. W. (1998). Qualitative inquiry and research design: Choosing among five traditions. Thousand Oaks: SAGE Publications.

Crisp, B. R., Jackson, A. C., Thomas, S. A., Thomason, N., Smith, S., Borrell, J., Ho, W. & Holt, T. A., (2001). Is more better? The relationship between outcomes achieved by problem gamblers and the number of counselling sessions attended, Australian Social Work, 54(3), 83-92.

206

XGAMBLE Report

Crisp, B. R., Thomas, S. A., Jackson, A. C., Smith, S., Borrell, J., Ho, W., Holt, T. A., & Thomason, N. (2004). Not the same: A comparison of female and male clients seeking treatment from problem gambling counselling services. Journal of Gambling Studies, 20(3), 283-299.

Crockford, D. N., & el-Guebaly, N. (1998). Psychiatric comorbidity in pathological gambling: A critical review. Canadian Journal of Psychiatry, 43, 43-50.

Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika. 16, 297-334.

Cultural Partners Australian Consortium (2000). The impact of gaming on specific cultural groups. Victoria: Author.

Currie, S. R., Hodgins, D. C., Wang, J., el-Guebaly, N., Wynne, H., & Chen, S. (2006). Risk of harm among gamblers in the general population as a function of level of participation in gambling activities. Addiction, 101, 570-580.

Daughters, S. B., Lejuez, C. W., Lesieur, H. R., Strong, D. R., & Zvolensky, M. J. (2003). Towards a better understanding of gambling treatment failure: Implications of translational research. Clinical Psychology Review, 23, 573-586.

Dickerson, M., Hinchy, J., & Legg England, S. (1990). Minimal treatments and problem gamblers: A preliminary investigation. Journal of Gambling Studies, 6(1), 87-103.

DiClemente, C. C. (2003). Addiction and change. How addictions develop and addicted people recover. London: Guildford Press.

Dyall, L. (2002). Kanohi kit e Kanohi: Face to face A Māori face to gambling. New Ethicals Journal New Zealand’s Journal of Patient Management, 5(1), 11-16.

Dyall, L., & Hand, J. (2003). Maori and gambling: Why a comprehensive Maori public health response is required in New Zealand. International Journal of Mental and Addiction. Retrieved June 24, 2007 from http://www.ijma-journal.com/content/full/1/1/5

Echeburua, E., Baez, C., & Fernandez-Montalvo, J. (1996). Comparative effectiveness of three therapeutic modalities in the psychological treatment of pathological gambling: Long-term outcome. Behavioral Cognitive Psychotherapy, 24, 51-72.

Echeburua, E., Baez, C., Fernandez-Montalvo, J. (1996). Minimal treatments and problem gamblers: A preliminary investigation. Journal of Gambling Studies, 6, 87-101.

Echeburua, E., Fernandez-Montalvo, J., & Baez, C. (2000). Prevention in the treatment of slot-machine pathological gambling: Long term outcome. Behavior Therapy, 31, 351-364.

Gee, P., Coventry, K. R., & Birkenhead, D. (2005). Mood state and gambling: Using mobile telephones to track emotions. British Journal of Psychology, 96, 53-66.

207

XGAMBLE Report

George, S. & Murali, V. (2005). Pathological gambling: an overview of assessment and treatment. Advances in Psychiatric Treatment, 11, 450-456.

Goodman, R. (1995). The luck business: The devastating consequences and broken promises of America’s gambling explosion. New York: Simon and Schuster.

Griffins, M., & Cooper, G. (2003). Online therapy: Implications for problem gamblers and clinicians. British Journal of Guidance and Counselling, 31(1), 113-123.

Griffiths M, Bellringer P, Farrell-Roberts K, & Freestone F. (2001) Treating problem gamblers: a residential therapy approach. Journal of Gambling Studies, 17(2):161-169.

Griffiths, M., & MacDonald, H. F. (1999). Counselling in the treatment of pathological gambling: An overview. Journal of Guidance and Counselling, 27, 179-190.

Haas, L. J., Benedict, J.G., Kobos, J. C. (1996). Psychotherapy by Telephone: Risks and Benefits for psychologists and Consumers. Professional Psychology: Research and Practice, 27, 154-160.

