Outpatient versus inpatient opioid detoxification: A randomized controlled trial

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

1

Outpatient Versus Inpatient Opioid Detoxification: A Randomised Controlled Trial

*Dr Ed Day, Senior Lecturer in Addiction Psychiatry, University of Birmingham, The Barberry, 25

Vincent Drive, Edgbaston, Birmingham B15 2FG. Tel: +44 121 301 2355, Fax: +44 121 301 2351,

e-mail: [email protected]

Professor John Strang, Director, National Addiction Centre, Institute of Psychiatry, 4 Windsor

Walk, Denmark Hill, London SE5 8AF

*Author for correspondence

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

2

Abstract

Opioid detoxification is not an effective standalone treatment for heroin dependence, but is

nevertheless an essential step on the path to recovery. There has been relatively little previous

controlled research on the impact of treatment setting on likelihood of successful completion of

detoxification. In this study 68 opioid dependent patients receiving community treatment

(predominantly with methadone) and requesting detoxification were randomly assigned to an

inpatient versus outpatient setting. Both groups received the same medication (lofexidine), and the

primary outcome measure was being opioid-free at detoxification completion. More inpatients (18,

51.4%) completed detoxification than outpatients (12, 36.4%), but this difference was not

statistically significant (2 = 1.56, p=0.21). However, the outpatient group received a significantly

longer period of medication, and when length of detoxification was controlled for the results

favoured the inpatient setting (Exp(B)=13.9, C.I. 2.6-75.5, p=0.002). Only 11 (16%) participants

were opioid-free at one-month follow-up, and 8 at six-month follow-up, with no between-group

difference. Inpatient and outpatient opioid detoxification were not significantly different in

completion or follow-up abstinence rates, but aspects of the study design may have favoured the

outpatient setting. Future studies should test patient characteristics that predict better outcomes in

each setting.

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

3

Introduction

Heroin addiction is a serious medical and social problem in the UK (HM Government, 2008).

Although medically assisted ‘detoxification’ is considered to have limited effectiveness as a stand-

alone treatment, it often acts as the bridge between maintenance treatment and abstinence (National

Collaborating Centre for Mental Health, 2007). There are several possible benefits of inpatient

admission for detoxification (Gossop, 2003; Kleber, 1999; Mattick & Hall, 1996; Weiss, 1999). A

hospital setting permits a high level of medical supervision and safety for individuals who require

intensive physical and/or psychiatric monitoring, and the greater intensity of treatment may also

help patients who do not respond to lesser measures. Inpatient treatment can help to interrupt a

cycle of drug use even in the absence of medically dangerous withdrawal symptoms. For some, the

safety of an inpatient environment and a period of respite from people and places associated with

drug use can help them in their attempt to make important life decisions. Withdrawal can often be

completed more quickly in an inpatient environment, but if the programme is a comprehensive one

more attention can also be paid to family, vocational, medical and psychiatric issues. However, the

protectiveness of an inpatient unit may also be one of its main disadvantages, as a major

determinant of craving is drug availability. Detoxification in an outpatient setting requires the

patient to cope with everyday situations that will be encountered on discharge and may promote

better coping skills. Inpatients are unable to work, care for their families, study or conduct their

normal daily activities. The stigmatization of some inpatient service settings may also be a deterrent

to some patients (Strang et al., 1997; Wilson, Elms, & Thomson, 1975).

Detoxification in an inpatient setting is up to twenty times more expensive than the equivalent

treatment in a community setting (Gossop & Strang, 2000), and the vast majority (90%) of opioid

users in the UK who want to become abstinent are offered community detoxification as the first-

line treatment (National Collaborating Centre for Mental Health, 2007). A national survey of the

provision of inpatient and residential treatment in England reported an estimate of 10,711 annual

admissions for drug detoxification in 2003/4, with approximately 60% of the activity occurring in

specialist inpatient units (Day, Ison, Keaney, Buntwal, & Strang, 2005). However, the geographical

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

4

distribution of these units was variable, with some areas of the country having limited or no access

to such a facility. Inpatient detoxification is usually only offered after community treatment has

repeatedly failed (SCAN Inpatient Treatment Working Party, 2006), and it is common practice to

offer inpatient detoxification to the service users with the most complex needs (SCAN, 2006).

Multiple publications have reported treatment outcomes for opioid detoxification conducted in both

an inpatient and an outpatient setting and using a variety of pharmacological treatment strategies

(see Day 2005 for a review (Day, 2005)). The reported rate of successful completion of opioid

detoxification varies between 4 and 100%, with higher rates in studies conducted in an inpatient

setting. Non-randomised studies comparing inpatients and outpatients typically show that the

former have more severe substance use histories and a greater prevalence of medical, psychosocial

and vocational difficulties, including less social stability, more unemployment and a greater

preponderance of medical and psychiatric disorders (Gossop, Marsden, Stewart, & Rolfe, 1999;

Weiss, 1999). Surprisingly, however, there have been few attempts to compare outcomes of opioid

detoxification between inpatient and outpatient settings using a randomised controlled design. Only

two RCTs comparing outcomes between inpatient (with 24-hour availability of medical and nursing

staff) and outpatient or community (with intermittent support from clinical staff during office hours

only) detoxification processes for opioid addicts have been published (Day, Ison, & Strang, 2005).

Both reported significantly higher rates of detoxification completion in inpatient settings, but both

had serious methodological limitations. A key problem was that both studies used a different

medication regimen in each clinical setting, thus missing the opportunity to study the impact of the

setting on the likelihood of success (Gossop, Johns, & Green, 1986; Wilson et al., 1975).

