Implementation of direct health facility financing in public ...

The University of Dodoma

University of Dodoma Institutional Repository http://repository.udom.ac.tz

Health Sciences Doctoral Theses

2020

Implementation of direct health facility

financing in public primary health

facilities in Tanzania: effects on health

system performance.

Kapologwe, Ntuli Angyelile

The University of Dodoma

Kapologwe, N. A. (2020). Implementation of direct health facility financing in public primary

health facilities in Tanzania: effects on health system performance (Doctoral thesis). The

University of Dodoma, Dodoma.

http://hdl.handle.net/20.500.12661/2826

Downloaded from UDOM Institutional Repository at The University of Dodoma, an open access institutional repository.

IMPLEMENTATION OF DIRECT HEALTH FACILITY

FINANCING IN PUBLIC PRIMARY HEALTH

FACILITIES IN TANZANIA: EFFECTS ON HEALTH

SYSTEM PERFORMANCE

NTULI ANGYELILE KAPOLOGWE

DOCTOR OF PHILOSOPHY IN PUBLIC HEALTH

THE UNIVERSITY OF DODOMA

DECEMBER, 2020

IMPLEMENTATION OF DIRECT HEALTH FACILITY

FINANCING IN PUBLIC PRIMARY HEALTH FACILITIES IN

TANZANIA: EFFECTS ON HEALTH SYSTEM PERFORMANCE

BY

NTULI ANGYELILE KAPOLOGWE

A THESIS SUBMITTED IN FULFILLMENT OF THE

REQUIREMENTS FOR THE DEGREE OF DOCTOR OF

PHILOSOPHY IN PUBLIC HEALTH

THE UNIVERSITY OF DODOMA

DECEMBER, 2020

i

DECLARATION AND COPYRIGHT

I, Ntuli Angyelile Kapologwe, declare that this is my own original work and that it

has not been presented and will not be presented to any other University for a similar

or any other degree award.

Signature

No part of this thesis may be reproduced, stored in any retrieval system, or

transmitted in any form or by any means without prior written permission of the

author or the University of Dodoma (UDOM). If transformed for publication in any

other format shall be acknowledging that, this work has been submitted for degree

award at the University of Dodoma.

ii

CERTIFICATION

The undersigned certify that they have read and hereby recommend for acceptance

by the University of Dodoma thesis titled “Implementation of Direct Health Facility

Financing in the Public Primary Health Facilities in Tanzania: Effect on Health

System Performance” in fulfillment of the requirements for the degree of PhD in

Public Health of the University of Dodoma.

Dr. Stephen Kibusi

Signature _ Date_____03/12/2020__________

(SUPERVISOR)

Dr. Albino Kalolo

Signature Date______03/12/2020__________

(Co - SUPERVISOR)

iii

ACKNOWLEDGEMENTS

To Almighty God, may there always be Glory. I am grateful to Him and I thank Him

for His mercy and grace.

Contributions and moral support from several people, organization and institutions

made this thesis successful. I would like to give my sincere thanks and appreciation

to each and everyone who in one way or another made this journey successful. First,

I would like to give my thanks to the Permanent Secretary, President’s Office-

Regional Administration and Local Government Authority, for granting me

permission to pursue my PhD studies at the University of Dodoma.

I would also like to express my deep gratitude to Dr. Stephen Kibusi and Dr. Albino

Kalolo, my research supervisors, for their guidance, encouragement, and mentorship

towards achieving this research work. My special tributes go to Professors Josephine

Borghi and Tuntufye Mwamwenda for their scholarly comments and valuable time in

the evaluation and examination of the Thesis. From their national and international

University careers, I benefited immensely.

I wish to acknowledge and thank the Swiss Development Cooperation (SDC) and

UNICEF for sponsoring my PhD Studies. Furthermore, I would like to register my

gratitude to the Center for Reforms, Innovation, Health Policy and Implementation

Research (CeRIHI) for their massive support in making this thesis a success. I would

also like to extend my sincere thanks to Regional Administrative Secretaries of the

regions of Pwani, Mtwara, Dodoma, Mbeya, Shinyanga, Manyara and Katavi - for

their support of this study. I am grateful to the District Executive Directors of

Dodoma City Council and Bahi District Council, Kibaha Town Council and

Kisarawe District Council, Mtwara Municipal Council and Nanyumbu District

Council, Mbeya City Council and Rungwe City Council, Shinyanga Municipal

Council and Ushetu District Council, Babati Town Council and Hanang District

Council, Mpanda Municipal Council and Mlele District Council, District Medical

Officers and all the staff and participants of the respective primary health facilities in

respective District Councils for their support and assistance during baseline, mid line

and terminal data collection exercises.

iv

I would also like to send my sincere thanks to my friends, Sally Lake, Ally

Kananika, Hendry Samky, Ally Kinyanga and Abdallah Ally, for their tireless

support in shaping the thesis.

Lastly, I would like to thank Eng. Joseph M Nyamhanga with whom I shared my

ideas and who supported me to the end of this important journey of my life.

v

DEDICATION

This thesis is dedicated to my lovely wife, Dr. Julieth Simon Kabengula, and our

lovely daughters, Tusekile and Twitike, and to our son, Rodney. Without their

support, it would not have been possible to complete this dissertation.

I am also dedicating my work to my parents, Mr. Angyelile Mwasomola Kapologwe

and Mrs. Imani Ambakisye Mwasomola, and my siblings, Lugano, Rhoda and

Nisalile, for their moral support and encouragement in any step I took in making this

thesis a success. May Almighty God protect them and give them the strength they

need.

Lastly, I dedicate this work to all health care workers of United Republic of Tanzania

for their continued effort of offering services to the Tanzanians. To them, I will

always be grateful.

vi

ABSTRACT

Background: Tanzania, like many countries globally, has continuously been

implementing health sector reforms with the intent of ensuring affordable health

services to the population, in alignment with the shift towards Universal Health

Coverage as well as improving the quality of health services, among these reforms

has been the introduction of Direct Health Facility Financing (DHFF), which has

limited evidence on its implementation and outcomes with respect to its intended

goals. This study aimed to assess the implementation of DHFF in the Public Primary

Health Facilities (PPHFs) as well as its effects on health system performance in

Tanzania.

Methods: This study was conducted from January 2018 to September 2019 utilized a

before and after non-controlled study design with a process evaluation embedded at

midline. Both quantitative and qualitative methods were used for data collection. A

structured questionnaire was used to obtain the quantitative data from exiting

patients, and health care workers in 42 PPHFs. Qualitative data was collected using

an interview guide through 14 in depth interviews and seven focus group discussions

to health managers at council level from 14 councils in seven regions. Quantitaive

data analysis was done using SPSS™ version 25 while for qualitative NVIVO

QSR™ version 12 was used.

Results: A total of 844 exiting patients were studied on seven domains of Health

System Responsiveness (HSR) at baseline and endline. All but one domain (i.e.

communication) of HSR were found to have positive significant difference between

baseline and endline. Overall HSR improved significantly from 62.84% at baseline to

71.20% at end line (p<0.0001). In a multivariable logistic regression model, patients

who were studied at the end line were four times more likely to have experienced

positive responsiveness as compared to those at the baseline.

Forty-two (42) PPHFs were studied based on structural quality of maternal health

services (MHS). The majority (71%) of the facilities were within 10 kilometers of

participants’ residences. Of note, the majority (88.9%) of dispensaries and 60% of

health centers were below the required staffing level. There were significant

vii

differences on structural quality of MHS between baseline and endline (2.38% and

30.9% respectively).

With respect to maternal health services utilization, 42 facilities were studied on nine

maternal health service consumption indicators before and after the introduction of

the DHFF initiative. There were significant differences in all indicators after DHFF

introduction except those for intermittent presumptive treatment for malaria second

dose, use of modern family planning methods, and mebendazole use.

Of 238 health care providers who participated in the Fidelity of Implementation (FoI)

component of the study, the majority (76%) had adequate knowledge on the DHFF

implementation issues while only 28% had high Fidelity of Implementation (FoI

scores. For health service providers working in the rural settings, training and

knowledge on the DHFF and Facility Financial Accounting and Reporting System

(FFARS) were significantly associated with high FoI. In the multiple logistic

regression model, both positive HSR (AOR 3.4 [95%CI: 1.637, 7.064]), and high

maternal health service utilization (AOR 40.971[95%CI: 4.065, 412.927]) were

strongly associated with high structural quality of MHS.

Health service providers indicated an improvement in both governance and

accountability, especially in the areas of planning and budgeting, financial

management and transparency, with the implementation of the DHFF. However, the

majority of health care providers admitted to have not received adequate supportive

supervision and mentorship. The providers also felt that this initiative added an

additional burden to their routine work.

Conclusions: In general, the DHFF have affected the health system, specifically on

health system performance in Tanzania. FoI was low despite training being offered to

health care providers, which indicates a need for more investment on this aspect in

order to have a resilient health system. Improved governance and accountability have

been positive outcome on DHFF implementation, which indicate the merits in

ongoing efforts to invest on governance and accountability at PPHFs.

viii

TABLE OF CONTENTS

DECLARATION AND COPYRIGHT ........................................................... i

CERTIFICATION ........................................................................................ ii

ACKNOWLEDGEMENTS ......................................................................... iii

DEDICATION.............................................................................................. v

ABSTRACT ................................................................................................ vi

TABLE OF CONTENTS ........................................................................... viii

LIST OF TABLES ................................................................................... xviii

LIST OF FIGURES ................................................................................... xxi

LIST OF PUBLICATIONS ..................................................................... xxiii

LIST OF APPENDICES .......................................................................... xxiv

LIST OF ABBREVIATIONS .................................................................. xxvi

DEFINITIONS OF TERMS ..................................................................... xxix

CHAPTER ONE ......................................................................................... 1

INTRODUCTION ...................................................................................... 1

1.0 Overview ................................................................................................ 1

1.1 Background Information ......................................................................... 1

1.2 Problem Statement ................................................................................ 25

1.3 Objectives ............................................................................................. 26

1.3.1 Broad objective .................................................................................. 26

ix

1.3.2 Specific Objectives............................................................................. 26

1.3.3 Specific Research Questions ............................................................... 26

1.4 Significance of the Study ...................................................................... 27

CHAPTER TWO ...................................................................................... 29

LITERATURE REVIEW ......................................................................... 29

2.0 Overview .............................................................................................. 29

2.1 Theoretical Literature Review ............................................................... 29

2.1.1 Health System Performance Model (HSPM) ...................................... 31

2.1.2 Health System Reforms Model (HSRM) ............................................ 32

2.1.3 Fidelity of Implementation (FoI) ........................................................ 35

2.1.4 Fidelity of Implementation (Adherence) ............................................. 36

2.1.5 Participants’ Responsiveness towards DHFF implementation ............. 38

2.1.6 Context and Description of DHFF Implementation in Tanzania .......... 39

2.1.7 Implementation Research ................................................................... 41

2.1.8 Evaluation of Health Inteventions or Programs ................................... 43

2.2 Empirical Review .................................................................................. 45

2.2.1 Situation of Health Care Financing Reforms....................................... 45

2.2.2 Implementation of DHFF initiative in Tanzania .................................. 48

2.2.2.1 Delays in Disbursement of Funds .................................................... 48

x

2.2.2.2 Weaknesses in Planning and Budgeting of Activities ....................... 49

2.2.2.3 Weak Supportive Supervision and Mentorship ................................ 50

2.2.2.4 Failure to Maintain Buildings and Supply Chain of Medicines and

Supplies ........................................................................................... 50

2.2.3 HSR in PPHF ..................................................................................... 51

2.2.4 Structural Quality of MHS ................................................................. 51

2.2.5 Maternal Health Service Utilization .................................................... 53

2.2.6 Conceptual Frameworks Guiding the Study ........................................ 54

2.2.7 Theory of Change (ToC) .................................................................... 54

2.2.8 Governance and Accountability .......................................................... 57

2.2.9 Research or Knowledge Gap .............................................................. 59

CHAPTER THREE .................................................................................. 60

METHODOLOGY ................................................................................... 60

3.0 Overview .............................................................................................. 60

3.1 Study Settings ....................................................................................... 60

3.2 Research Approaches ............................................................................ 63

3.3 Study Design ......................................................................................... 63

3.4 Target Population .................................................................................. 64

3.5 Study Population ................................................................................... 64

3.5.1 Inclusion Criteria................................................................................ 65

xi

3.5.2 Exclusion Criteria .............................................................................. 65

3.6 Unit of Analysis .................................................................................... 65

3.7 Sampling Method and Sample Size Estimation ...................................... 66

3.7.1 Sampling Methods for Quantitative Data ............................................ 66

3.7.2 Quantitative Sample Size Determination ............................................ 69

3.7.3 Quantitative Data Collection .............................................................. 69

3.7.4 Sampling Methods for Qualitative Data .............................................. 70

3.7.5 Data Collection Methods and Tools for Qualitative ............................ 71

3.8 Sample Size Determination ................................................................... 72

3.9 Data Collection Tools............................................................................ 73

3.10 Pre-testing of Tools ............................................................................. 86

3.11 Data Collection Procedure ................................................................... 86

3.12 Data analysis.…………………………………………………………...87

3.12.1 Quantitative data analysis …………………………………………....87

3.12.1.1 Variables and their measures ......................................................…87

3.12.1.1.1 Dependent Variables................................................................... 87

i. HSR .................................................................................................... 87

ii. Structural quality of MHS ................................................................... 87

iii. MHS utilization

iv. Implementation fidelity of DHFF initiative

xii

v. Governance and accountability ............................................................ 87