Heineman, M. (1987). A comparison: The treatment of wives of alcoholics with the treatment of wives of pathological gamblers. Journal of Gambling Behavior, 3, 27–40.

Heineman, M. (1992). Losing your shirt: Recovery for compulsive gamblers and their families. Minneapolis, MN: CompCare.

Herd, R., & Richards, D. (2004). Wahine tupono: A kaupapa Maori intervention programme. In R. Tan & S. Wurtzburg (Eds.). Problem gambling: New Zealand perspective on treatment (pp.172-181). Wellington: Steele Roberts Ltd.

Hodgins D. C., & el-Guebaly, N. (2000) Natural and treatment-assisted recovery from gambling problems: a comparison of resolved and active gamblers. Addiction, 95, 777-789.

Hodgins, D. C., & el-Guebaly, N. (2004). Retrospective and prospective reports of precipitants to relapse in pathological gambling. Journal of Consulting and Clinical Psychology, 72(1), 72-80.

Hodgins, D. C., Currie, S. R., & el-Guebaly, N. (2001). Motivational enhancement and self help treatments for problem gambling. Journal of Consulting and Clinical Psychology, 69(1), 50-57.

Hodgins, D. C., Makarchuk, K., el-Guebaly, N., & Peden, N. (2002). Why problem gamblers quit gambling: A comparison of methods and samples. Addiction Research and Theory, 10, 203-218.

Hodgins, D., & Makarchuk, K. (2003). Trusting problem gamblers reliability and validity of self-reporting gambling behaviour. Psychology of Addictive Behaviors, 17, 244-248.

208

XGAMBLE Report

Hollander, E., Pallanti, S., & Allen A. et al., (2005). Does sustained release lithium reduce impulsive gambling and affectively instability versus placebo in pathological gamblers with bipolar spectrum disorders. American Journal of Psychiatry, 162(1), 137-145.

Hunt, P. A. (1993). Relateline: An Evaluation of a Telephone Helpline Counselling Service for Marital Problems. British Journal of Guidance and Counselling, 21(3), 277-289.

Jackson, A. C. & Thomas, S. A. (2005). Clients’ perspective of, and experiences with selected Australian selected problem gambling services. Journal of Gambling Issues. Retrieved October 22, 2007, from http://www.camh.net/egambling/issue14/pdf/jgi_14_jackson.pdf

Jackson, A. C., Thomas, S. A., Holt, T. A., & Thomason, N. (2005). Change and continuity in a help-seeking problem gambling population: A five-year record. Journal of Gambling Issues. Retrieved March 22, 2007, from http://www.camh.net/egambling/archive/pdf/JGI-issue13/JGI-Issue13-jackson.pdf

Jackson, A., Thomas, S., & Blaszczynski, A. (2003). Best practice in problem gambling services. Melbourne: Gambling Research Panel.

Jackson, A.C., Holt, T.A., Thomas, S.A., Crisp, B.R. (2003) ‘Development of an Instrument for the Analysis of Problem Gambling Counselling Practice’, International Gambling Studies, 3(1), 67-87.

Korn, D. A., & Shaffer, H. J. (2004). Massachusetts Department of Public Health’s practice guidelines for treating gambling-related problems: An evidence-based treatment guide for clinicians. Massachusetts Council on Compulsive Gambling.

Ladouceur, R., & Walker, M. (1996). A cognitive perspective on gambling. In P. M. Salkovskis (Ed.), Trends in cognitive and behavioural therapies (pp.89-120). New York: Wiley.

Ladouceur, R., Sylvain, C., Boutin, C., Doucet, C. (2002). Understanding and treating the pathological gambler. West Susssex, England: John Wiley & Sons, Ltd.

Ladouceur, R., Sylvain, C., Boutin, C., Lachance, S., Doucet, C., Leblond, J., & Jacques, C. (2001). Cognitive treatment of pathological gambling. Journal of Nervous and Mental Disease, 189, 773-780.

Ledgerwood, D. M. & Petry, N. M. (2005). Current trends and future directions in the study of psychosocial treatments for pathological gambling. Current Directions in Psychological Science, 14(2), 89-94.