The study reported in this paper took place in a specialist clinical addiction treatment service and

compared the same medical treatment regimen (using the alpha-2 adrenergic agonist lofexidine, the

only non-opioid medication licensed for opioid detoxification in the UK at the time of the study

(Strang, Bearn, & Gossop, 1999)) administered for the same period of time in both an inpatient and

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

5

outpatient setting. The service treated a mixture of primary opioid users and more complex poly-

drug users, including both those that had relapsed after a previous detoxification and previously

untreated cases of heroin dependence, and so appeared to be representative of many treatment

services in England at the time (Department of Health, 1996; National Treatment Agency for

Substance Misuse, 2005).

Materials and Methods

The OutPatient versus Inpatient OpIoid Detoxification (OPIOiD) study was conceived and

conducted as a randomised controlled trial to test whether there is a difference in the number of

opioid dependent patients who achieve abstinence from opioids (heroin and methadone) at the end

of the medicated detoxification period using a lofexidine assisted withdrawal process in an inpatient

setting compared with an outpatient setting. Secondary areas of interest were the influence of

treatment setting on levels of patient satisfaction, and also rates of uptake of aftercare or relapse to

regular heroin use after detoxification.

Setting

At the time of the study the Addictive Behaviours Centre (ABC) provided both inpatient and

outpatient drug and alcohol treatment to approximately 1 million people in Birmingham, focusing

on medically supported treatment to patients considered too complex for community services. The

inpatient service had a capacity of 12 beds (6 each for alcohol and drug problems) and was staffed

by nurses, an Occupational Therapy assistant and a psychology assistant. There was input from a

Consultant addiction psychiatrist and junior psychiatrists in training during office hours, with cover

from doctors on the general psychiatry rota at other times. The outpatient service was an assessment

and treatment program delivered through medical outpatient clinics, with two clinical nurse

specialists providing support for opioid detoxification at home. Both nurses had more than ten years

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

6

experience of working in the addiction treatment field, and had been trained in providing opioid

detoxification using lofexidine.

Recruitment

Patients were recruited into the study between April 2001 and May 2003, with the final follow-up

interview conducted in January 2004. They were initially assessed in the ABC outpatient clinics by

one of the medical staff. Most patients seen in these clinics were under the care of the ABC, but

four community drug treatment teams could refer patients for assessment for opioid detoxification.

All potential participants were therefore receiving some form of community treatment prior to

referral, predominantly with methadone. Where relevant each clinician completed a screening form

incorporating the inclusion and exclusion criteria for the study. All patients meeting the entry

criteria were given an information leaflet and were invited to attend an assessment interview with

the researcher within one week of the screening appointment. There were three inclusion criteria:

meeting ICD-10 criteria for opioid dependence (World Health Organisation, 1992), requesting a

medically-assisted opioid detoxification, and consenting to randomisation and to receive treatment

and monitoring within the study. Patients were excluded if they were homeless or unable to identify

any source of support, currently pregnant, dependent on alcohol or psychostimulants as defined by

ICD-10, had a history of coronary artery disease or cardiac arrhythmias, had current symptoms of

psychosis or severe affective disorder, or were unwilling to consider one of the treatment settings.

Each person assessed was given written information about the rationale for the study and the

proposed procedures by the researcher, and all those agreeing to participate signed a consent form.

All interviews were conducted by the researcher who was not blind to the treatment condition, but

was independent of provision of treatment to participants in either arm of the study. On entering the

study, each participant was asked to give the names and contact details of others who may know

their whereabouts (e.g. friends or family members) at the time of follow-up, and to confirm that

they would be agreeable to the researcher contacting these individuals if the participant themselves

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

7

could not be reached. The study was approved by the North Birmingham Local Research Ethics

Committee (LREC 01/01/203).

Intervention

Participants in both treatment settings received the same medication regime, which was the routine

structured lofexidine detoxification used in the service at the time (outlined in Table 1). The same

clinical protocol was used by both inpatient and outpatient staff, with participants in the outpatient

group having their medication brought to their home by the nurse on each day during the working

week, with weekend doses supplied on a Friday. Both inpatient and outpatient staff were given

some flexibility in administering the protocol. Clinicians could increase the period of full dose

lofexidine by up to 7 days if clinically indicated, but after 19 days of full-dose medication the

detoxification was considered to have ended and the outcome was assessed on day 23. Other

medication was also available as necessary, including ibuprofen (up to 400mg 8-hourly for muscle

aches), Lomotil (diphenoxylate & atropine, up to 2 tablets 6-hourly for diarrhoea), zopiclone (up to

15mg) or nitrazepam (up to 10mg at night for insomnia), and diazepam (up to 5mg 8-hourly for

agitation).

[Insert table 1 about here]

Both inpatient and outpatient groups received at least 1 hour of individual face-to-face interaction

with a nurse each day from Monday to Friday. At least 10 days after the last dose of opioids, the

participant was offered a ‘naloxone challenge’ test to determine whether it was appropriate to

commence a regular prescription of oral naltrexone. This was conducted on the inpatient unit

(outpatients were brought to the unit by the community nurse), where an intravenous cannula was

inserted and a dose of 400 micrograms of naloxone was administered. If the participant experienced

mild or tolerable opioid withdrawal symptoms they were commenced on a dose of naltrexone

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

8

50mg/day on the next day. If they experienced more severe symptoms, the challenge test was

repeated in 2 days time.