3.12.1.1.2 Independent Variables ................................................................ 87

3.12.1.2 Categorization of Variables ........................................................... 88

3.12.1.2.1 Categorization of Health System Responsiveness ....................... 88

3.12.1.2.2 Categorization of Structural Quality of MHS .............................. 91

3.12.1.2.3 Categorization of Health Service Utilization ............................... 92

3.12.1.2.4 Categorization of Fidelity of Implementation (FoI) ..................... 92

3.13 Data Processing and Statistical Analysis .............................................. 93

3.13.1 Statistical analysis ........................................................................... 93

3.13.2 Governance and Accountability ....................................................... 94

3.14 Qualitative Data Analysis .................................................................... 94

3.15 Integration of Quantitative and Qualitative methods ............................ 95

3.16 Trustworthiness and Credibility ........................................................... 96

3.17 Validity and Reliability ....................................................................... 97

3.17.1 Validity ........................................................................................... 97

3.17.2 Reliability ........................................................................................ 97

3.18 Ethical Consideration ........................................................................ 102

CHAPTER FOUR................................................................................... 103

RESULTS ................................................................................................ 103

xiii

4.0 Introduction ........................................................................................ 103

4.1 Perception of Patients on Health System Responsiveness (HSR) in PPHF

...................................................................................................... 104

4.1.1 Socio-Demographic Characteristics of Patients in PPHFs ................. 104

4.1.2 Status of Health System Responsiveness (HSR) as perceived by patients

in PPHF (PPHFs) before and after implementation of Direct Health

Facility Financing (DHFF) ............................................................ 106

4.1.3 Factors Associated with change in overall perception of patients on

Health System Responsiveness in PPHF before and after

implementation of Direct Health Facility Financing. ...................... 110

4.2 Level of Structural Quality of MHS in the PPHF before and after Direct

Health Facility Financing Implementation ..................................... 119

4.2.1 Demographic Characteristics of PPHF before and after implementation

of Direct Health Facility Financing ................................................ 119

4.2.2 Level of Structural Quality of Maternal Health Service in PPHF before

and after implementation of Direct Health Facility Financing. ....... 128

4.2.3 Other factors that influenced change in structural quality of MHS in

PPHF before and after implementation of Direct Health Facility

Financing....................................................................................... 133

4.3 Level of maternal health service utilization in PPHF before and after

implementation of Direct Health Facility Financing ....................... 136

4.3.1 Other factors associated with Change of Maternal Health Service

Utilization in PPHF before and after implementation of Direct Health

Facility Financing. ......................................................................... 137

4.4 Level of Direct Health Facility Financing (DHFF) Fidelity of

Implementation (FoI) and its potential moderators ......................... 141

xiv

4.4.1 Socio-Demographic Characteristics of Health Service Providers in

PPHFs ........................................................................................... 141

4.4.2 Knowledge of Health Service Providers on the implementation of

Direct Health Facility Financing in PPHF. ..................................... 144

4.4.3 Fidelity of Implementation of DHFF by Health Service Providers . 149

4.5 The association between Fidelity of Implementation (FoI) and overall

Structural quality of maternal health service in PPHF. ................... 151

4.6 The association between Fidelity of Implementation (FoI) and overall

Maternal health service utilization in PPHF before and after

implementation of Direct Health Facility Financing. ...................... 153

4.6.1 Health Care Providers’ experience towards Direct Health Facility

Financing...........................................................................................154

4.6.2 Thematic Findings for Qualitative Research......................................155

4.6.2.1 Training on DHFF, FFARS, Planning and Budgeting .................... 156

4.6.2.2 Knowledge (Content) on DHFF initative ....................................... 157

4.6.2.3 Coordination of DHFF activities .................................................... 157

4.6.2.4 Attitude of Health care providers towards DHFF ........................... 159

4.6.2.5 Timeliness in addressing issues due to DHFF presence .................. 159

4.6.2.6 Timely availability of health service provider’s benefits, tools and

other resources .............................................................................. 160

4.6.2.7 Timely fund utilization and service provision ................................ 160

4.6.2.8 Capacity of health service providers on management of DHFF

initiative ........................................................................................ 161

4.6.2.9 Timeleness in disbursement of funds ............................................. 162

xv

4.6.2.10 Improvement in working environment ......................................... 162

4.7. Thematic Number 1: Governance…………………………………….162

4.7.1 Governance of DHFF ....................................................................... 163

4.7.2 Funds Management .......................................................................... 164

4.7.3 Capacity of Health Facility Governing Committee (HFGC) Members

...................................................................................................... 164

4.7.4 Planning and Budgeting ................................................................... 164

4.8 Thematic Area Number 2: Accountability ........................................... 165

4.8.1 Transparency and Community Involvement ..................................... 165

4.8.2 Financial and performance report review and approval ..................... 166

4.8.3 Oversight ......................................................................................... 166

4.9 Thematic Number 3: Supportive Supervision and Mentorship ............. 166

4.9.1 Regular monitoring of primary health facilities, Consistency provision

of guidance………………………………………………………….168

4.9.2 Clarification/ interpretation of policies, strategies and operational

guidelines…………………………………………………………..168

4.9.3 Integration of Quantitative and Qualitative Results ........................... 168

4.9.3.1 Framework for Integration ............................................................. 168

4.9.3.2 Interpretation of the Integrated Results .......................................... 169

CHAPTER FIVE .................................................................................... 170

DISCUSSION .......................................................................................... 170

xvi

5.0 Introduction ........................................................................................ 170

5.1 Health System’s Responsiveness ......................................................... 170

5.2 Structural Quality of MHS .................................................................. 172

5.3 Maternal Health Service Utilization .................................................... 173

5.4 FoI towards DHFF initiative................................................................ 174

5.5 Health Care Providers’ responsiveness towards implementation of Direct

Health Facility Financing in PPHF. ............................................... 177

5.6. Governance and Accountability on DHFF implementation....................178

5.6.1 Governance ...................................................................................... 178

5.6.2 Accountability .................................................................................. 179

5.6.3 Theoretical reviews .......................................................................... 181

5.6.4 Conceptual frameworks .................................................................... 182

CHAPTER SIX ....................................................................................... 184

CONCLUSION, RECOMMENDATIONS AND SUGGESTION FOR

FURTHER RESEARCH ............................................................. 184

6.1 Conclusion .......................................................................................... 184

6.2 Recommendations ............................................................................... 185

6.2.1 Recommendation to Policy makers ................................................... 185

6.2.2 Recommendation to Policy Implementers ......................................... 186

6.2.3 Recommendations to the DHFF Initiative ......................................... 188

xvii

6.3 Strengths of the Study ......................................................................... 189

6.4 Limitations of the Study ...................................................................... 189

6.5 Suggestions for Further Research ........................................................ 190

REFERENCES ......................................................................................... 192

PUBLICATIONS ..................................................................................... 217

APPENDICES .......................................................................................... 219

xviii

LIST OF TABLES

Table 1: Star Rating and Grading Criteria of Primary Health Facilities .................... 19

Table 2: Allocation of funds pre and post-introduction of DHFF ............................. 24

Table 3: Scoring Criteria/Cut-offs for HSR Performances ....................................... 74

Table 4: Health Care Responsiveness Performance Criteria and their Categorization

........................................................................................................................ 90

Table 5: Reliabilities of each Domain of HSR ......................................................... 90

Table 6: Data Analysis Plan and Measurement of Variables for Baseline and Endline

Study............................................................................................................... 98

Table 7: Objective number 4 ................................................................................. 101

Table 8: Social Demographic Characteristics of the Patients in PPHF (PPHFs) before

and after implementation of Direct Health Facility Financing (DHFF) (n=844)

...................................................................................................................... 104

Table 9: Two Sample Independent T-test for Perception of Patients towards Health

System Responsiveness before and after implementation of Direct Health

facility Financing (DHFF) in PPHF (PPHF) (n=844, p=0.05) ........................ 109

Table 10: Chi-squire test for factors associated with change in overall perception of

Health System Responsiveness among patients who attended in the public

primary health care facilities before and after DHFF implementation (n=844,

p=0.05). ........................................................................................................ 112

Table 11: Logistic regression model for factors associated with change in overall

perception of Health System Responsiveness before and after implementation of

Direct Health Facility Financing in PPHF (p<0.05, 95% CI). ........................ 115

Table 12: Characteristics of the Public Primary Health Facilities before and after

implementation of Direct Health Facility Financing (n=42). .......................... 120

xix

Table 13: Descriptive Results on Change of Structural quality of MHS in Tanzania

before and after DHFF implementation (n= 42). ............................................ 124

Table 14: Paired T-test Results for Change in Structural Quality of MHS in PPHFs

before and after implementation of DHFF (n=42, p=0.05, CI = 95%). ........... 130

Table 15: Fisher’s exact test results for other factors that attributed to change in

overall structural quality of MHS in PPHF before and after implementation of

Direct Health Facility Financing (n= 42). ...................................................... 134

Table 16: Cell analysis results for for other factors that attributed to change in overall

structural quality of MHS in PPHF before and after implementation of Direct

Health Facility Financing (n= 42) .................................................................. 135

Table 17: Wilcoxon signed rank test for Change in Maternal Health Service

Utilization before and after implementation of Direct Health Facility Financing

in PPHF (n=42, p= 0.05). .............................................................................. 138

Table 18: Poisson Regression Model for other factors associated with change of

maternal health service utilization in Public Primary Health Facilities before and

after implementation of DHFF. ..................................................................... 140

Table 19: Social Demographic characteristics of Health Care Workers (HCWs) in

PPHF implementing DHFF. .......................................................................... 142

Table 20: Descriptive results of knowledge among health care workers in PPHF

during the implementation of Direct Health Facility Financing ...................... 144

Table 21: Level of FoI among Health Service Providers in PPHFs ........................ 146

Table 22: Logistic regression model for factors associated knowledge (p=0.005) .. 147

Table 23: Chi-squire test for factors associated with FoI ....................................... 149

Table 24: Logistic regression results for factors associated with FoI in PPHF before

and after implementation of DHFF (n=234). .................................................. 151

xx

Table 25: Association between Changes of Structural Quality by Fidelity of

Implementation ............................................................................................. 152

Table 26: Logistic regression results for association between FOI and overall

structural quality of HS and in PPHFs before and after implementation of DHFF

(n= 42). ......................................................................................................... 152

Table 27: Association between overall changes of health service utilization and

Fidelity of Implementation of Direct Health facility financing in PPHFs. ...... 153

Table 28: Logistic regression results for the association between FoI and overall

maternal health service utilization in PPHFs before and after implementation of

DHFF. ........................................................................................................... 154

Table 29: Socio-demographic Characteristics of Interviewees ............................... 155

Table 30: Results from Indepth interviews by thematic and sub thematic areas ..... 155

Table 31: Characteristics of Interviewees .............................................................. 163

xxi

LIST OF FIGURES

Figure 1. Health System Building Blocks (Adam & Savigny, 2012) ...................... 10

Figure 2: Estimated health sector expenditures, projections and sources of funding

(million TZS). ................................................................................................. 15

Figure 3: Tanzania Health Sector Milestones and Reform Timelines from 1880s to

2019 ................................................................................................................ 17

Figure 4: Cascade Training on Direct Health Facility Financing (DHFF) across

Tanzania. ........................................................................................................ 21

Figure 5: Funds disbursement before and after DHFF implementation. ................... 24

Figure 6: Opening the ‘black box’ of the DHFF through assessing the

implementation processes. ............................................................................... 28

Figure 7: The health reform cycle (modified from Roberts et al., 2002)

31

Figure 8: Health System Performance Model (Berman & Bitran, 2011a) ................ 33

Figure 9: Direct Health Facility Financing Management Framework ...................... 41

Figure 10: Health Financing Arrangement, Intermediate Objectives and UHC goals

(WHO, 2013). ................................................................................................. 46

Figure 11: Direct Health Facility Financing (DHFF) Funds Flow and Stakeholders

Relationship in Tanzania (Kapologwe. et al., 2020). ........................................ 49

Figure 12: Theory of Change of DHFF Implementation in Tanzania ....................... 56

Figure 13: Conceptual Framework for Fidelity of Implementation .......................... 56

Figure 14: Map of Tanzania that depicts the distribution of District Councils and

PPHF that participated in the study (Kapologwe et al., 2019). ......................... 61

xxii

Figure 15: The Organization of Health System of Tanzania .................................... 62

Figure 16: Schematic Presentation of Sampling Technique ..................................... 68

Figure 17: Schematic Presentation of Data Collection Processes ............................. 70

Figure 18: Schematic Presentation of Process and Outcome Evaluation of the DHFF

Implementation ............................................................................................... 81

Figure 19: Modified Conceptual FoI Framework after the Midline Study (Modified

from Original Framework by Kapologwe et al., 2019). .................................... 85

Figure 20: Summary of Study Variables ................................................................ 88

Figure 21: Schematic Presentation of Integration of Mixed Methods Triangulation

Design for Measuring DHFF Fidelity of Implementation (FoI) ........................ 96

Figure 22: Regional pefromance on structural quality of MHS before and after

implementation of Direct Health Facility Financing (n=42). .......................... 108