Legge, K. (1996). Jackpot society. In M. Cathcart & K. Darrian-Smith (Eds.), Place your bets: gambling in Victoria. Parkville: The Australian Centre, The University of Melbourne.

Leitner, M. J., & Leitner, S. F. (2005). The use of leisure counselling as a therapeutic technique. British Journal of Guidance and Counselling, 33(1), 37-49.

209

XGAMBLE Report

Lesieur, H. R., & Blume, S. B. (1987). The South Oaks Gambling Screen (SOGS): A new instrument for the identification of pathological gamblers. American Journal of Psychiatry, 144, 1184-1188.

Lesieur, H. R., & Custer, R. L. (1984). Pathological gambling: Roots, phases and treatment. Annals of the American Academy of Political and Social Sciences, 474, 146-156.

Lesieur, H. R., & Rosenthal, R. J. (1991). Pathological gambling: A review of the literature. Journal of Gambling Studies, 7, 5-39.

Litt, M. D., Kadden, R. M., Cooney, N. L., & Kabela, E. (2003). Coping skills and treatment outcomes in cognitive-behavioral and interactional group therapy for alcoholism. Journal of Consulting Clinical Psychology, 71, 118-129.

Lopez Viets, V. C., & Miller, W. R. (1997). Treatment approaches for pathological gamblers. Clinical Psychology Review, 17(7), 689-702.

Lorenz, V. C., & Yaffe, R. A. (1988). Pathological gambling: Psychosomatic, emotional and marital difficulties as reported by the spouse. Journal of Gambling Behavior, 4, 13–26.

Marotta, J. (2005). What is the role of problem gambling helpline services? Journal of Gambling Issues. Retrieved October 23, 2007, from http://www.camh.net/egambling/pdf/jgi_15_marotta.pdf

McCown, W. G., & Howatt, W. A. (2007). Treating gambling problems. Hoboken, New Jersey: John Wiley & Sons.

Milton, S., Crino, R., Hunt, C., & Prosser, E. (2002). The effect of compliance-improving interventions on the cognitive-behavioural treatment of pathological gambling. Journal of Gambling Studies, 18(2), 207-229.

Ministry of Health (2004). Preventing and minimising gambling harm: strategic plan 2004-2010, needs assessment, proposed three-year funding plan, proposed problem gambling levy rates. Wellington, Ministry of Health.

Ministry of Health (2005). Problem Gambling Intervention Services in New Zealand: 2004 national statistics. Wellington, Ministry of Health.

Ministry of Health (2006). Problem gambling intervention services in New Zealand: 2006 service-user statistics. Public Health Intelligence, Monitoring Report, No. 14. Wellington: Author.

Morasco, B. J., & Petry, N. M. (2006). Gambling problems and health functioning in individuals receiving disability. Disability and Rehabilitation, 28(10), 619-623.

Morrison, L. (2005). Unfolding and dealing with the effects on partners and whanau members of Maori women’s gambling. International Journal of Mental Health and Addiction. Retrieved November 1, 2007, from http://www.ijma-journal.com/content/proc/abstracts_presentations/4/00023

210

XGAMBLE Report

Moskowitz, J. A. (1980). Lithium and lady luck – Use of Lithium-Carbonate in compulsive gambling. New York State Journal of Medicine, 80(5), 785-788.

Najavits, I. M., & Weiss, R. D. (1994). Variations in therapist effectiveness in the treatment of patients with substance use disorders: An empirical reviews. Addiction, 89, 679-688.

National Research Council (1999). Pathological gambling: A critical review. Washington, DC: National Academy Press.

Neighbors C., Lostutter, T. W., Larimer, M. E., Takushi, R. Y. (2002). Measuring gambling outcomes among college students. Journal of Gambling studies, 18(4), 339-360.

Newscan (November 29, 2002). Asian immigrants, elderly face gambling problems says Christchurch Casino chief, 4(46). Ontario: Responsible Gambling Council.

Obeymayer, J. L., Riley, W. T., & Jersino, J. (2004). College smoking-cessation using cell phone text messaging. Journal of American College Health, 53(2), 71-80.

Ogborne, A. C., Wild, C., Braun, K., Newton-Taylor, B. (1998). Measuring treatment process beliefs among staff of specialised addiction treatment services. Journal of Substance Abuse Treatment, 15(4), 301-312.