Both inpatient and outpatient nurses had been trained in providing information about opioid

withdrawal, supportive counselling techniques and the principles of relapse prevention (Marlatt &

Donovan, 2005), and all participants could have auricular acupuncture and relaxation training on

request. All participants were encouraged to discuss their plans for on-going treatment with their

nurse, and an appointment with a doctor at the ABC outpatient service was made routinely within

the first two weeks post-detoxification. Both groups were also provided with information about

residential rehabilitation services and were assessed for suitability for funding of such a placement

on request. At various points during the 2 year duration of the trial other non-residential aftercare

was also available, including a weekly open relapse prevention group run over 6 sessions by the OT

at the inpatient unit, and a day care programme run by a non-statutory agency. All aftercare options

were equally available to both treatment groups.

The two treatment conditions differed in the availability and intensity of additional interventions

during the detoxification period. The ‘standard’ inpatient admission package involved a three-week

stay, and in addition to the daily individual session with a nurse, the participant could attend a daily

(Monday to Friday) 2-hour group programme delivered by the nursing staff and the Occupational

Therapy assistant. This included a combination of educational lectures or discussions about drugs

and preventing a return to daily use post-detoxification, and other activities such as art and pottery.

This was not available to the outpatient group, but during each home visit the nursing staff

encouraged family members and friends to become involved in the detoxification process wherever

possible, particularly focusing on developing social support for non-drug using activities.

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

9

Outcome measures

Each participant was interviewed at the point of randomisation, and 1-month and 6-months after

detoxification completion. The primary outcome measure was successful completion of the

medicated detoxification process, measured by 1) self-report and 2) urinary drug screening on the

day after the last dose of lofexidine. Participants dropping out of treatment during the detoxification

process were considered to have been unsuccessful in their attempt to become abstinent. Secondary

outcome measures were: 1) patient satisfaction with the detoxification process, measured at the 1-

month follow-up point, 2) uptake of aftercare and further treatment reported at the 1- and 6-month

follow-up points, 3) use of opioids at 1 month and 6 months after completion of detoxification.

The participant was asked at baseline about previous experience of medically assisted withdrawal

from opioids, past admissions to psychiatric hospital, past medical history, and potential social

supports to help with withdrawal. The level of drug use and related problems was measured at

baseline and follow-up using the Maudsley Addiction Profile (MAP). This is a brief, interviewer-

administered questionnaire for treatment outcome research (Marsden et al., 1998) that covers the

30-day period before interview and includes problems in four broad domains: substance use, health

risk behaviour, health symptoms and personal/social functioning. A urine sample was taken

randomly at least once per week in both treatment conditions, on the day after the last dose of

lofexidine, and at both follow-up points. It was sent to the local National Health Service toxicology

laboratory for both screening and confirmatory testing (using gas chromatography-mass

spectroscopy (GCMS)) for opioids, cocaine, amphetamines, benzodiazepines and cannabis.

Satisfaction with the treatment process was measured using the Treatment Perception Questionnaire

(TPQ), which examines the patient’s views towards: 1) the nature and extent of their contact with a

treatment programme's staff team (5 items); and 2) aspects of the operation of the treatment service

and its rules and regulations (5 items), giving a total score of between 0 and 40 (Marsden et al.,

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

10

2000; Marsden et al., 2000). A semi-structured interview at both follow-up points was used to

assess the type and intensity of aftercare treatment received by each participant.

Randomisation

Participants were randomised by an independent telephone randomisation service at the University

of Birmingham Clinical Trials Unit. After confirming eligibility and consent, participants were

allocated to either inpatient or outpatient detoxification in a 1:1 ratio. Randomisation was stratified

to ensure that equal numbers of people undertaking their first medically assisted withdrawal

procedure and those who had previously attempted such treatment appeared in each treatment arm.

The randomisation list was prepared using variable block sizes. Allocation was concealed until the

participant was committed to the trial and the investigator was unaware of the block sizes, to

prevent foreknowledge of the treatment.

Power calculation

In a review of the literature, Lipton and Maranda found detoxification completion rates of between

50 and 77% for inpatient detoxification, and of about 20% for outpatient detoxification (Lipton &

Maranda, 1983). Later studies in the UK also showed that approximately 20% of outpatients

achieve abstinence from opioids (Dawe, Griffiths, Gossop, & Strang, 1991; Gossop et al., 1986),

compared with initial abstinence rates of 80-85% for inpatient detoxification (Gossop et al., 1986;

Gossop & Strang, 1991). However, the development of the alpha-2 adrenergic agonists has

improved the rate of successful detoxification in community settings e.g. Kleber et al found

equivalent rates of successful detoxification of approximately 40% between a methadone reduction

regime and a clonidine substitution technique (Kleber et al., 1985). Therefore, assuming an 80%

success rate in the inpatient group and a 40% success rate in the outpatient group, a two-group

continuity corrected X2 test with a 0.05 two-sided significance level would have 90% power to

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

11

detect the difference between the two groups when the sample size was 70 (1:1 allocation, 35 in

each group).

Data Analysis

All data were analysed using SPSS version 14.0. Differences in continuous variables between the

two treatment groups were compared using independent t-tests for normally distributed continuous

variables, Mann-Whitney U tests for non-normally distributed continuous variables, and chi-square

tests for categorical variables. Initial assessment and follow-up comparisons for patients were

analysed using paired data sets. Changes for categorical data were assessed with the McNemar test,

and repeated measures t-tests (for normally distributed data) or Wilcoxon signed-rank tests (for

non-normally distributed data) were used to assess within group changes in the target behaviours.

Results

The results section will begin by describing the study population, before comparing the levels of

detoxification completion between the two treatment settings. Secondary outcome measures will

then be considered, including the influence of treatment setting on levels of patient satisfaction,

rates of uptake of aftercare, and relapse to regular heroin use after detoxification.