Figure 23: Percentage Change of Structural Quality of Maternal Health service in

PPHF before and after implementation of Direct Health Facility Financing

(n=42). .......................................................................................................... 133

xxiii

LIST OF PUBLICATIONS

Publication Paper 1: Understanding the implementation of direct health facility

finacing and its effect on health system performance in Tanzania .................. 217

Publication Paper 2: Assessing health system responsiveness in primary health care

facilities in Tanzania ..................................................................................... 218

xxiv

LIST OF APPENDICES

Appendix I: A descriptive results of Health System Responsiveness as perceived by

patients in PPHF before and after implementation of DHFF .......................... 219

Appendix II: Prompt to Attention .......................................................................... 221

Appendix III: Respect to Dignity .......................................................................... 223

Appendix IV: Clear Communication ..................................................................... 224

Appendix V: Respect to Autonomy ....................................................................... 226

Appendix VI: Access to Care ................................................................................ 227

Appendix VII: Respect to Confidentiality ............................................................. 229

Appendix VIII: Basic Amenities ........................................................................... 230

Appendix IX: Association between Structural Quality and HSR in PPHF after

implementation of DHFF (p<0.05; CI = 95%) ............................................... 234

Appendix X: Table of Random Numbers .............................................................. 235

Appendix XI: Coding Table for Indepth Interviews ............................................... 236

Appendix XII: Coding table for Focus Group Discussion ...................................... 240

Appendix XIII: Objective number One: Health System Responsiveness as perceived

by the end users of PPHF .............................................................................. 244

Appendix XIV: Objective number two: Structural Quality of MHS in the PPHF ... 257

Appendix XV: Objective number three: Maternal Health Service Utilization in the

PPHF ............................................................................................................ 284

Appendix XVI: Objective Number Four: To explore Governance and Accountability

on DHHF in PPHFs ....................................................................................... 289

xxv

Appendix XVII: Objective Number Five (Quantitative): Level of Fidelity of

Implementation and its potential moderators ................................................. 293

Appendix XVIII: Objective number five (Qualitative): Participant’s Responsiveness

to DHFF ........................................................................................................ 313

Appendix XIX: Consent Form .............................................................................. 315

Appendix XX: Ethical Clearance Forms................................................................ 317

xxvi

LIST OF ABBREVIATIONS

AIDS Acquired Immunodeficiency Syndrome

ANC Antenatal Care

AOR Adjusted Odds Ratio

ART Antiretroviral Therapy

CHF Community Health Fund

CHMT Council Health Management Team

CHS Community Health Syestem

CHSB Council Health Service Board

CHW Community Health Workers (CHWs)

CORPs Community Own Resource Persons

CTC Centre for Treatment and Counselling

D-by-D Decentralization by Devolution

DED District Executive Director

DFF Direct Facility Financing [Kenya]

DHFF Direct Health Facility Financing

DHIS -2 District Health Information System -2

DMO District Medical Officer

FFARS Facility Financial Accounting and Reporting System

FoI Fidelity of Implementation

xxvii

GoT Government of Tanzania

HCWs Health Care Workers

HF Health Facility

HFGC Health Facility Governance Committee

HIV Human Immunodeficiency Virus

HRH Human Resource for Health

IDI In-Depth Interview

IPD Inpatient Department

LGA Local Government Authority

LMIC Lower- and Middle-Income Countries

MOHCDGEC Ministry of Health, Community Development, Gender, Elderly

and Children

OPD Outpatient Department

PO-RALG President’s Office – Regional Administration and Local

Government

PPHF Public Primary Health Facility

PHC Primary Health Care

P4P Payment for Performance

RBF Results-Based Financing

RMNCH Reproductive, Maternal, Newborn, and Child Health

SDGs Sustainable Development Goals

xxviii

ToC Theory of Change

ToT Training of Trainers

URT United Republic of Tanzania

WHO World Health Organization

xxix

DEFINITIONS OF TERMS

Acceptability Determining how well an intervention will be received by the

target population and the extent to which the new intervention

meet the needs of the target population and organizational setting

(Ayala & Elder, 2011).

Attitude A settled way of thinking or feeling about something.

Caretakers

People from within the family environment taking care of basic

needs required by the patient.

Client Person who uses services available at health facility.

Cost Center Refers to levels of service provision and associated budgets. For

example, in the health sector there are 5 cost centers – Council

Health Management Team (CHMT), Council Hospital, Voluntary

Agency Hospital (VAH), Health Centre and Dispensaries.

Cost Sharing Money paid by clients ‘out of pocket’ and through various

insurance schemes

Decentralization

by Devolution (D

by D)

One of three types of decentralization; the other types are de-

concentration and delegation. Under D-by-D policy the central

government is expected to devolve functions, transfer authority for

decision-making, finance, and management to the local

government authorities. Responsibility for services to city,

municipalities and district councils that elect their own mayors and

councils, raise their own revenues, and have independent authority

to make investment decisions.

Direct Health

Facility

Financing

The process of disbursing funds directly to the health facilities by

the central Government of Tanzania (GoT).

xxx

Effect

Exercised

Outcomes or Results from implementation of DHFF initiative.

Means how implementation of DHFF in PPHF adhered to the

principles of Governance and Accountability.

Fidelity of

Implementation

Extent to which an intervention, initiative, or program is delivering

as intended by the intervention model (Carroll et al., 2007).

Health Basket

Fund

Joint funding mechanism in Tanzania whereby Development

partners pool un-earmarked resources to support non-salary

recurrent costs of primary level health services.

Health

Management

Information

System

Data collection and dissemination system specifically designed to

support planning, management, and decision making in

health facilities and organizations.

Health Planning Process of getting agreed priorities and direction for health sector

in the light of available resources.

Health service Any service (i.e. not limited to medical or clinical services) aimed

at contributing to improved health or to the diagnosis, treatment,

and rehabilitation of sick people.

Health System

Performance

Indicators

Indicators used to measure progress of the health care system. In

this study, the 5 indicators studied were health system

responsiveness, structural quality of maternal health services,

maternal health service utilization, governance, and accountability.

Health system

responsiveness

(HSR)

Assessment of how well the health system meets the legitimate

expectations of the population for the non-health enhancing

aspects of the health system. It includes seven domains: prompt to

attention, autonomy, dignity, quality of basic amenities of care,

communication, respect of confidentiality, and access to care.

Implementa-tion An integrated concept that links research and practice to accelerate

xxxi

Research the development and delivery of public health approaches.

Implementation research involves creation and application of

knowledge to improve the implementation of health policies,

programmes, and practices. Implementation research as a

“scientific inquiry into questions concerning implementation, i.e.,

the act of carrying an intention into effect which in health research

can be policies, programs or individual practices (collectively

called interventions)”(Peters, Tran, & Adam, 2013, p # please).

Knowledge

Information and skills acquired through experience or training on

DHFF, which can be either explicit or implicit.

Morbidity State of illness and disability in a population

Mortality rate Number of deaths within a specific population group and within a

particular period of time.

Outcome A quantitative or qualitative factor or variable that provides a

simple and reliable means to measure achievement, to reflect the

changes connected to an intervention, or to help assess the

performance of an institution.

PlanRep A web-based system designed to assist local government

authorities and public health facilities in planning, budgeting,

projecting revenue from all sources, and tracking of and reporting

on funds received, physical implementation, and expenditures.

Primary Health

Care

Essential preventive and curative health services offered from the

District Council level, Health Centers, Dispensaries up to the

Community level.

Public Health

Facility

All health facilities that are owned and managed by the GoT.

xxxii

Quality of Health

Care Services

Degree to which health services for individuals and populations

increase likelihood of desired health outcomes and consistent with

current professional knowledge (Mainz, 2003).

Results-Based

Financing (RBF)

A cash payment or non-monetary transfer made to a national or

sub-national government, manager, provider, payer, or consumer

of health services after predefined results have been attained and

verified. Payment is conditional on measurable actions being

undertaken." (www.rbfhealth.org). RBF is an umbrella term

because the definition is general and characterizes various

programs in many countries. Different labels exist for essentially

the same concept or are associated with different incentives and

payment arrangements. Any mode of payment that tries to base

financing beyond inputs is called results-based financing.

1

CHAPTER ONE

INTRODUCTION

1.0 Overview

This chapter presents background information on health system performance, health

care financing related reforms and implementation of a decentralization policy in the

health sector in Tanzania. In addition, this chapter introduces the concept of

implementation research, problem statements, research objectives, research and

evaluation questions and significance of the study.

1.1 Background Information

Health system performance and indicators measures

Health system performance, refers the measure of whether the health systems meet

expectation of the beneficiaries, is an important metric for any health care system in

the world. There is a multiplicity of models to measure health system performance,

but measures that consistently reflect the objectives of the system, the nature and

quality of the data, the incentives for stakeholders to scrutinize and act upon the data,

and the culture of the organization within which the data are deployed, are

considered to be more reliable (Smith, 2002; Smith & Busse, 2008). Indicators of a

perfoming system include: HSR, clinical quality, utilization, population health

outcomes, equity, and productivity. Other essential indicators for measuring health

system performance are structural outcomes and process indicators (Musgrove et al.,

2000; Smith, 2002; Smith & Busse, 2008).

Maternal Health Services (MHS)

Women and children are the main users of health services in any given settings. The

reason for their main use of services are due to their predisposition to different

disease conditions (Chakraborty, Islam, Chowdhury, Bari, & Akhter, 2003). This is

one of the reasons which accounts for their high level of attendences in the health

facilities. The empirical evidence has shown that when you address health challenges

that are maternal and child related tend to impact other groups in the general

population (Braveman & Gottlieb, 2014).

2

Globally, it is estimated that 295,000 maternal deaths occurred in the year 2017

(WHO & UNICEF, 2019), with haemorrhage (44,200 deaths) and abortions (43,700

deaths) being the leading causes of mortality (Kinney et al., 2010; Naghavi et al.,

2015). Sub-Saharan African women constitute about 66% (196, 000) of maternal

mortality worldwide (WHO, UNICEF, UNFPA, World Bank Group,

2019). Tanzania ranks among the countries with the highest maternal mortality rates

worldwide (WHO, UNICEF, UNFPA, World Bank Group, 2019). The current

Tanzanian estimated maternal mortality ratio is 556/100,000 live births

(MoHCDGEC et al., 2016).

Tanzania like many other Sub-Saharan Africa (SSA) countries, has been challenged

by high maternal mortality rates since 1990s despite significant government and

partner efforts to minimize the deaths. There have been several strategies,

interventions and reforms dedicated to reduction of mortality with negligible

successes such as use of traditional birth attendants (MOHCDGEC, 2016). In an

effort to reduce maternal deaths and to increase maternal health service utilization

among mothers, the GoT introduced exemptions of costs to cover for health services

among pregnant women and also subsidizing the cost of some essential commodities

like insecticides treated nets through the ‘Hati Punguzo’ program and anti malarial

medicines. Moreover, several strategies have been undertaken, such as One Plan I, II,

III and Health Sector Strategic Plan-IV (2015-2020), which have demonstrated some

success such as increase in immunization coverage and also increase in reproductive

and child health outreach and mobile clinics. Also there have been some programs

including the Results-Based Financing (RBF) which is currently implemented in

eight regions aimed at improving maternal health indicators (MOHSW, 2009).

Results-Based Financing as an intervention is aiming at improving the quantity and

quality of MHS, raising health facility utilization, and improving community

participation on maternal health related activities (MoHSW, 2015a).

In 2017/2018, the Direct Health Facility Financing (DHFF) initiative was introduced

with aim of improving quality of health service delivery at the PPHF with a main

focus being maternal and child health services.

3

Empirical evidence shows that poor quality facility-based health care for women and

their newborns contributes to increased deaths and morbidity in their respective areas

(MoHCDGEC, 2015; Naghavi et al., 2015).

Health System Responsiveness (HSR)

Health system responsiveness (HSR) ‘tends to measure the non-health aspect of care

relating to the environment and the way healthcare services are provided to clients’

(Robone, Rice, & Smith, 2011; Shé et al., 2020; Smith, Rice, Robone, & Smith,

2008, ). The intrinsic goal of responsiveness is to ensure “that people are treated

promptly, with respect for their dignity and their wishes, and that patients receive

adequate physical and effective support while undergoing treatment” (WHO, 2000)

while seeking meaningful improvement of the health care system for the well-being

of the population (Darby, Valentine, Murray, & de Silva, 2001; Smith et al., 2008).

Responsiveness, as it pertains to the health system, mainly focuses on seven domains

that have been divided into two major components: (a) respect for persons (including

dignity, clear communication, confidentiality and autonomy of individuals and

families to decide about their own health); and (b) client orientation (prompt

attention, access to social support networks during care and quality of basic

amenities) (Darby et al., 2001; Murray & Frenk, 2000). In Tanzania, the overall

health system performance by the PPHF has been low even before the introduction of

DHFF initiative (Kapologwe et al., 2020).

Structural Quality of Health Services

Quality of care can be defined as the “degree to which health services for individuals

and populations increase likelihood of desired health outcomes and consistent with

current professional knowledge” (Gary, 1990). Structural quality of MHS are all

structural quality issues that are related to maternal health service provision in the

health facilities. A study done in five countries in SSA including Tanzania showed

that the quality of the MHS in the majority of primary health care (PHC) facilities is

poor and needs improvement (Kruk et al., 2016; Yahya & Mohamed, 2018).