Parke, J., Griffiths, M., Parke, A. (2007). Positive thinking among slot machine gamblers: A case of maladaptive coping? International Journal of Mental Health and Addiction, 5, 39-52.

Paton-Simpson, G. P., Gruys, M. A., & Hannifin, J. B. (2002). Problem gambling counselling in New Zealand. 2001 National statistics. Palmerston North. Problem Gambling Purchasing Agency.

Patton, M. Q. (1990). Qualitative evaluation and research methods. Newbury Park, CA: SAGE.

Peele, S. (2001). Is gambling an addiction like drug and alcohol addiction? Developing realistic and useful conceptions of compulsive gambling. Electronic Journal of Gambling Issues: eGambling. Retrieved June 24, 2002, http://www.camh.net/egambling/.

Perese, L., & Faleafa, M. (2000). The impact of gambling on some Samoan peoples’ lives in Auckland. Auckland: Compulsive Gambling Society of New Zealand.

Peters, J. (2007). “We need to talk”: Talking therapies, a snapshot of issues and activities across mental health and addiction services in New Zealand. Auckland: Te Pou, The national Centre for Mental Health Research, Information and Workforce Development.

Petry, N. M. (2006). Pathological gambling: Etiology, comorbidity, and treatment. Washington, DC: American Psychological Association.

211

XGAMBLE Report

Petry, N. M., Armentano, C., Kuoch, T., Norinth, T., & Smith, L. (2003). Gambling participation and problems among South East Asian refugees to the United States. Psychiatric Services, 54, 1142–1148.

Petry, N. M., Litt, M. D., Kadden, R. M, & Ledgerwood, D. M. (2007). Do coping skills mediate the relationship between cognitive-behavioral therapy and reductions in gambling in pathological gamblers? Addiction, 102, 1280-1291.

Potenza, M. N., Steinberg, M. A., Wu, R., Rounsaville, B. J., O’Mallery, S. S. (2006). Characteristics of older adult problem gamblers calling a gambling helpline. Journal of Gambling Studies, 22 (2), 241-254.

Prochaska, J. O., & DiClemente, C. C. (1986). Toward a comprehensive model of change. In W. R. Miller & N. Heather (Eds.), Treating addictive behaviors: Processes of change (pp. 3-27). New York: Plenum Press.

Pulford, J., Adams, P., & Sheridan, J. (2006). Unilateral treatment exit: A failure of retention or a failure of treatment fit? Substance Use and Misuse, 41, 1901-1920.

Pulford, J., Adams, P., & Sheridan, J. (2007). Client/ clinician discrepancies in perceived problem improvement and potential influence on dropout response. International Journal of Mental Health and Addiction. Retrieved November 11, 2007, http://www.springerlink.com/content/9110nwun8w5mq428/.

Raghunathan, R. & Pham, M.T. (1999). All negative moods are not equal: motivational influences of anxiety and sadness on decision making. Organizational Behavior and Human Decision Processes, 79 (1), 56-77.

Rankine, J., & Haign, D. (2003). Social impacts of gambling in Manukau City. Report prepared for Manukau City Council.

Raylu, N., & Oei, T. P. (2004). Role of culture in gambling and problem gambling. Clinical Psychology Review, 23, 1087-1114.

Riley, W. T., & Jean-Mary, J. (2004). College smoking-cessation using cell phone text messaging. Journal of American College Health. 53(2), 71-78.

Robertson, P. Pitmana, S., Huriwai, T., Ahuriri-Driscoll, A., Haitana, T., Larsen, J., & Uta’I, S. (2005). Developing services in Te Rohe o Ngai Tahu for Maori with Gambling Related Problems. New Zealand Journal of Psychology, 34(1), 35-40.

Rodgers, A., Corbett, T., Bramley, D., Riddell, T., Wills, M., Line, R. B., & Jones, M. (2005). Do u smoke after txt? Results of a randomised trial of smoking cessation using mobile phone text messaging, Tobacco Control, 14(4), 255-261.

Rollnick, S., Heather, N., Gold, R., & Hall, W. (1992). Development of a short “readiness to change” questionnaire for use in brief, opportunistic interventions among excessive drinkers. British Journal of Addiction, 87, 743-754.