Study Participants

The baseline demographics, drug use and related problems for the whole sample are shown in table

2. All participants fulfilled ICD-10 criteria for opioid dependence, and urinary drug screening

corroborated opioid use. Three quarters (53, 77.9%) were being prescribed methadone at a mean

daily dose of 33mg (range 15 - 60), but only 27 of these (50.9%) had taken methadone every day in

the preceding month. Use of more than one substance was common, and 28 (41.2%) had used one

illicit substance in addition to heroin, 20 (29.4%) had used two, 8 (11.8%) had used three, and 1

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

12

(1.5%) had used four. The participants had first started using heroin at a mean age of 21.7 years

(range 14-45 years), and the mean length of heroin-using career was 6.1 years (range 1-23). Just

over half (38, 55.9%) had made a previous attempt at a medically assisted detoxification.

The progress of participants through the study is shown in figure 1. The ABC was a secondary

referral centre, and all potential participants were receiving some form of treatment prior to referral

for detoxification. The majority were existing patients of the ABC outpatient methadone

programme, but some (approximately 15%) were referred by other community drug treatment

teams. A total of 168 patients screened at the ABC between 2001 and 2003 requested a

detoxification (113 (67.3%) men and 55 (32.7%) women). Fifty (26.9%) refused study entry, 37

(22%) had no source of social support, 4 (2.3%) were pregnant, 16 (9.5%) met ICD-10 criteria for

psychostimulant dependence, 3 (1.8%) had a history of cardiac problems, and 8 (4.8%) had serious

active mental illness, resulting in a potentially available study population of 79 subjects. A further

11 patients failed to attend the meeting with the researcher, leaving a final sample size of 68.

[Insert figure 1 about here]

Due to the limited availability of the inpatient resource, patients at the ABC would routinely wait

between 4 and 12 weeks before commencing their detoxification programme, but would continue to

receive their exisiting community treatment during this period. The mean time between

randomisation and starting the detoxification was 34 days (range 2 to 114 days), with no statistical

difference in the two study groups (inpatient mean=33 days, range=7-114 days, outpatient mean=36

days, range=2-113 days, U=549, p=0.73). A comparison between self-reported measures of drug

use (number of days of heroin use and amount per using day) taken at randomisation and at the start

of the detoxification process showed no significant differences. Follow-up was scheduled for both 1

and 6 calendar months after the day that the medically assisted detoxification was completed. The

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

13

mean length of time between the end of detoxification and the 1-month and 6-month interviews was

33 days and 195 days respectively, with no statistical difference between treatment groups.

[Insert table 2 about here]

Completion of detoxification

Eighteen (51.4%) of the inpatient group completed the medicated detoxification successfully,

compared with 12 (36.4%) of the outpatient group (2 = 1.56, p=0.21). Treatment in an inpatient

setting was associated with an increase in the odds of detoxification completion by a factor of 1.85

(95% C.I. = 0.7 – 4.89, p=0.21). Thirteen participants discharged themselves from the inpatient unit

before the end of the medication regime, and a further 4 (11.4%) were discharged by the staff (two

for smoking cannabis, one for smoking heroin and one for aggressive behaviour). The nurses made

a mean of 10.9 visits per participant in the outpatient group (range 2-19), and a mean of 23.7 phone

contacts (range 5-39). Although the ABC treatment protocol used during the study required the

clinician to terminate the detoxification if the patient tested positive for opioids, cocaine,

amphetamine or unprescribed benzodiazepines, there was no guidance for cannabis. This led to an

unanticipated difference in practice between the two groups, as the outpatient nurses routinely

ignored urine tests that were positive for cannabis, whereas the inpatient staff adopted a strict zero

tolerance approach to all illicit drugs. Four of the participants that completed the outpatient

detoxification successfully tested positive for cannabis, and if these cases are reclassified as

detoxification failures the result significantly favours the inpatient setting (18 of 35 (51.4%) vs. 8 of

33 (24.2%), 2 = 5.3, p=0.02).

Furthermore, the outpatient group had a significantly longer mean medicated period (17.9 days)

than the inpatient group (11.2 days, t=4.96, d.f.=66, p<0.001), again linked to the greater flexibility

applied by the outpatient staff. For example, two of the outpatient group had full-dose lofexidine for

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

14

more than the 19 days (i.e. including the extra 7 days) allowed by the treatment protocol, but neither

successfully completed the detoxification process. When logistic regression was used to control for

the length of medication period, inpatient treatment was associated with an increase in the odds of

detoxification completion of 13.9 (95% C.I. = 2.6 – 75.5, p=0.002).

There was no difference between the two groups in mean quantity of lofexidine (1.2mg) or

diazepam (13mg) used per detoxification day, but patients in the inpatient group received more

ibuprofen than those in the outpatient group (mean dose per treatment day = 486mg vs. 80mg,

p=0.05). However, participants in the inpatient group were supervised in taking all of their

medication whereas outpatient participants were not, and so these figures may not represent what

was actually taken. All patients who successfully completed the detoxification process in either

setting were offered naltrexone as a relapse prevention measure. Of the 30 successful completers,

22 (73.3%) started naltrexone (13 of 18 (72.2%) inpatients and 9 of 12 (75%) outpatients, 2 = 0.76,

p=0.38).

Satisfaction with the detoxification process

Overall, the outpatients rated their satisfaction with the treatment process significantly higher than

the inpatients (mean=27.7 vs. 23.5, t=2.44, d.f.=54, p=0.02). The outpatient group had higher mean

scores on both the ‘staff’ sub-scale (14.2 vs. 12.8, t=1.57, d.f.=51, p=0.12) and the ‘operation’ sub-

scale (13.5 vs. 10.7, t=2.88, d.f.=62, p=0.005). There was a significant correlation between the total

TPQ score and the length of medicated detoxification period (r=0.34, p=0.005), the mean total

lofexidine dose (r=0.32, p=0.01) and the mean total zopiclone dose (r=0.41, p=0.023). There was

no statistically significant correlation with the amount of diazepam or nitrazepam prescribed.