4

According to Mainz (2003), ‘structures’ refers to

Health system characteristics that affect the system’s ability to meet the

healthcare needs of individual patients or a community. Structural indicators

describe the type and amount of resources used by a health system or

organization to deliver programs and services, e.g. the presence or number of

staff, clients, funds, beds, supplies, and buildingsincluding those which offer

safe surgery facilities (p.525).

A body of evidence exists which suggests there is an association between health care

structures and the quality of health services being offered by those facilities (WHO &

OECD, 2018).

Poor quality of health services provision has been a central challenge affecting the

Tanzanian health system experienced by the health providers working in those health

facilities (Mboya et al., 2016; Renggli, 2017). This quality gap has consequently

resulted in dissatisfaction of service users. A study done in in Dar es Salaam at

Mwananyamala hospital among outpatients on quality of health services rendered to

them, they expressed the poor quality concerns (Khamis & Njau, 2014). Despite

those challenges, the GoT has been investing heavily in improving the structural

quality of the primary health facilities through several interventions including the

Primary Health Service Development Programme (PHSDP) commonly known as

MMAM (Mpango wa Maendeleo wa Afya ya Msingi) in Kiswahili. This has resulted

in renovation and construction of both old and new facilities respectively, as well as

equipping the facilities with essential medical equipment (Kapologwe et al., 2020).

Maternal Health Service Utilization

Health service utilization is the extent to which people are accessing and making use

of health services available within and outside their immediate catchment areas.

Access can be either through admission or through the outpatient services. Optimal

service utilization is a core indicator for any highly performing health care system in

any given setting. Patient service utilization is an important indicator of health

system performance or a proxy indicator for quality of services provision and HSR.

Additionally, health service utilization data can inform planning and budgeting as

well as forecasting and quantification of health commodities and supplies. Moreover,

5

the data can help to identify utilization trends and bypassing of health services by the

members of the community in their catchment areas.

Maternal health service utilization has been a challenge in many resource-limited

countries due to low education levels, poor income and wealth, and rurality (Alam,

Hajizadeh, Dumont, & Fournier, 2015; Kanyangarara, Munos, & Walker, 2017;

Saxena, Vangani, & Mavalankar, 2013). Many studies point to maternal health

service indicators as showing particularly low performance in areas of post-natal care

[i.e., within 3-7 days], institutional deliveries, use of modern family planning

methods, and uptake of a second and third dose of intermittent preventive treatment

for malaria during pregnancy (IPT2) (Kibusi, Kimunai, & Hines, 2015; Martin et al.,

2019; Saxena et al., 2013). Morever, one maternal health service utilisation indicator

that has increased in Tanzania over recent years is the prevalence of women

experiencing facility-based deliveries, shifting from 50% in 2010 to 63% in 2015.

However, this trend varies depending on the area of residence whereby those who are

in urban areas had an attendance of 86% while those in rural areas had an attendance

of 54% (MoHSW, 2015b).

Health System Governance and Accountability

As part of DHFF implementation, the PPHF are authorized to have a facility-level

account for the receipt of all type of funds from both government and non-state

actors. However, opening of these accounts needs authorization from the Bank of

Tanzania (BoT) (Kapologwe et al., 2019). Management of facility funds is done by

the health service providers together with Health Facility Governing Committee

(HFGC) members by following the guidelines for Comprehensive Health Facility

Plans (CHFP) and Comprehensive Council Health Plan (CCHP) (MoHSW, 2011;

United Republic of Tanzania (URT), 2016). The GoT has developed a simple

Facility Financial Accounting and Reporting System (FFARS) which is used as a

guide for public financial management for the public primary health facility

(MoHSW, 2014b). FFARS works alongside other financial systems, such as

Planning and Reporting System (PlanRep) and the LGA accounting software

6

(EPICOR), to aid planning, budgeting, expenditures tracking and reporting as per

schedules (URT, 2016).

Governance can be defined as ‘how societies make and implement collective

decision’ (Saltman, Bankauskaite, & Vrangbaek, 2007), and is reflected in

Sustainable Development Goal (SDG) number 16 (UN, 2015). Governance helps to

enhance accountability and ownership of programs and initiatives (Savedoff, 2011).

The governance structures vary considerably across the globe. For example, in

Kenya, they have Health Facility Committees which are responsible for some

management and governance activities at the primary health facility level.

Governance in the Tanzanian health sector is guided by the existing governance

structures at central and lower levels of government. In Tanzania, health system

governance structures at all levels combine members from both technical and

political sides, which create checks and balances during implementation decision

points. For primary health care, this places governance at the district council level,

and lower level governments, such as wards and villages where the facilities are

based. At the district council level there is a standing committee for social services

with technocrats (composed of heads of departments for health, education,

agriculture, environment, accounts, internal audit, procurement, water, land, law,

human resources and planning) whose roles are to approve plan and report on all

technical issues that are implemented at the ward and village levels, they are also

answerable to councillors. Alongside this committee there are governing entities

from the regional level down to the local level, including the Council Health Service

Board (CHSB) that is composed of members from the community including the

chairperson, the other members include the chairperson to the Social Committee and

Council Management Team (CMT) members. The Secretariat to this meeting comes

from the Council Health Management Team (CHMT) (Kessy, 2014). At the facility

level, there is HFGC that serves a health system governance structure that links the

community’s decisions to that of the primary health facility within the catchment

area.

Accountability in the health system can be defined as “procedures and processes by

which one party justifies and takes responsibility for its activities such as for

7

achieving various organizational goals” (Emanuel & Emanuel, 1996). It is

considered to be one of the intermediate goals of attaining Universal Health

Coverage (UHC) along with transparency and equitable distribution of health

systems resources (Kutzin, 2013; Kutzin, Cashin, & Jakab, n.d.).

The concept of internal accountability mechanisms refers to the relationship within

and between different levels of the health care system. It is important to look into

how key players were involved in the planning and budgeting processes, adherence

to standards of submissions of technical and financial reports, and displays of

revenues and expenditures at the primary health facility level while complying to all

available instructions. The World Bank (2015) found that many HFGC members and

some health facility providers had limited knowledge on planning and budgeting and

few adhered to the budget guidelines (World Bank, 2015a, 2015b).

External accountability mechanisms refer to existing relationship between health care

workers and the communities within and around their catchment areas. The current

study assessed how implementation of the DHFF initiative has influenced the

performance of HFGCs and HSR. The HFGC are mainly responsible for oversight

of the health services provided by the primary health facilities. However, several

assessments conducted in Tanzania showed that they have limited capacities in terms

of their functionality and performance of their assigned roles and responsibilities

(Mayumana et al., 2017; World Bank, 2015b). Moreover, in many primary health

facilities, the complaints handling mechanism, display of revenues and expenditures,

and price list of services were non-existant (Mayumana et al., 2017; World Bank,

2015b).

The way health systems perform vary due to a number of factors which such as

expenditure rates in the health sector, disbursement modalities, and human resources

for health (Murray & Frenk, 2000; Tashobya et al., 2014). To improve health system

performance globally and specifically in low- and middle-income countries (LMIC),

several reforms have been implemented with varying levels of success. Such reforms

include introduction of a prime vendor system in order to ensure a constant supply of

health commodities and supplies, improved community health fund (CHF), and

8

direct health facility financing (Kalolo, Radermacher, Stoermer, Meshack, & De

Allegri, 2015b; Kapologwe et al., 2019; Wiedenmayer et al., 2019). These reforms

have been implemented and evaluated alongside the World Health Organization’s

(WHO) conceptual framework for health system performance assessements (HSPA).

The HSPA are conducted through selected indicators that reflects the performance of

the health care system (Murray & Frenk, 2000; Sun, Ahn, Lievens, & Zeng, 2017;

Tashobya et al., 2014).

The performance of the country’s health care system largely depends on the health

financing modality adopted in a given country (Murray & Frenk, 2000). The health

financing block, which is one of the six WHO health system building blocks (Figure

1) (WHO, 2010), is a cornerstone of health system performance and improvements in

health outcomes of the population. There are two types of health care financing

systems that include: 1) Bismarck model, which is based on social insurance; and 2)

Beveridge model, that is based on the tax collection (Liaropoulos & Goranitis, 2015).

The functions of the health care financing systems include: i) resource mobilization

from different sources such as out of pocket payment, community health financing,

social health insurance, general revenue and donor funding; ii) risk pooling; and iii)

resource allocation (payment mechanisms) [i.e. through global budget, fee-for-

service and capitation] (Kutzin et al., n.d.; WHO, 2017).

Although funds may be mobilized using different sources and ensuring risk-pooling

approaches, paying health services providers (organizations) and how to disburse the

payments presents a challenge (Frumence, Nyamhanga, Mwangu, & Hurtig, 2013).

Adopting innovative payment and fund disbursement approaches can solve these

challenges. The DHFF initiative represents an innovative fund disbursement

mechanism adopted by the GoT since 2017 (Kapologwe et al., 2019).

Direct Health Facility Financing represents a payment and disbursement mechanism

where mobilized funds are disbursed directly to the health facility from any funding

source or the government treasury without going through any other channel. The

objective is to improve efficiency, accountability, transparency, autonomy and

service delivery while also adhering to the financial guidelines, regulations, and

9

laws. Others define DHFF initiative as “the direct provision of government or

external funds to a health facility to meet the operational requirements of the health

facility”(Asian Development Bank, 2016). Direct Health Facility Financing is one of

the disbursement approaches within the health-financing component of the health

system and one of the approaches to facilitate fiscal decentralization. Evolving

evidence indicates that DHFF is linked to improvements in efficient use of resources

and promoting accountability (Kapologwe et al., 2019).

Under the DHFF initiative, funds are directly sent to the facilities instead of going

through a district council’s head office with the aim to address the delays and

misappropriation of funds that was experienced when funds were disbursed through

the district’s office. A setback of this approach is that the CHMTs are not likely to

own the disbursement process as they may feel that they have been disempowered

although upon disbursement of funds to the primary health facilities the Distict

Council also get an exchequer. This may lead to lower perfomance of the DHFF-

related interventions due to unacceptability by managers; hence, there is low fidelity

of implementation.

Given the intertwining and non–lineality of the health system building blocks (see

Figure 1) (Adam & Savigny, 2012), the DHFF initiative, although directly related to

the health financing block, is expected to impact all other blocks, such as service

delivery, human resources, and governance. DHFF is also viewed as a health sector

reform that aims to empower PHC actors by enhancing autonomy in decision making

to improve health system performance (Kapologwe et al., 2019).

10

Figure 1. Health System Building Blocks (Adam & Savigny, 2012)

Health Sector Reform

Health sector reform, defined as ‘sustained, purposeful change to improve the

efficiency, equity and effectiveness of the health sector’ (ANDREW, 1995). Reforms

have ostensibly been implemented globally to improve health system performance

(Berman & Bitran, 2011a; Shewade, 2012). Both LMICs and high income countries

are implementing different reforms and evaluations to improve effectiveness,

efficiency, accountability, responsiveness, and equity in service delivery and

coverage as a strategy to move towards the goal of UHC, particularly as addressed in

goal number three of the SDGs (Gilson & Mills, n.d.; Kruk et al., 2016; Tomson &

Biermann, 2015; WHO, 2016). The reform package for achieving UHC includes

health financing strategies particularly targeting health insurance schemes, user fee

removal or exemption, performance-based financing, and fiscal decentralization

(Kolehmainen-Aitken, 2004; Panda & Thakur, 2016). The WHO considers health

financing reforms to be important in the movement towards achieving UHC, with

such reforms influencing the intermediate UHC objectives of equity in the

11

distribution of health systems resources, efficiency, transparency, and accountability

(Kutzin, 2013; Kutzin et al., n.d.). Many on-going health financing reforms intend to

improve service delivery, especially at primary healthcare facilities given that these

facilities are critical as the gateway to health care for many and health for all.

Reforms are necessary opportunities for continuous improvement of health care

systems while paving the way for continuous evaluation of the on-going

implementation of various interventions for the better health outcome of the

population (Roberts, Reich, Hsiao, & Berman, 2002).

Tanzania, like any other LMICs, has undertaken numerous innovations and reforms

to ensure the health system delivers the best to its people through effective and

affordable approaches. Historically, the health agenda has been the main priority of

the GoT since its independence in 1961 (Semali, 2005; Massoi & Norman, 2009).

This agenda had been implemented through various trials, reforms, and evaluations

to make it responsive to people’s demand (WHO & World Bank, 2012) and has been

reaffirmed through the Tanzanian Vision 2025 document which identifies health as

sectors contributing to a higher quality livelihood for all Tanzanians (URT, 1999).

Improving services at the PHC facilities was one of the priorities after independence

(Semali, 2005). This goal first seriously conceived in 1967 and later it was

strengthened through the Decentralization Act of 1972 (Frumence et al., 2013;

Massoi & Norman, 2009) which led to construction of health posts across the

country with the aim of improving access of health care and equity to the general

public (Massoi & Norman, 2009; MOHSW, 2009). The PHC concept was cemented

by the Alma-Ata Declaration of 1978 that emphasized the importance of the ‘health

for all’ agenda (Rifkin, 2018). More than forty years later, after vigorous reviews and

evaluation, the ‘Health for All’ agenda has slightly changed and been re-affirmed in

the Astana Declaration of 2018. This declaration states that the ‘Health for All’

agenda should be a priority in all settings as the international community move

towards UHC (Walraven, 2019). In the Tanzanian context, primary health care

(PHC) is defined as ‘the essential promotive, preventive and curative health services

12

offered from the community level, Health Centers up to the district level facilities’

(URT, 2007).