Shepherd, P. (1987). Telephone therapy: an alternative to isolation. Clinical Social Work Journal, 15 (1), 56-65.

212

XGAMBLE Report

Stewart, S.H. &. Kushner, M.G. (2003). Recent research on the comorbidity of alcoholism and pathological gambling. Alcoholism: Clinical and Experimental Research, 27 (2), 285-291.

Stinchfield, R., Govoni, R., & Frisch, G. R. (2001). An evaluation of diagnostic criteria for pathological gambling. Prepared for the Ministry of Health and Long term Care, Substance Abuse Bureau, Canada.

Strauss, A., & Corbin, J. (1990). Basics of qualitative research: Grounded theory procedures and techniques. Newbury Park, CA: Sage Publication Ltd.

Sylvian, C., Ladouceur, R., Bosivert, J. M. (1997). Cognitive and behavioral treatment of pathological gambling: A controlled study. Journal of Consulting and Clinical Psychology, 65, 727-732.

Tan, R., & Tam, S. (2003). Survey of problem gambling amongst restaurant workers in Christchurch (Unpublished report). Christchurch, New Zealand: South Island Services, Problem Gambling Foundation of New Zealand.

Tan, R., & Wurtzburg, S. (Eds.), Problem gambling: New Zealand perspectives on treatment. Wellington: Steele Roberts Ltd.

Tang, C. SK., Wu, A. MS., & Tang, J. YC. (2007). Gender differences in characteristics of Chinese treatment-seeking problem gamblers. Journal of Gambling Studies, 23, 145-156.

The Werry Centre for Child and Adolescent Mental Health (2005). New Zealand population projection statistics (2006, 2011, 2016). University of Auckland: Author.

Thomas, D. R. (2006). A general inductive approach for analyzing qualitative evaluation data. American Journal of Evaluation, 27(2), 237-246.

Thomas, S. (1995). ‘More than a flutter’: Women and problem gambling. In J. O’Connor (Ed.), High stakes in the nineties. Perth: Curtin University.

Toce-Gerstein, M., Gerstein, D. R., & Volberg, R. A. (2003). A hierarchy of gambling disorders in the community. Addiction, 98, 1661-1672.

Toneatto, T. & Ladouceur, R. (2003). Treatment of pathological gambling: A critical review of the literature. Psychology of Addictive Behaviors, 17(4), 284-292.

Toneatto, T., & Millar, G. (2004). Assessing and treating problem gambling: Empirical status and promising trends. Canadian Journal of Psychiatry, 49(8), 517-525.

Tse, S., Abbott, M., Clarke, D., Townsend, S., Kingi, P., & Manaia, W. (2005). Examining the determinants of problem gambling. Prepared for Health Research Council of New Zealand. Auckland UniServices Limited, University of Auckland, 210pp.

Tse, S., Wong, J., & Kim, H. (2004) A public health approach for Asian people with problem gambling in foreign countries. Journal of Gambling Issues, 12. Retrieved January 20, 2005, from http://www.camh.net/egambling/issue12/jgi_12_tse.html

213

XGAMBLE Report

Varah, C. (1973). The Samaritans in the 70’s. London: Constable.

Walker, M. (2005). Problems in measuring the effectiveness of cognitive therapy for pathological gambling. Journal of Gambling Studies, 21(1), 81-92.

Walters, G. (1994). The gambling lifestyle: II. Treatment. Journal of Gambling Studies, 10(3), 219-235.

Wells, K. B. (1999). Treatment research at the crossroads: The scientific interface of clinical trials and effectiveness research. American journal of Psychiatry, 156, 5-10.

Westphal, J. R. (2006). Are the effects of gambling treatment overestimated? International Journal of Mental Health and Addiction. Retrieved October 30, 2007, from http://www.springerlink.com/content/w66r4nn418m0w234/

Wong, J., & Tse, S. (2003). The face of Chinese migrants’ gambling: A New Zealand perspective. eGambling: Electronic Journal of Gambling Issues, issue 9. Retrieved November 3, 2003, from thttp://www.camh.net/egambling/issue9/opinion/wong/

214