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

15

Post-detoxification care

At the 1-month follow-up point 11 (16.9%) participants were taking naltrexone (5 inpatients and 6

outpatients), and by the 6-month point this number had fallen to 2 (1 inpatient and 1 outpatient).

Although the aim of both ABC programmes was to achieve abstinence from opioids and to link

patients into aftercare to help them maintain this, resource constraints on the service meant that this

was not always possible. None of the participants had health insurance or sufficient resources to

fund a residential rehabilitation placement, and so were reliant on obtaining funding from the local

social services department. The prolonged assessment for funding was an obstacle to many

participants, and only 2 detoxification completers started a residential rehabilitation programme

post-detoxification (1 inpatient and 1 outpatient). A further three participants (all outpatients) were

unsuccessful in becoming opioid-free, but later obtained a place at a residential unit for further

detoxification and rehabilitation. Only 1 participant (outpatient) attended a local abstinence-based

day programme, but this service closed 6 months after the start of the trial. Five participants (all

inpatients) attended outpatient appointments at the ABC to receive naltrexone but received no other

aftercare. The majority (36, 52.9%) of participants returned to an opioid substitution programme

post-detoxification (21 of 35 inpatients and 15 of 33 outpatients), and 21 (30.9%) attended no

follow-up treatment at all in the 6 month follow-up period. In summary, 47 participants (69.1%)

attended any professional aftercare treatment post-detoxification, with no significant difference

between the treatment groups (27 of 35 (77.1%) inpatients vs. 20 of 33 (60.6%) outpatients, 2 =

2.18, p=0.14).

Opioid use at the 1-month follow-up point

Sixty five participants (95.6%) were interviewed about their heroin use, and only 3 urine test results

were incompatible with the self-report (1 reported heroin use but tested negative, and 2 reported no

heroin use but tested positive). If those not followed up were assumed to be using heroin, 13 out of

68 (19.1%) were heroin-free 1 month after the end of the detoxification (8 of 35 (22.9%) inpatients

vs. 5 of 33 (15.1%) outpatients, 2 = 0.65, p=0.42). Of these, 2 had been recommenced on opioid

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

16

substitution treatment, leaving 11 (16.2%) abstinent from all opioids (7 of 35 (20%) inpatients vs. 4

of 33 (12.1%) outpatients, 2 = 0.78, p=0.38).

Opioid use 6-month follow-up point

Sixty four participants (94.1%) were interviewed about their heroin use in the preceding 30 days.

Assuming participants not followed up were using heroin, 11 (16.2%) were abstinent (8 of 35

(22.9%) inpatients vs. 3 of 33 (9.1%) outpatient, 2= 2.37, p=0.12). Three of these individuals were

receiving opioid substitution treatment, so 8 (12.5%) were abstinent from all opioids (6 of 35

(17.1%) inpatients vs. 2 of 33 (6.1%) outpatients, 2= 2.01, p=0.16). Four of the 11 (36.4%)

participants who were abstinent from all opioids at 1 month were still abstinent at 6 months.

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

17

Discussion

In common with the two previous RCTs published in this area, the OPIOiD trial demonstrated that

the rate of completion of detoxification is higher in an inpatient setting than in the community, but

in this study the difference between the two groups was not statistically significant. However, these

results must be interpreted with a certain amount of caution. It is likely that the relative advantage

of inpatient treatment over outpatient treatment was overestimated in the power calculation for this

study (an absolute difference of 40%), and so the likelihood of a type II error is high. When Finney

et al reviewed outcome studies comparing inpatient and outpatient settings for treatment in the

alcohol field, a higher power to detect (especially medium-sized) setting effects was one factor

differentiating those studies yielding significant results from those showing no effects of treatment

setting (Finney, Hahn, & Moos, 1996). It is possible that inpatient detoxification does produce

better outcomes in terms of completion of detoxification, but this study failed to detect them.

Rates of completion of opioid detoxification in this study were lower than in others published in the

UK and the USA (Day, 2005). This may have been because other studies have reported the results

of controlled comparisons of different methods of detoxification, and so utilized samples selected

for their suitability for treatment. Rates of completion of detoxification are generally lower in

reports from non-research settings (Sheridan, Cook, & Strang, 1999). The majority of the

participants in this study were receiving methadone treatment prior to starting detoxification, and

only 6 were abstinent from heroin despite opioid substitution. During the period in which the study

took place the inpatient unit experienced problems with a shortage of suitably qualified staff, and

this may have led to a dilution of the possible benefits of this setting. However, staffing problems

were on average no more common in the Birmingham unit than in any other unit in England at the

time (Day, Ison, Keaney et al., 2005).

Participants in the outpatient setting also received an individually tailored ‘home’ detoxification

with visits by experienced nursing staff over a 3-4 week period. This far exceeds the outpatient

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

18

detoxification in other studies, where participants were expected to attend a clinic to receive

medication and support. Another factor that reduced any potential advantage of the inpatient setting

was a different response to cannabis use between the two treatment groups. If the outpatient

detoxification process had been subject to the same strict criteria around cannabis as the inpatient

process, four of the successful participants may have failed, and the results would have favoured the

inpatient setting. Likewise, the outpatient treatment staff were more inclined to extend the period of

medication than those in the inpatient unit, where a 3 week admission period was usually adhered

to. As the primary outcome measure was opioid abstinence on the day after the last dose of

medication, this may have had the effect of improving completion rates in the outpatient setting.