Therefore, any investments that seek to promote a high performing health care

system should focus on the PHC system including their reforms and evaluations at

any given point. As the entry point to the health care system for many, primary

health facilities need to offer quality health services and must be responsive to the

people’s needs. In Tanzania, 95% of patients encounter the PHC facility before

moving to the next level (NIMR, 2010). Fortunately, the majority of countries

including Tanzania (Kress et al., 2016) are signatories to a number of international

agreements that promote primary health care, including the SDGs (UN, 2015) and

the Astana Declaration of 2018 which are essential for improvement of health care

delivery at the lower level (Rifkin, 2018; Walraven, 2019). Both agreements require

signatories to have sustainable health financing strategies and strong PHC system

that responds to the people’s needs towards achievement of UHC by 2030 (UN,

2015; WHO, 2013).

The evidence has suggested that primary health facilities need to be social enterprises

to attract patients including women to visit those health facilities when needs arises

so that they can access quality health care services (Addicott, 2011; Chang, Zangle,

& Hunter, n.d.). The ability of primary health facilities to deliver quality health

services requires sustainable and adequate financing coupled with good governance

and accountability (Wang et al., 2015; Macinko & Starfield, 2009). Moreover, PHC

systems, like many other cogs in the machinery of the health systems, should

continuously be subjected to reforms and evaluations so that they can respond to

evolving demands and needs of the general public.

The Tanzanian health sector has passed through several milestones and challenges,

which are reflected in Figure 3. The main challenge has been around health care

financing as an important health system building block. The health system in

Tanzania has not been able to sustain itself due to inadequate domestic tax funding

and over-reliance on the donor communities. In addition, inadequate collection of

user fees and community health funds, as well as delays in receipt of funds as a result

13

of long channels of disbursement have further contributed to low potential of self-

reliance (MoHCDGEC, MoH [Zanzibar], 2016). Following the Structural

Adjustment Programs (SAP) of 1993, the public sector embarked on its Health

Sector Reforms of 1994 which included, among other things, cost-sharing for select

health services which was a significant challenge for the lowest wealth quintile

(Abel-Smith & Rawal, 1992; Brunelli, 2007; Semali, 2005). The on-going litany of

reforms have all been geared towards improving Tanzania’s health sector delivery.

For example, piloting of the Community Health Fund in Igunga District (Tabora

Region) targeted people working in the informal sector so that they can afford health

care (MoHSW, 2015a). Additionally, the introduction of the National Health

Insurance Fund (NHIF) was intended for people working in the formal sector. These

schemes were institutionalised under Acts of 2001 and 1999 respectively (Kamuzora

& Gilson, 2007; Kamuzora, Maluka, Ndawi, Byskov, & Hurtig, 2013a). Another

reform was the introduction of HFGCs in 1999 as responsible for oversight of the

health facility affairs, planning and budgeting, and endorsement of requests and

transactions that are made at the health facility level (Kapologwe et al., 2019; Kessy,

2014; S. Maluka, 2017; S. Maluka et al., 2018). The communities have been

participating enthusiastically and materially towards ensuring availability and access

to health services through construction, renovation, and maintenance of health

facilities through their Health Boards and Health Facility Governing Committees.

Since the year 2000, there has been implementation of several health care reforms in

the area of health care financing. One of the areas which has undergone major

reforms is the Community Health Fund (CHF) which was revamped in 2011 by

improving the minimum benefit packages, introducing regional resource pooling

level, purcher and provider split. These reforms led to rebranding of CHF to be

called ‘improved Community Health Fund (iCHF)’ and is currently being rolled out

across the country. Under iCHF there is a separation of roles between the providers

and purchasers of the health services (Kalolo et al., 2015b; Kapologwe et al., 2017).

Another reform is the movement from input-based financing to results-based

financing. The Tanzanian RBF is a strategic program funded by the World Bank and

USAID launched as a pilot in Kishapu District Council in 2014 and since then it has

been rolled out to eight regions. The program relies on offering financial complement

14

other sources of funds in the primary health facilities based on their achievements

(MoHSW, 2014a; World Bank, 2013). Initiated as a pilot scheme in the Pwani

Region under the name “Payment for Performance (P4P)” through the support of the

Norwegian government, the program was transformed into RBF after having proven

its effectectiveness (Binyaruka et al., 2015; Borghi et al., 2015; Mayumana et al.,

2017; Olafsdottir et al., 2014). The achievements are usually obtained after quarterly

monitoring of agreed qualitative and quantitative indicators that are based on pre-

defined outputs to improve the delivery of health services (Grittner, 2013). The RBF

Program categorizes its resources for development grant for minor renovations and

constructions (75%) and incentives (25%) paid to health care providers with the dual

aim of improving provider performance and increasing utilization of maternal health

services by the local catchment population (World Bank, 2005, 2015b).

The concept of decentralization is derived from Article 8 of the Tanzanian

Constitution of 1977 that stipulates clearly that the government will derive its

mandate from the citizens. Most of the health reforms in Tanzania are in line with

decentralization processes. There have been notable success stories since re-

introduction of Decentralization by Devolution (D-by-D) policy in 1984. The re-

introduction of decentralization policy in 1984 was subsequently suspended for about

11 years following disappointing results due to circumstances such as low capacity

for resource management by the local government officials (Khaleghian, 2004;

Massoi & Norman, 2009). Financial decentralization entailed financial discretionary

powers to local councils, through a mandate to levy local taxes and the entitlement to

receive adequate unconditional and other forms of grants from the central

government. It warrants local councils to plan, budget, and spend financial resources

in the way that reflects their own priorities and complies with the national standards.

Prior to 1998, the government financed the provision of services and local

development through the Regional Administration. The enactment of the Regional

Administration Act No. 19 of 1997 reduced the role of the Regional Administration

and, as a result, some fiscal powers and service delivery functions were transferred to

Local Government Authorities. The amendment of the Local Government Finance

Act No. 9 of 1982 in 2000 redefined sources of revenue for local government

authorities.

15

The DHFF is considered as one of the financial reforms that bypasses funds to be

channeled via the district while sending money directly the health facilities. Under

this arrangemen the district council headquarters are provided with exchequer,

whenever there is disbursement of funds to PHC facilities to help in coordination and

followup. Consequently, there has been an increase in the percentage of sectoral

block grants provided to LGAs over time for education, health, agricultural, water

supply and road sectors. For instance, for the health sector there has been an increase

in spending from 2005 – 2010 (Figure 2).

Figure 2: Estimated health sector expenditures, projections and sources of

funding (million TZS).

Financial decentralization to the public primary health facilities has been the difficult

one due to fact that majority of people at the central are concerned with capacity of

those working at the lower level despite its notably significant impact in terms of

bringing autonomy, transparency, creativity and innovation to the lower levels of

health care provision (Massoi & Norman, 2009; URT, 2001). Prior to DHFF

introduction in 2017/2018, the health sector financial decentralization had just

reached the level of the district council level but not the primary health facility level.

At this level, the District Executive Director (DED) and CHMTs through the

leadership of DMO decided on spending and planning on behalf of the facilities

(Boex, 2013; Boex, Fuller, & Malik, 2015a; 2015b) . This persistent top down

16

approach contributed to delays that affected quality of service delivery at the facility

level in areas, such as availability of health commodities, and minor rehabilitation of

health facilities to improve basic amenities, hygiene, and sanitation (Figure 3). With

DHFF initiative introduction these challenges are expected to be addressed.

Under decentralization policies, it is recognised that the local communities have a

better understanding and knowledge of their local needs and challenges than national

or regional authorities. Hence, local communities are better placed to make sound

decisions and propose innovative solutions to their existing challenges. However,

this challenge is not the case if they are granted autonomy, and equipped with the

authority and capacity to identify problems, to prioritize them, and to plan/manage

all resources at their facility level (Kress et al., 2016; Tsofa, Goodman, Gilson, &

Molyneux, 2017). The community structures, like the HFGC, have the mandate to

deliver tasks on behalf of the communities and they can act as social accountability

representatives.

Boex, et al. (2015a) and MoHCDGEC (2016) both revealed that disbursement of

funds into district council accounts instead of primary health facility accounts created

a loophole for reallocation and misuse of facility funds by LGAs rather than

improving the services delivery of health facilities and frontline workers.

17

Fig

ure 3

: Tan

zan

ia H

ealth

Secto

r Milesto

nes a

nd

Refo

rm T

imelin

es from

1880s to

2019

Source: K

apolo

gw

e et al., 2017.

18

Through fiscal decentralization, primary healthcare facilities are supposed to receive

adequate resources inplanning, management,and using resources; however, many of

these facilities are often underfunded in LMICs (Panda & Thakur, 2016).

Despite an increase in funding, service delivery at public PHC facilities was poor as

evidenced through a star rating exercises of 2014/2015 and 2017/2018. The star

rating approach for primary health facilities is a method to gauge PHC system

quality. Star rating assessment is usually based on the minimum score out of four

domains [(A = Facility management and staff performance (20); B = Service charters

& accountability (30); C = Safe and conducive facilities (20); D = Quality of care &

services (30)]. Zero-star facilities are those scoring less than 20% in any one of the

four domains (Table 1) (Yahya & Mohamed, 2018). There are percentage scores

attached to each grade in order to classify facilities into five grades starting with zero

star rated to five star rated facilities. For the zero-star rating, the scores are between

zero and below 20%. For one star, the score is between 20 and less than 40%. For

two stars, the score is between 40 and less than 60%. For three stars, it is between 60

and less than 80%. For four stars, it is between 80 and less than 90%, and for five

stars, it is between 90 and 100%. The star rating is used to grade all the primary

health facilities in Tanzania (Table 1).

In 2014/2015, fewer than 2% of primary healthcare facilities (131/6,993) met the

desired quality standard of three stars or above. Just over half (51%) of all assessed

facilities were rated one-star, and about one-third were rated zero-star (34%), with

the latter category requiring urgent attention (Yahya & Mohamed, 2018). The

repeated star rating assessment exercise of 2017/2018 findings showed that 456 (6%)

of primary health facilities scored zero; 2396 (33%) scored one star; 3067 (42%)

score two stars; 1276 (18%) scored three stars; 94 (1%) scored for stars and 0

facilities scored five stars (Yahya & Mohamed, 2018).

19

Table 1: Star Rating and Grading Criteria of Primary Health Facilities

Star Status Grade Scoring Criteria

0 <20 A = Facility management and staff performance

(20)

B = Service charters & accountability (30)

C = Safe and conducive facilities (20)

D = Quality of care & services (30)

Zero-star facilities are those scoring less than

20% in any one of the four domains

1 20 -39

2 40 – 59

3 60 – 79

4 80 – 89

5 90 -100

Source: MoHSW, 2014.

In addition to these star ratings, the Service Availability and Readiness Assessment

(SARA) of 2012 and Tanzania Service Provision Assessment (TSPA) of 2015/2016

revealed that the physical condition of the health facility buildings was poor, such as

electricity, piped water, and basic amenities (Kapologwe et al., 2020; MoHSW,

2015b; Tanzania MoHSW, 2013).

Reforms in the Tanzanian Health Financing System

Direct Health Facility Financing

As a result of the above-mentioned challenges, in 2017, the GoT decided to introduce

DHFF in order to ensure flexible and timely funding at the level of service delivery

points to ensure increased efficiency in financial use, accountability, transparency,

and quality of service delivery to the public. The DHFF initiative further aligns with

global health initiatives, such as the (UHC and SDGs).

As part of the DHFF implementation arrangement, the facility is required to prepare

a budget and plan that undergoes rigorous scrutiny. Once responsible authorities

approve the plan, in this case the HFGC, spending by facilities will not be subject to

pre-approval by the District Council. There is a requisite for a certain level of

20

approval for the set financial threshold by the District Executive Director (DED). As

part of financial accountability and compliance mechanisms, all PHC facilities will

be subject to audits by the Internal Auditor General (IAG) and the National Audit

Office as per Public Act Number 11 of 2008 (Opwora, Kabare, Molyneux, &

Goodman, 2010). Training on the DHFF initiative was delivered using a cascade

approach from the national to the community level. As part of preparation to DHFF

initiative implementation, training was conducted in December 2018 with the aim

being to impart knowledge and skill sets to enable health service providers to

implement DHFF accordingly. The diagram below shows every step of the training

and the associated activities (Figure 4).

The training involved teams from Ministry of Health, Community Development,

Gender, Elderly and Children (MoHCDGEC), Ministry of Finance and Planning

(MoFP), and President’s Office-Regional Administration and Local Government

(PORALG) who took part in the masters of training exercise. It also involved

Regional Health Management Teams (RHMTs) and CHMTs. Lastly, training was

conducted with the health service providers and members of the HFGCs. It was a

three days training session that involved introduction to the DHFF initiative, roles

and responsibilities of the DHFF, financial management and issues relating to the

supportive supervision and mentorship. All these events utilized the cascade

approach depicted in Figure 4.

The introduction of the DHFF intiative was expected to affect purchasing

arrangements, because it allows facilities to make local purchases for medical

supplies, employment of casual workers, to pay for health facility operational costs

including paying electricity and water bills, minor infrastructure renovations, and to

provide community outreach services of PPHF in Tanzania. These outcomes were

shown in Kenya’s parallel intervention (Kapologwe et al., 2020; Kapologwe et al.,

2019; Opwora et al., 2010, 2015; URT, 2001).