When length of detoxification was controlled for, the odds ratio in favour of the inpatient setting

increased significantly. This difference in staff attitudes may have also contributed to the greater

levels of treatment satisfaction in the outpatient setting.

In most studies exploring treatment setting to date, patients with significant medical, psychiatric or

social problems or additional substance dependence disorders have been excluded. Paradoxically, it

is this very group of patients who may be most appropriate for inpatient care and may be most

likely to benefit from it (Miller & Hester, 1986), a phenomenon that Seivewright calls the ‘severity

paradox’ (Seivewright, 2000). Due to the concern about completing opioid detoxification in an

outpatient environment if there were severe mental health problems, co-dependence on other

substances or no social support at home, such patients were excluded from the OPIOiD trial. Of 186

patients screened as opioid dependent and requesting detoxification, 57 (31%) had at least one of

the exclusion criteria. Clinical practice suggests that outcomes are better in an inpatient setting for

this group, and so by excluding them the potential benefits of the inpatient setting are lost.

The results of the OPIOiD trial confirm previous research findings that detoxification is an

ineffective stand-alone treatment for opioid dependence (Simpson & Sells, 1990). Less than half of

the participants who began the detoxification process were opioid-free at the end of the medication

period, and only two of these managed to remain opioid-free for the following 6 months. When

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

19

Gossop et al followed up 80 opioid addicts after inpatient treatment in 1984-1986, they found that

half of the sample had lapsed to some opioid use by the end of the first week post detoxification,

with 81% relapsing at some point in the 6-month follow-up and 32% returning to daily

heroin/opioid use at 6 months (Gossop, Green, Phillips, & Bradley, 1989). In the OPIOiD trial 81%

had used heroin at least once in the 30 days after detoxification, and 28% had returned to daily

heroin use at 6 months. Three-quarters of the patients who were successful in their detoxification

attempt commenced naltrexone therapy, but only half were still taking the medication 1 month later,

and only 2 at the 6-month follow-up. This is consistent with previous studies of unsupervised

naltrexone prescribing as a relapse prevention measure (Adi et al., 2006).

Long-term drug treatment offers the most promising route to reduction in drug dependence, and

substantial numbers of addicted individuals begin their contact with the treatment system via

detoxification. Although even admissions for detoxification as short as 3 days can have

considerable benefits up to 6 months later (Chutuape, Jasinski, & Fingerhood, 2001), longer-term

treatment is required in order to achieve the best outcomes. Indeed, detoxification for opioid

dependence can also be problematic when not integrated into a comprehensive treatment system, as

the reduction in opioid tolerance that it produces can lead to accidental overdose and death (Davoli

et al., 2007; Strang et al., 2003). Detoxification is not a treatment in its own right, but rather the first

(and often necessary) step in a pathway to recovery. Patients who receive aftercare once they have

completed opioid detoxification have better outcomes in terms of drug abstinence (McCusker,

Bigelow, Luippold, Zorn, & Lewis, 1995) and re-admission rates (Daley, Salloum, Zuckoff, Kirisci,

& Thase, 1998) than those who do not enter aftercare. An extensive evaluation of the nationwide

Department of Veterans Affairs programme in the USA found that drug dependent patients who

entered a community residential facility (CRF) after discharge from inpatient care obtained more

outpatient mental health care and had lower readmission rates than comparable patients discharged

directly to the community (Moos, Pettit, & Gruber, 1995). Furthermore, patients with longer

episodes of CRF care had lower 6-month and 1-year readmission rates than patients who dropped

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

20

out of CRFs and than matched controls. Similar findings have come from the UK and Australia

(Ghodse et al., 2002; Sannibale et al., 2003).

In the OPIOiD trial, detoxification in the inpatient setting conferred no advantages over the

outpatient setting in terms of medium-term outcomes, including entry into residential aftercare.

However, there was a poor uptake of residential or non-residential rehabilitation treatment by the

whole sample, and these results partly reflect the nature of UK treatment services at the time of the

study. Funding for residential treatment was relatively scarce and mainly focused on a short-term

medically-managed episode rather than longer term residential rehabilitation. The process of

applying for such funding was lengthy and difficult to sequence at the correct phase of the client’s

care journey. Although the results did not reach the level of statistical significance, aftercare uptake

was higher in the inpatient group, and with greater power these results may have favoured the

inpatient group.

Previous research has shown that many patients fail to make the transition from inpatient treatment

to rehabilitative aftercare, with average rates of transition for patients receiving standard referral

instructions (i.e. given a list of services and encouraged to attend one of them) across six published

studies only 35% (Chutuape, Katz, & Stitzer, 2001). Familiarising the patient with the aftercare

programme has been found to enhance contact rates relative to standard referral procedures for both

alcohol and drug problems (Lash, 1998), as has addressing practical barriers to treatment entry such

as transport (Booth, Crowley, & Zhang, 1996). Patients who failed to complete the detoxification

process in the outpatient setting in the OPIOiD trial were more likely to have a second successful

attempt within the follow-up period, possibly reflecting the close one-to-one relationship formed

with the treating nurse, who continued to advocate for the patient even after the detoxification

ended.

The straightforward conclusion from this study is that a larger trial is needed to provide a definite

answer about the role of treatment setting. However, a complex range of variables affect the

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

21

likelihood of detoxification completion, whatever the setting. There has been interest in the study of

‘matching’ effects in which specific patient attributes interact with different types of treatment to

determine outcomes (Moos, 1997). In a naturalistic, multi-site sample of patients seeking treatment

for substance use disorders in the Department of Veteran’s Affairs treatment system in the USA,

Tiet et al found that the severity of substance use at baseline predicted follow-up substance use

severity and the setting that treatment took place in had no main effects. However, patients with

more severe problems experienced better alcohol and drug outcomes following inpatient/residential

treatment when compared with outpatient treatment, and those with lower baseline severity scores

showed the opposite pattern (Tiet, Ilgen, Byrnes, Harris, & Finney, 2007). Finney and others have

therefore suggested that the focus of research needs to move away from the main effects of

treatment and onto an exploration of whether intensive inpatient treatment may be differentially

beneficial for some sub-groups of patients e.g. those with psychiatric co-morbidity or lower levels

of social stability. Large multi-site naturalistic studies are ideal for identifying moderators of

treatment setting effects as they tackle the severity paradox effect. Designs using ‘real world’

settings contain more severe patients at baseline that would be excluded from randomized trials on

ethical grounds (Tiet et al., 2007), and may be the preferred research design option of the future.

Whichever design is chosen, the incorporation of an economic analysis is increasingly recognised

as important when expensive inpatient interventions are being considered.

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

22

Acknowledgements

The authors received no financial support for this study.

The findings were included in a systematic review of the evidence for opiate detoxification for drug

misuse conducted by the National Collaborating Centre for Mental Health in the UK (National

Collaborating Centre for Mental Health, 2007).

References

Adi, Y., Juarez-Garcia, A., Wang, D., Jowett, S., Frew, E., Day, E., et al. (2006). Oral Naltrexone

as a Treatment for Relapse Prevention in Formerly Opioid Dependent Drug Users: A

Systematic Review and Economic Evaluation. . Health Technology Assessment, 11(6), 1-85.

Booth, R. E., Crowley, T. J., & Zhang, Y. (1996). Substance abuse treatment entry, retention, and

effectiveness: out-of-treatment opiate injection drug users. Drug and Alcohol Dependence,

42, 11-20.

Chutuape, M. A., Jasinski, D. R., & Fingerhood, M. I. (2001). One-, three-, and six-month

outcomes after brief inpatient opioid detoxification. American Journal of Drug and Alcohol

Abuse, 27(1), 19-44.

Chutuape, M. A., Katz, E. C., & Stitzer, M. (2001). Methods for enhancing transition of substance

dependent patients from inpatient to outpatient treatment. Drug and Alcohol Dependence,

61, 137-143.

Daley, D. C., Salloum, I. M., Zuckoff, A., Kirisci, L., & Thase, M. E. (1998). Increasing treatment

adherence among outpatients with depression and cocaine dependence: results of a pilot

study. American Journal of Psychiatry, 155(11), 1611-1613.

Davoli, M., Bargagli, A. M., Perucci, C. A., Schifano, P., Belleudi, V., Hickman, M., et al. (2007).

Risk of fatal overdose during and after specialist drug treatment: the VEdeTTE study, a

national multi-site prospective cohort study Addiction, 102, 1954-1959.

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

23

Dawe, S., Griffiths, P., Gossop, M., & Strang, J. (1991). Should opiate addicts be involved in

controlling their own detoxification? A comparison of fixed versus negotiable schedules.

British Journal of Addiction, 86, 977-982.

Day, E. (2005). Opiate Detoxification in an Inpatient Setting (Research Briefing 9). London:

National Treatment Agency for Substance Misuse.

Day, E., Ison, J., Keaney, F., Buntwal, N., & Strang, J. (2005). A National Survey of Inpatient Drug

Treatment Services in England. London: National Treatment Agency for Substance Misuse.

Day, E., Ison, J., & Strang, J. (2005). Inpatient versus other settings for detoxification for opioid

dependence. Cochrane Database of Systematic Reviews Issue 2. Art. No.: CD004580

Department of Health. (1996). Task Force to Review Services for Drug Misusers: Report of an

Independent Review of Drug Treatment Services in England. London: Department of

Health.

Finney, J. W., Hahn, A. C., & Moos, R. H. (1996). The effectiveness of inpatient and outpatient

treatment for alcohol abuse: the need to focus on mediators and moderators of setting

effects. Addiction, 91(12), 1773-1796.

Ghodse, A. H., Reynolds, M., Baldacchino, A. M., Dunmore, E., Byrne, S., Oyefeso, A., et al.

(2002). Treating an opiate-dependent inpatient population: A one-year follow-up study of

treatment completers and noncompleters. Addictive Behaviors, 27, 765-778.

Gossop, M. (2003). Withdrawal and Detoxification. In Drug Addiction and its Treatment (pp. 115-

136). Oxford: Oxford University Press.

Gossop, M., Green, L., Phillips, G., & Bradley, B. (1989). Lapse, relapse and survival among opiate

addicts after treatment. A prospective follow-up study. British Journal of Psychiatry, 154,

348-353.

Gossop, M., Johns, A., & Green, L. (1986). Opiate withdrawal: inpatient versus outpatient

programmes and preferred versus random assignment to treatment. British Medical Journal,

293, 103-104.

Gossop, M., Marsden, J., Stewart, D., & Rolfe, A. (1999). Treatment retention and 1 year outcomes

for residential programmes in England. Drug and Alcohol Dependence, 57, 89-98.

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

24

Gossop, M., & Strang, J. (1991). A comparison of the withdrawal responses of heroin and

methadone addicts during detoxification. British Journal of Psychiatry, 158, 697-699.

Gossop, M., & Strang, J. (2000). Price, cost and value of opiate detoxification treatments.

Reanalysis of data from two randomised trials. British Journal of Psychiatry, 177, 262-266.

HM Government. (2008). Drugs: Protecting Families and Communities. The 2008 Drug Strategy.

London: The Home Office.

Kleber, H. D. (1999). Opioids: Detoxification. In H. D. Kleber & M. Galanter (Eds.), Textbook of

substance abuse treatment. (pp. 251-269). Washington: American Psychiatric Press.