21

Figure 4: Cascade Training on Direct Health Facility Financing (DHFF) across

Tanzania.

Source: Kapologwe et al., 2017.

Master training on DHFF

Regional Trainers of Trainers

(ToTs)

Instructions were sent to the

Local Government Authorities

on DHFF

Training to Health Facility

Governing Committee on DHFF

Health Service Providers’

Training on DHFF

Local Government Authorities

Training in DHFF

22

PPHF countries, there is increased efficiency in financial use, and there are

transparency and accountability mechanisms instituted at all levels. These should be

accompanied by utilization levels relative to need and quality of services rendered

while ensuring that there is universal financial protection to the service users (Kutzin,

2013; Murray & Frenk, 2000; WHO, 2017).

At this point, DHFF is an approach that is envisaged to set a foundation and platform

for resource mobilization, utilization, and accountability for committed resources at

the PHC facility level. Its main aim is to improve the quality of services delivered to

the people while at the same time advancing towards achieving UHC. Under the

DHFF initiative it is expected that they will be leveraging of all resources at the PHC

levels and accountability of other sources of funds which are generated locally like

the National Health Insurance Fund (NHIF), RBF, Improved Community Health

Fund (iCHF) and user fees (Kapologwe et al., 2017).

Kenya is considered to have well implemented DFF program by starting it as a pilot

intervention in some counties. The Government of Kenya through the devolved

system started to disburse the Health Sector Service Fund (HSSF) to more than 6,000

health facilities with some notable successes (Opwora et al., 2010; Waweru et al.,

2013).

The WHO considers ‘better health’ as the main objective of any health system.

Improvement of health services has two social goals: the goal of ‘goodness’, which

means the health system responds well to what people need; and the goal of

‘fairness’, which implies that the response is equal to all without discrimination

(Musgrove et al., 2000). Both concepts should lead to better health outcomes as they

create an environment for improved access to the essential services at the affordable

cost. However, achievement of goodness in the health system is better off as

compared to fairness. A health outcome is defined as a change in the health of an

individual, group of people, or population that is attributable to an intervention or

series of interventions. Outcome measures (i.e., quality of health services, morbidity,

mortality etc.) are quality and cost targets that healthcare organizations are trying to

improve (Murray & Frenk, 2000; Robone, Rice, & Smith, 2011). Therefore, it is

23

important that, health system should be robust and able to respond to the challenges

that are faced by people with intent to access services at any health delivery point.

District council health funds are usually allocated through pre-determined cost

centers namely CHMT, District Council Hospitals, Health Centers, and Dispensaries.

However, prior to the introduction of DHFF, CHMT received a larger share allowing

them to conduct some of the activities that were initially meant for frontline facilities

(Table 2) (URT, 2016).

After introduction of DHFF initiative there has been a devolution of many tasks that

were initially done by the CHMT on behalf of frontline workers. This has been

accompanied by the reduction of funds allocation from the CHMT cost center to go

to the PHC facilities where most of activities are implemented. However, CHMTs

consolidate all health facility plans and budget to a Comprehensive Council Health

Plan (CCHP) before its submission to the regional level for scruitnization. The

Comprehensive Council Health planning process is conducted at two levels; Health

Facility level and CHMT level, whereby plans from these two levels are later

consolidated to CCHP. Thereafter, the plan is presented to the CHSB, Council

Management Team (CMT), Council Social Services Committee (CSSC), Council

Finance, Administration and Planning Committee (CFAPC) and Full Council (FC)

for endorsement. After endorsement at the Full Council level the plan is submitted to

the Regional Secretariat for assessment, approval, and forwarding to national level.

At the national level, the plan is assessed by national assessors comprised of

members from PORALG-Health and MoHCDGEC and then, if approved, it is

recommended for funding (MoHSW, 2011a; URT, 2016).

24

Table 2: Allocation of funds before and after introduction of DHFF

SN Cost Centre Allocation

Before DHFF After DHFF

1 CHMT 15-20% 15 %

2 Council Hospital /Council

Designated Hospital

25-30% 20 %

3 Voluntary Agency Hospitals

(VAH)

10-15% 10 %

4 Health Centre 15-20% 25 %

5. Dispensary 20-25% 30%

The differences existing in funds allocation in the respective cost centre before and

after the introduction of the DHFF initiative have been shown in Table 2. With

resource allocation after DHFF more resources are directed to health centers and

dispensaries, with combined now both get 55% of the total allocation from the

Health Basket Fund (HBF) compared with a maximum of 45% before DHFF

introduction.

Figure 5: Funds disbursement before and after DHFF implementation.

Source: Kapologwe et al., 2017.

25

1.2 Problem Statement

Since the DHFF’s introduction in Tanzania, there has not been any attempt to

understand its effects on health system performance variables. While there is a

widespread belief that this initiative may solve challenges that exist in the PPHF,

there is no systematic evidence to substantiate this conjecture. Existing evidence on

the financial management system prior to DHFF introduction showed there were

leakage of funds in the process of disbursement to the target primary health facilities

(Boex et al., 2015b), this was coupled by insufficient decisions making at the district

level which resulted into erratic supply of health commodities and other health

system inputs that lead to poor quality of services rendered at the primary health

facilities (Kamuzora et al., 2013a; URT, 2016).

The delays and misappropriation of funds led to the compromise in the governance

and accountability mechanisms at the PHC level, and also led to poor HSR. There

has been some attempt to improve health system responsiveness through health

system reforms such implementation of RBF program that among other things

directed financial resources to the health facilities to address challenges that are

experienced at the lower level. However, these reforms have been implemented in

few (8) regions and they were not evaluated to know areas for improvement.

Moreover, there is limited evidence on the RBF program effects and how major

reforms can improve performance of health systems. This study is needed at this time

to understand whether the introduction of the DHFF initiative had any effects on the

health system performance and addressed problems that existed before its

introduction.

This study therefore evaluated both the processes after 6 months and outcomes of

DHFF 18 months after its introduction in order to understandits performance on

other health systems components. The process evaluation was conducted as it is

important for a better understanding of the ‘black box’ of the DHFF initiative and the

Fidelity of Implementation (FoI). The study also included the outcome evaluation to

understand the effects of DHFF on various indicators of health systems at PPHF in

Tanzania.

26

1.3 Objectives

1.3.1 Broad objective

To determine effects of DHFF on selected health system performance indicators

(HSR, governance and accountability, health service utilization and the structural

quality of maternal health services) and how DHFF is implemented in PPHF in

Tanzania.

1.3.2 Specific Objectives

1. To assess the level of HSR as perceived by patients attending the select PPHF

before and after DHFF implementation.

2. To assess the level of structural quality of MHS in the select PPHF before

and after DHFF implementation.

3. To assess the level of maternal health service utilization in the select PPHF

before and after DHFF implementation.

4. To measure FoI of DHFF and its potential moderators in select PPHF.

5. To measure the association between FoI and overall change of structural

quality of MHS before and after implementation of DHFF.

6. To measure the association between FoI and overall change of maternal

health service utilization in select PPHF.

7. To explore how governance and accountability of DHFF is exercised in the

select PPHF.

1.3.3 Specific Research Questions

1. What are patients’ perceptions on the level of health system responsiveness in

the PPHFselect PPHF before and after DHFF implementation?

2. What is the level of structural quality of MHS in the select PPHF before

anmaternal health service utilization in the PPHFselect PPHF before and after

DHFF implementation?

3. What is the level of FoI and its potential moderators towards DHFF

implementation?

4. What is the association existing between FoI and Structural quality?

27

5. What is the association existing between FoI and maternal health service

utilization?

6. How are governance and accountability of DHFF exercised in the PPHF?

1.4 Significance of the Study

The rationale for conducting this study is informed by the evidence that timely

funding at the level of service delivery can unlock important barriers to effective and

efficient healthcare delivery and contribute to better health outcome for the

population. There has been minimal research on the effects of DHFF on health

system performance.

DHFF is expected to improve the quality of health services including maternal health

that is crucial in Tanzania. It is expected that DHFF initiative will improve

efficiency, foster innovations. And bring more autonomy while promoting

accountability and governance structures such as HFGC.

Findings from this study may inform future designing of interventions and improve

the performance of the current initiative through use of this implementation science

experience. This may also help to strengthen the implementation research discipline

in Tanzania.

The findings may also help to re-design the DHFF initiative in this phase of the

implementation. Will also help to lay down strategies for other health system

interventions of this nature.

Therefore, this process and outcome evaluation was expected to find out the effect of

DHFF on some health system components (HSR, governance and accountability and

structural quality of MHS) and evaluated the processes and outcome by using

selected indicators for health service utilization, structural quality, responsiveness,

knowledge and practice of health service providers and HFGC members. The process

and outcome evaluation (Figure 6) were further achieved through development and

use of a Theory of Change (ToC) and application of the concept of Fidelity of

Implementation (FoI). Other researchers and policy makers who work to improve the

28

DHFF implementation in Tanzania will potentially use the findings of this research

and evaluation.

Figure 6: Opening the ‘black box’ of the DHFF through assessing the

implementation processes.

29

CHAPTER TWO

LITERATURE REVIEW

2.0 Overview

This chapter reviews literature relevant to this study. It starts with the presentation of

the definitions of key theories on health sector reforms and reform cycles, related

theoretical review, and empirical literature review. This chapter aims to reveal what

is known and not known on the topic globally, in Africa, and in Tanzania,

specifically. The second part provides a discussion of the two conceptual frameworks

that guided the conduct of this study, namely: the theory of change (ToC) and fidelity

of implementation (FoI) framework. Finally, the chapter presents a review of

evaluation (Process and Outcome) and a review of the current state of the

implementation of research on complex public health initiatives.

2.1 Theoretical Literature Review

In this study, two related models that help to understand health system performance

and reforms are reviewed. The two theoretical constructs (health system performance

model and health reforms model) are both integrated through the Health Reform

Cycle that has six components to show how DHFF reform was carried out (Figure 6)

(Roberts et al., 2002). These six components are:

1) Problem identification. This is an important step in the health reform process,

whereby the stakeholders identify the health sector problems that need to

have a reform solution. In this study, the main identified problem was delays

in timely disbursement of funds and misappropriation of funds that led to

poor quality health service delivery and lowered the level of autonomy

amongst frontline workers of the PHC facilities.

2) Problem diagnosis. This step enables the identification of the determinants or

causes of the identified health sector problem. To arrive at a diagnosis there

must be exploration of the control knobs that guide the performance of the

health care system. In this regard the control knobs that are used are in the

areas of: financing, payment, organization, regulation and behaviors or

30

persuasion (Roberts et al., 2002). Therefore, the DHFF initiative qualifies as a

major reform as it has incorporated all five control knobs.

3) Policy development. Under this step the health problems were diagnosed

through problem tree analysis and then reform was gauged by using control

knobs. As for this study, a policy meeting that took place in December 2016,

it was used to make resolution for the introduction of DHFF initiative as a

strategy to improve quality of health care services in the primary health

facility level in Tanzania. The performance of the Health Sector is reviewed

on annual basis under Sector Wide Approach (SWAp) by different

stakeholders. The review is conducted to determine Sector’s performance

against HSSP targets.

4) Political will and commitment. This is a critical input for addressing the

health problem or challenge. As for the introduction of DHFF initiative in

Tanzania there were series of meetings from both technical (technical

working group and annual joint technical review meeting) and political levels

(annual joint policy meeting and discussion with social services

parliamentary committee). Both sets of meetings aimed at ensuring political

decisions and commitment to the introduction of the DHFF initiative

(MOHCDGEC, 2016).

5) Implementation of the reform agenda. In this study, the reform is the DHFF

initiative that was implementated from the national to the primary health

facility level. The implementation started in 2017/2018 after the buy in from

all key stakeholders in Tanzania (Figure 4). Implementation started by

cascade training.

6) Evaluation of the implemented reform. Through this stage, we come to

understand how the outcome of interest changes over time. Moreover, this

stage helps to monitor progress of an initiative and also provides room for

modification while exploring necessary factors for sustainability (Roberts et

al., 2002). In this study, baseline and endline cross-sectional studies were

conducted to evaluate the effect of the DHFF on select health system areas. A

midline/midpoint study was done to evaluate processes and also the FoI of

31

DHFF to get a better understanding on how the initiative evolved over a

period of time (Kapologwe et al., 2019).

Figure 7: The health reform cycle (modified from Roberts et al., 2002)

In the next section, a review of theories (health system performance model and

health reforms model) used in this study and their relationship to the health reform

cycle is profferred.

2.1.1 Health System Performance Model (HSPM)

The health system performance model (HSPM) helps to identify the causes of poor

performance of any project/program or an approach and to propose policy changes to

improve performance (Berman & Bitran, 2011a). The HSPM is presented in the four

blocks of Figure 8 with each block is representing control knobs in health system

performance. Reading fromleft to right, the left-most positioned block contains

categories of ultimate and intermediate outcomes that are often used to measure

health system performance. Based on these measures, analysts assess performance

1. Problem identification

4. Political decision -

making

5. Implementation of

DHFF

6. Evaluation

(Process & Outcome)

2. Problem diagnosis

by using control

knobs

3. Policy

development and

resolution on DHFF

32

and identify problems (i.e., poor performance) that must be addressed by health

reforms of an organization or within thepublic sector. The second and third blocks

from the left portrays the analysis of the causes of health system performance and the

formulation of hypotheses. The fourth block focuses mainly on the policy

interventions which are necessary to improve health system performances through

implementation of each control knob, particularly, financing, payment, organization,

regulation and persuasions (FPORP). This DHFF initiative was approved during the

policy meeting of December 2016, and then started its implementation in 2017/2018.