Kleber, H. D., Riordan, C. E., Rounsaville, B., Kosten, T., Charney, D., Gaspari, J., et al. (1985).

Clonidine in outpatient detoxification from methadone maintenance. Arch Gen Psych, 42,

391-394.

Lash, S. J. (1998). Increasing participation in substance abuse aftercare treatment. Am. J. Drug

Alcohol Abuse, 24(1), 31-36.

Lipton, D. S., & Maranda, M. J. (1983). Detoxification from heroin dependency: an overview of

method and effectiveness. Advances in Alcohol and Substance Abuse, 3, 31-55.

Marlatt, G. A., & Donovan, D. M. (2005). Relapse Prevention: Maintenance Strategies in the

Treatment of Addictive Behaviors. New York: Guilford Press.

Marsden, J., Gossop, M., Stewart, D., Best, D., Farrell, M., Lehmann, P., et al. (1998). The

Maudsley Addiction Profile (MAP): a brief instrument for assessing treatment outcome.

Addiction, 93(12), 1857-1868.

Marsden, J., Nizzoli, U., Corbelli, C., Margaron, H., Torres, M., Castro, I. P. d., et al. (2000). New

European instruments for treatment outcome research: reliability of the Maudsley Addiction

Profile and Treatment Perceptions Questionnaire in Italy, Spain and Portugal. European

Addiction Research, 6, 115-122.

Marsden, J., Stewart, D., Gossop, M., Rolfe, A., Bacchus, L., Griffiths, P., et al. (2000). Assessing

client satisfaction with treatment for substance use problems and the development of the

treatment perceptions questionnaire (TPQ). Addiction Research, 8(5), 455-470.

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

25

Mattick, R. P., & Hall, W. (1996). Are detoxification programmes effective? The Lancet, 347, 97-

100.

McCusker, J., Bigelow, C., Luippold, R., Zorn, M., & Lewis, B. F. (1995). Outcomes of a 21-day

drug detoxification program: retention, transfer to further treatment, and HIV risk reduction.

American Journal of Drug and Alcohol Abuse, 21(1), 1-16.

Miller, W. R., & Hester, R. K. (1986). Inpatient alcoholism treatment. Who benefits? American

Psychologist, 41(7), 794-805.

Moos, R. H. (1997). Evaluating Treatment Environments. Somerset, NJ: Transaction Publishers.

Moos, R. H., Pettit, B., & Gruber, V. (1995). Longer episodes of community residential care reduce

substance abuse patients' readmission rates. Journal of Studies on Alcohol, 56, 433-443.

National Collaborating Centre for Mental Health. (2007). Opiate Detoxification for Drug Misuse.

London.

National Treatment Agency for Substance Misuse. (2005). Tier 4 Drug Treatment in England:

Summary of Inpatient Provision and Needs Assessment (Research Briefing 7). London:

National Treatment Agency for Substance Misuse.

Sannibale, C., Hurkett, P., Van den Bossche, E., O'Connor, D., Zador, D., Capus, C., et al. (2003).

Aftercare attendance and post-treatment functioning of severely substance dependent

residential treatment clients. Drug and Alcohol Review, 22, 181-190.

SCAN Inpatient Treatment Working Party. (2006). SCAN Consensus Project: Inpatient Treatment

of Drug And Alcohol Misusers in the National Health Service. London: Specialist Clinical

Addiction Network.

Seivewright, N. (2000). Community Treatment of Drug Misuse: More Than Methadone.

Cambridge: Cambridge University Press.

Sheridan, J., Cook, C., & Strang, J. (1999). Audit of the in-patient management of opioid

withdrawal using lofexidine hydrochloride. Journal of Substance Use, 4, 29-34.

Simpson, D., & Sells, S. B. (1990). Opioid Addiction and Treatment: A 12-Year Follow-Up.

Malabar, Florida: Robert E. Krieger Publishing Co.

Day E & Strang J (2011) Journal of Substance Abuse Treatment, 40(1) 56-66 DOI: http://dx.doi.org/10.1016/j.jsat.2010.08.007

http://www.journalofsubstanceabusetreatment.com/article/S0740-5472(10)00189-3/abstract

26

Strang, J., Bearn, J., & Gossop, M. (1999). Lofexidine for opiate detoxification: Review of recent

randomised and open controlled trials. The American Journal on Addictions, 8, 337-348.

Strang, J., Marks, I., Dawe, S., Powell, J., Gossop, M., Richards, D., et al. (1997). Type of hospital

setting and treatment outcome with heroin addicts. British Journal of Psychiatry, 171, 335-

339.

Strang, J., McCambridge, J., Best, D., Beswick, T., Bearn, J., Rees, S., et al. (2003). Loss of

tolerance and overdose mortality after inpatient opiate detoxification: follow up study. BMJ,

326(7396), 959-960.

Tiet, Q. Q., Ilgen, M. A., Byrnes, H. A., Harris, A. H. S., & Finney, J. W. (2007). Treatment setting

and baseline substance use severity interact to predict patients' outcomes. Addiction, 102(3),

432-440.

Weiss, R. D. (1999). Inpatient Treatment. In M. Galanter & H. D. Kleber (Eds.), Textbook of

Substance Abuse Treatment (Second ed.). Washington DC: The American Psychiatric Press.

Wilson, B. K., Elms, R. R., & Thomson, C. P. (1975). Outpatient vs hospital methadone

detoxification: an experimental comparison. The International Journal of the Addictions,

10(1), 13-21.

World Health Organisation. (1992). The ICD-10 Classification of Mental and Behavioural

Disorders. Geneva: WHO.