This reform qualifies as major health financing reform as it has all five control

knobs. The model identifies more than one of these control knobs (health financing,

payment, organization, regulation and persuasion) as drivers in achieving major

reforms and performance of health sector of a state (Figure 8) (Berman & Bitran,

2011a).

2.1.2 Health System Reforms Model (HSRM)

Health system reforms are important for any growing health sector that is expected to

meet the people’s expectations. Direct Health Facility Financing is one of the health

financing reforms to be implemented in Tanzania, makes the health systems reform

model important for conceptualizing this study. The process and outcome study of

DHFF used the HSRM to gauge the magnitude of this health reform in which it

qualified as a big ‘R’.

33

Figure 8: Health System Performance Model (Berman & Bitran, 2011a)

Health system control knobs are things ‘that can be adjusted by government action.

And those adjustments or changes in the control knob must be significant causal

determinants of health system performance’(Roberts et al., 2002). Control knobs

describe the discrete areas of health system structures and functions that matter

significantly for health system performance and are subject to change as part of

health system reform.

Looking at the items in the Control Knob Blocks in Figure 8,

1. The financing knob determines the resources which are available to run the

health care system particulary in the primary health facilities. For this study,

funds came from health basket fund, block grants, national health insurance

funds, council own source and improved community health fund (iCHF).

2. The payment knob determines resources that are available to providers. As

for this study, payment knob referred to payments made by Ministry of

Finance and Planning (MoFP) directly to the primary health facilities.

3. The organization knob determines the kinds of provider organizations that

exist and their internal structures that shape how these organizations perform.

In this study, the organizational structures that are responsible for

34

implementation of DHFF are CHMT, CHSB, Health Facility Management

Teams (HFMT) and HFGC.

4. The regulation knob imposes constraints on desired behaviors to allow the

health system to perform. In this study we used circulars, standard operating

procedures and guidelines to implement the DHFF initiative. Direct Health

Facility Financing is governed by existing laws and legal framework present

in Tanzania, including the Constitution of 1977, Budget Act 2015 and its

regulations, annual budget guidelines, Public Procurement Act of 2011 (as

amended 2016) and its regulations of 2013, Local Government Authorities

Tender Board (Establishment and Proceedings Regulations, 2018, Local

Government Finances Act, 1982 (as amended in 2017), Local Government

Financial memorandum of 2009.

5. Finally, the efforts to change behavior knob influences how individuals

respond to health sector organizations, which, in turn, shapes the

opportunities that organizations confront (Roberts et al., 2002). In this study

that was checked through measuring process evaluation by using fidelity of

implementation framework of the DHFF implementers.

The analysis of introduction of DHFF initiative as the health financing reform shows

that; all five control knobs are key components of the DHFF initiative making it a

major reform and also helped in assessing the effects of the DHFF initiative on HSR,

governance, and accountability, and structural quality of MHS in Tanzania.

Hsiao (2000) reform models categorises reforms into two types i.e. ‘R’ and small ‘r’,

and explains that; big “R” reforms are those that are derived from strategic,

purposeful reform programs that introduced change in two or more of the control

knobs, thus affecting health system performance across several parts of the health

system. On the other hand, small “r” reforms may still be strategic and purposeful,

but are more narrowly focused on only one control knob and only one part of the

system (International Health System Group- Harvard School of Public Health, 2000).

Given these definitions, this study examined big “R” reform to explain DHFF

reforms because it touches more than one control knobs. This research studied issues

of current health financing modalities and payment mechanisms used in DHFF

35

initiative. The issues studied included: organizations benefiting from DHFF,

specifically the PPHF, compliance with regulations governing the operation of

DHFF, and behaviours of the health service providers towards DHFF implementation

since its introduction. The success of many reforms depends on the context, design,

policy and fidelity of implementation.

2.1.3 Fidelity of Implementation (FoI)

All policies/programs/initiatives around the globe have a higher likelihood of being

successful and impactful if they are implemented with fidelity (i.e. faithful to the

design). The concept of fidelity has been used to gauge the level of performance

against the design of the initiative and its intended modus operandi (mode of

operation).

Implementation fidelity refers to “the degree to which an intervention or program is

delivered as intended” (Carroll et al., 2007). Only by understanding and measuring

whether an intervention has been implemented with fidelity can researchers and

practitioners gain a better understanding of how and why an intervention works, and

the extent to which health outcomes improved. It is therefore important that fidelity

is measured for any implemented project or program (Carroll et al., 2007). Carroll et

al. (2007) developed a conceptual framework to measure the FoI by allocating a

score based on the adherence to the proposed intervention. The adherence is

understood as all those practices or actions that comply with the set standards.

However, within this conceptualisation of FoI, adherence is defined as whether "a

program service or intervention is being delivered as it was designed or written"

(Mihalic, 2002).

Implementation fidelity helps to establish the relationship between interventions and

their intended outcomes (Carroll et al., 2007). Evaluation of implementation fidelity

is important because this variable may not only moderate the relationship between an

intervention and its outcomes, but may prevent potential false conclusions from

being drawn about an intervention’s effectiveness. Fidelity of any program can be

assessed by conducting a process evaluation (Mihalic, 2002).

36

The concept of fidelity of implementation is currently described and defined in

terms of five elements, namely: adherence to intervention, exposure, quality of

delivery, participant responsiveness, and programme differentiation (Compaoré,

Yameogo, Millogo, Tougri, & Kouanda, 2017; Dusenbury, Brannigan, Falco, &

Hansen, 2003; Kalolo, Radermacher, Stoermer, Meshack, & De Allegri, 2015).

However, according to Carroll et al (2007), adherence is considered to be the

cornerstone to FoI measurement. A policy becomes successful if it adheres to content

and frequency, duration, and coverage, which are collectively grouped as ‘dose’

(Carroll et al., 2007). The content of an intervention is considered as skill sets and

the knowledge that the participants have to deliver the prescribed task.

The score of implementation fidelity can be obtained through measurement of the

adherence to the task that is related to the delivarence of outcomes of interest. Keith

et al. (2010) added contextual factors to other factors presented previously by Carroll

et al. (2007) and it has shown to be influencing the fidelity of implementation of the

programme and its outcomes (Keith et al., 2010). Implementation also offered

multiple insights into contextual factors affecting sustainability of program

implementation (Ojemeni et al., 2017). Despite being a relatively new study area in

Tanzania, there has been an increased interest in measuring FoI of different

initiatives, projects, and program in Tanzania (Mmari et al., 2019).

2.1.4 Fidelity of Implementation (Adherence)

According to Carrol et al (2007), adherence tends to measure who receives an

intervention, and the frequency and magnitude (size) of the given intervention

(Carroll et al., 2007). In this regard, adherence of any initiative can be measured by

looking at how the implementers are abiding to the design while at the same time

looking for amount of dose given and duration with which it was offered.

Adherence to the implementation of the programs or initiatives varies considerably

from one place to the other depending on the quality of delivery which depends on

the awareness and level of knowledge of the implementers (Hasson, 2010; Toomey,

Matthews, & Hurley, 2017). For the DHFF inititative the dose was considered as

funds that are disbursed from Ministry of Finance and Planning to the public primary

37

health facilities across the country. These funds are disbursed through use of

allocative efficiency formular by taking into account the issue of equity for district

councils and public primary health facilities. Therefore, resource allocation formular

takes into account several factors depending on whether is a district council or a

public primary health facility. For district council the following factors are put into

considerations which are: Population (60%): Funds are distributed in proportion to

the population of each district council by 60%; Poverty (10%): Councils receive

additional resources for the special needs of a poor population (10% of the grant

resources); Under-five mortality (10%): The 10% of resources are earmarked to

places with high burden of diseases; the under-five mortality (U5M) is considered as

an appropriate proxy for burden of diseases; Capped land factor (20%): The formula

recognizes the higher expenditure needs of rural areas by directing 20% for the route

mileage regularly travelled by health sector vehicles within their locality and to and

from the district council head office; For that case this formula takes into

consideration the issue of higher operational cost of delivering health services to a

rural population and to sparsely populated areas within Tanzania; this formula also

includes health commodities distribution costs.

Allocation ceiling for Dispensaries and Health centres (30% and 25% respectively) is

combined to provide a pool for lower level health facilities. Parameters used for

allocating the pooled Health Sector Basket Fund and Health Block Grant to health

facilities includes; Catchment Population (40%): Takes into account the population

expected to be served by the Health facility; Distance of the individual

Dispensary/Health Centre to the Council’s headquarter (10%): It is important in

budgeting funds to be used for referral, and logistics requiring health facility staff to

travel to head quarters; Service utilization (40%): Takes into consideration the

intention to improve quality of services. It is measured by a composite indicator, the

“Care unit”; that measures service outputs. The composite indicator consists of six

indicators and is calculated using DHIS2 data as total of (number of attendances x

weight). The six indicators and their weights are: Outpatient attendances – 01,

Antenatal attendances – 01, Institutional deliveries – 06, Postnatal attendances – 01,

Admissions – 10, C-sections – 27; Performance (10%): Takes into account facility’s

contribution towards LGAs performance. It is calculated based on two indicators

38

(use of modern family planning methods and availability of 30 tracer medicine)

using DHIS2 data. Each of the two indicators weight 0.5; Modern Family Planning

users are calculated as percentage of total Family Planning users of public Health

Centres and dispensaries.

On the other hand, duration of funds disbursement from the Ministry of Finance and

Planning to the public primary health facilities is done on quarterly basis for all

health facilities. Therefore, this study focused mainly on the quality of delivery by

looking into the level of knowledge (content), participation of HFGCs and

availability of working tools where as for fidelity of implementation of DHFF, this

study focused mainly on the adherence to the implementation design as dose and

duration was the same for all public primary health facilities.

2.1.5 Participants’ Responsiveness towards DHFF implementation

Several studies have tried to define participant responsiveness. Following review of

13 studies, Schaap et al. (2018) defined participant responsiveness as ‘satisfaction,

appreciation, acceptability, or enjoyment of the programme’ (Schaap, Bessems,

Otten, Kremers, & van Nassau, 2018).

Participant responsiveness is achieved when those who have been assigned to deliver

a certain initiative, deliver it with passion and enthusiasm thereby achieving high

fidelity (Dusenbury et al., 2003; Hasson, 2010). In this DHFF study, participants

were defined to be the health service providers and members of health facility

governing committee (Kapologwe et al., 2019).

Quality of Delivery

Quality of delivery is an important component of the FoI and reflects the way in

which an initiative was delivered according to the set protocols. For this study, the

quality of delivery was assessed by looking into the level of knowledge (content) of

health care workers, participation of HFGCs in decision making processes and

planning and budgeting and availability of working tools especially financial

management tools which are necessary for the implementation of DHFF initiative.

39

Program Differentiation

Program differentiation is ‘the degree to which the critical components of a program

are distinguishable from each other and from other programs’(Dusenbury,

Brannigan, Falco, & Hansen, 2003). Program differentiation can also be referred to

as the process of identifying the critical components of a program that are essential

for producing positive outcomes (Dusenbury, Brannigan, Falco, & Hansen, 2003),

which essentially discriminates between successful and unsuccessful programs. For

this study, the program differentiation was achieved through supportive supervision

and mentorship and also through participator decision making processes by

expanding decision making space.

Strengths of Implementation Fidelity in Initiative Management

Fidelity of Implementation (FoI) helps to measure how an implementation of an

initiative or intervention was achieved and its relationship to the outcomes of interest

(Hulleman & Cordray, 2009; Mowbray, Holter, Gregory, & Bybee, 2003), and

increase external validity of tools that are used for data collection (Stains & Vickrey,

2017). In other words, using FoI improves causal pathway of the program

implementation and its anticipated outcomes while identifying predictors for the

initiative’s success (Donnell, 2015). Stains and Vickrye (2017) outlined that, through

use of FoI, one can improve empirical and descriptive evidences of how and under

what circumstances the program can lead to the anticipated results. It is not easy to

establish the adherence of the program implementation if we do not monitor its

performance. However, it is difficult to establish whether failure to achieve the

outcomes reflects a failure of the model or failure to implement the model as

intended (Carroll et al., 2007; Linda Dusenbury et al., 2003; Mowbray et al., 2003).

2.1.6 Context and Description of DHFF Implementation in Tanzania

The DHFF initiative was launched in 2017/2018; however, actual disbursement and

receipt of funds to the public primary health facilities started in February 2018 across

all 184 district councils of Tanzania mainland. The DHFF initiative was a

government approach to facilitate financial decentralization in the health sector with

40

the aim to improve the quality of health service delivery and autonomy at the public

primary health facilities. Some of the DHFF essential implementation activities

included: 1) training about the program; 2) supportive supervision and mentorship; 3)

employment of assistant accountants; 4) dissemination of facility financing

accounting and reporting systems (FFARS) and guidelines; and 5) opening of bank

accounts that are approved by the BoT. The Regional Health Management Team

(RHMT) and CHMT where mandated to provide technical support and mentorship

on financial management, implementation of annual plans and budget to the health

facilities in accordance with the guidelines.

This study was conducted in PHC facilities (i.e., health centers and dispensaries) and

communities through HFGC as their representative. In Tanzania, health centers are

expected to have mixed set of skilled staff (i.e., doctors, nurses, assistant

accountants) from 39 to 52 workers. At the dispensary level, there is a range of 15 to

20 staff. The introduction of DHFF targeted these facilities to increase efficiency and

accountability and to improve health services provision. Finally, other moderating

factors were supportive supervision and mentorship, inclusiveness in decision-

making, functionality of HFGCs, and operation of BoT approved bank accounts. The

resource allocation for the public primary health facilities considers workload

(Service utilization 70% and health system performance 30%). As for Health centres

and Dispensaries, Service utilization are factored into six indicators (Outpatient

attendances (1), Antenatal attendances (1), Deliveries (6), Postnatal attendances (1),

Admissions (10) and C-section (27). Where as for the Council Hospital performance

are calculated on availability of 30 tracer medicines (0.5) and modern family

planning use (0.5).

41

Figure 9: Direct Health Facility Financing Management Framework

Link between theoretical frameworks and the study

In this study the expected effects of DHFF outcomes are:

1. Provision of structurally quality MHS;

2. Increased HFGC accountability and governance;

3. Increased HSR as perceived in the patient’s experience; and

4. Increased maternal health service utilization.

2.1.7 Implementation Research

Peters et al. (2013) define implementation research as “scientific inquiry into

questions concerning implementation, i.e., the act of carrying an intention into effect

which in health research can be policies, programs or individual practices

(collectively called interventions)”(Peters et al., 2013). The intent is to understand

42

what, why, and how an intervention works in real world settings and to test

approaches to improve it.

Health systems are complex in nature as their blocks are intertwined in nature, so

they need a clear and robust approach in dealing with its evaluations so that to get

clear picture of its processes and outcome of implementation of an initiative or

reform (Adam & Savigny, 2012; Moore, Audrey, Barker, & Bond, 2014). The

interventions that comprise multiple components like the DHFF initiative are

considered as complex interventions and needs holistic and robust approaches to be

successfully evaluated (Moore et al., 2014). Therefore, any design for the evaluation

of any intervention need a good crafting so that they can give results for initiative in

quest. The learnings from such applications are potentially transferrable to other

settings and may contribute to project sustainability. Through use of the FoI

framework, complex interventions will benefit from a rigorous process evaluation

that will enable the opening of a black box. The opening of the black box helps to

understand how the DHFF was implemented in the public primary health facilities.

The essence of the process evaluation is to explore the quality and quantity of the

project implemented and why and how. The inclusion of a process evaluation adds a

measure of accountability and integrity for an organization and individuals

participating in the implementation of the intervention (Moore et al., 2014).

Implementation research uses theories to conceptualize the ideas, therefore applying

a theory is an important component of any health system implementation research

(Brousselle & Champagne, 2011; Rogers, 2008). Theory application of any

intervention usually presents an important step in conducting a robust theory driven

implementation research. A theory of intervention refers to causal pathways through

which change is produced as a result of intervention strategies and implementation

actions (Rogers, 2008). The use of the theory of intervention helps to guide to test

evaluation questions in the course of conducting evaluations.

This study is an example of implementation research, as it studies the effects of an

ongoing initiative relatively recent implementation. Implementation research studies

aim to get a better understanding on what, why, and how an initiative like DHFF

works under normal conditions rather than trying to maneuver or controlling these

43

conditions or remove their influences as a cause and effect (Peters et al., 2013).

According to Peters et al. (2013), some of the studied variables under

implementation research are: fidelity, acceptability, sustainability, and coverage. As

implementation research tends to be conducted in the real world, studying health

systems which are complex in nature align with this approach (Peters et al., 2013).

As the introduction of DHFF has the potential to affect various health systems

building blocks, implementation research is an appropriate method to study the

effects of DHFF initiative on other health system building blocks and the complexity

of the system as a whole.

2.1.8 Evaluation of Health Inteventions or Programs

There have been many approaches that are used to classify different types of

evaluation (Sun, Ahn, Lievens, & Zeng, 2017; WHO, 2012). However, the common

types of evaluation are: (1) formative evaluation - used during the development of

new modification of an existing program; (2) process evaluation that be used to

measure if an initiative in implemented as per design or as per intent; (3) outcome

evaluation - can be used to measure the effect of an initiative or program to the

population in question; and (4) impact evaluation - used to assess whether the

initiative or program is effective enough to attain the ultimate goal.

Globally many studies have focused on the outcomes of an intervention but ignore

the processes that can explain the how and why for the outcomes. The importance of

investing in both the process and the outcome evaluation, it helps to know how the

initiatives are emplemented and the effects evolved from an initiative (Moore et al.,

2014; Nielsen & Randall, 2013). Outcome evaluations do not offer an opportunity to

open the “black box” of an initiative or program hence leading to constrained

knowledge on how the effect of the program or an initiative came into being

(Harachi, Abbott, Catalano, Haggerty, & Fleming, 1999; Nielsen & Randall, 2013).

Having the right frameworks for evaluation can uncover whether an initiative has

failed due to design or due to implementation (Carroll et al., 2007a; Harachi, 1999;

Health, Rychetnik, & Building, 2006; Patton, 2008).

44

The UK Medical Research Council (MRC) defines process evaluation as an

evaluation study that aims to understand the functioning of an intervention by

examining implementation processes, mechanisms of impact, and contextual factors

(Moore et al., 2014). There are several benefits of conducting implementation

process assessments. Four things that make assessment of the implementation

process beneficial are: 1) assists with interprtaton of the outcomes of the

intervention; 2) provides feedback that can help to improve intervention

improvement; 3) helps to replicate an intervention to other settings; and 4) it helps to

appraise the generalizability and the transferability of the intervention (Nielsen &

Randall, 2013).

For the process evaluation focusing on a FoI framework (which mainly focuses on

whether a project is implemented as per design), it can offer more information on the

outcome of interest, as FoI tends to mediate outcomes (Carroll et al., 2007; Hasson,

2010). Process evaluations that use the FoI framework can help to avoid type-III

errors, which falsely occurs due to lack of effect of a given measure to the

intervention rather than due to weaknesses in the course of its implementation

(Carroll et al., 2007; Hasson, 2010)

Research and evaluation contribute to the improvement of health services of the

general population (CDC, 2011). Globally, most countries have been implementing a

number of reforms, projects, and programs with the aim of improving health care

services. However, many organizations or institutions fail to conduct evaluations of

performance of such initiatives (WHO, 2010). Conducting an evaluation of the

initiative or program helps in measuring the performance and also tracks progress of

the implementation, while enhancing the accountability of the said implemented

initiative (Frieden, 2014; Lindeman, 2014; Nielsen & Randall, 2013).

A monitoring and evaluation framework is important when starting a new initiative

or program as it helps in the monitoring of progress towards the set targets or goals

and also offers room for the improvement of the program as it gives necessary

feedback to the decision makers and implementers.

45

2.2 Empirical Review

Most countries are striving to make sure that public sector systems are both effective

and responsive to the peoples’ need. This effort explains why there are a number of

innovations and reforms emerging and being implemented with the aim of improving

improve health service delivery to the public. Most of these reforms aim at

empowering people who are residing at the community level or empowering workers

who are working at the grassroots level. A frequent approach, which has met with

varying levels of success, has been decentralization policy which purport to bring

more autonomy to the lower levels of government and the communities at large.

The empirical review shares findings on the situation of implementation of various

form of reforms on health financing globally and regionally. Also help to understand

essential factors that are necessary for successful implementation of any reform in

any given settings. Thi section will mainly focus on situation of health financing,

HSR, Structural quality of maternal health services, maternal health service

utilization, fidelity of implementation and governance and accountability.

2.2.1 Situation of Health Care Financing Reforms

Health financing reforms are an important ingredient for achievement of UHC by

improving equity and access to the health services. The WHO acknowledges that, in

order to make progress towards achieving UHC, countries should embrace health

financing reforms through development of their national health financing strategies

(Kutzin, 2013). It is important to invest in health financing reforms as health

financing influences achievement of the intermediate objectives of UHC of equity in

distribution of health systems resources, efficiency, transparency, and accountability

(Kutzin, 2013; Kutzin et al., n.d.). WHO has defined three specific functions of a

health financing system, specifically revenue collection, pooling, and purchasing,

each of which may require definition of a reform strategy (Figure 10).

46

-

Figure 10: Health Financing Arrangement, Intermediate Objectives and UHC

goals (WHO, 2013).

The DHFF initiative, being a health financing reform, aligns well with the WHO’s

agenda of health financing reform as it encompasses all three specific functions of

health financing arrangement as seen in the Figure 10. It is also highlighted in the

SDG number three, under target 3c which states that “Substantially increase health

financing and the recruitment, development, training, and retention of the health

workforce in developing countries, especially in the least developed countries and

Small Island Developing States” (UN, 2015).

In SSA, there has been several policy decentralisations which are implemented in

different settings, however very few of these decentralizations is on financial

decentralization. For example, Kenya has been implementing financial

decentralization through local governments since 2008 (Opwora, Kabare, Molyneux,

& Goodman, 2009). The pilot phase was followed by a scale up at the the county

level through devolved government functions. Studies conducted in Kenya have

shown some improvement in terms of financial management and autonomy, quality

47

of health care, health worker staff motivation, and patient satisfaction (Waweru,

Molyneux, Goodman, & Tsofa, 2016). Urban facilities reported water, sanitation

(toilets and cleaners) and minor renovations as their most important uses of funds,

while rural facilities named casual labourers, essential drugs, food, and referrals

(Goodman, Waweru, Kedenge, Tsofa, & Molyneux, 2010; Opwora et al., 2009).

As part of the decentralization process, Ghana has been channelling funds to the

district through regions (Kolehmainen-Aitken, 2004).

The Decentralization by Devolution (D-by-D) policy implementation in the health

sector of Tanzania started in 1970s and was strengthened in the 1990s.

Implementation of these reforms laid foundation for the health sector reforms of

1994 which substantially altered health system financing mechanisms through re-

introduction of cost sharing, soon after initiation of the adjusted structural program

(Frumence et al., 2013; Maluka & Bukagile, 2016; URT, 1999). Under this D-by-D

policy and reform, powers to manage health services delivery at the council level

were devolved to Local Government Authorities (LGAs) and CHMTs. Thus CHMTs

became responsible for budgeting and planning, conducting supportive supervision

and mentorship to the lower level health facilities, and facilitating administrative

activities of the primary health facilities within their jurisdiction (MOHCDGEC

2017).

Under the D-by-D policy implementation, there were still some challenges. These

included: a top-down approach in management on different issues that denied the

frontline workers from exercising their full roles and responsibilities; delays in

disbursement of funds to the PHC facilities from CHMTs; the misappropriation of

some funds; and delays in distribution of medicines and supplies to reach health

facilities (Boex et al., 2015b). Other challenges were inadequate funding, delayed

disbursement from central government, insufficient and unqualified personnel at the

PHC level, and lack of community participation in the budgeting and planning

process. Of all the D-by-D policy achievements, health financing decentralization

has been the most difficult to implement due to worries around capacities and

accountability at the lower level (Fjeldstad, 2001; Panda & Thakur, 2016).

48

2.2.2 Implementation of DHFF initiative in Tanzania

The idea for the DHFF initiative originated from the Joint Annual Health Sector

Planning meeting of December 2016, where participants passed a resolution on the

introduction of DHFF in Tanzania. The decision was based on positive results from

the Results Based Financing that was being implemented in eight regions

(Kapologwe, et al., 2019; Maluka, et al., 2018; World Bank, 2013), as well as

experience from the Ministry of Education and Vocational Training of Tanzania,

that had started to send money directly to the both primary and secondary schools

(Komba, 2012). However, Kenya which is neighbouring Tanzania started

implementation this approach as a pilot intervention in 2008 (Opwora et al., 2009).

Some of the challenges that led to the introduction of DHFF in Tanzania were delays

in disbursement of funds; weak planning and budgeting activities by primary health

facilities; weak supportive supervision and mentorship; failure to carry out minor

renovation; and erratic supply of health commodities and supplies. The detailed

explanations of each challenge are in the following sections.

2.2.2.1 Delays in Disbursement of Funds

Several reports were issued on the delays of the funds reaching the frontline health

facilities challenging expenditures for important health activities (Boex et al.,

2015b). Delays in fund disbursement were observed by World Bank officials when

they were conducting fiduciary system assessments, with showed funds being

delayed up to 16 days enroute to the LGAs. This delay was amplified when it came

to availability of funds at the health facilities level as they were disbursed by LGAs

(World Bank, 2015). In most cases facilities that missed resources for some time

became reliant on other internal sources, such as Community Health Fund (CHF),

National Health Insurance Fund (NHIF) and user fees, to run services at the primary

health facility. Also, the burden for access to medical commodities was transferred to

district hospital where most of the primary health facilities went for support. All

these issues caused unnecessary referrals and poor service delivery at primary health

facilities (Boex et al., 2015b).

49

Figure 11 shows how funds were disbursed from the Ministry of Finance to the

respective health care facilities prior to the introduction of DHFF.

Figure 11: Direct Health Facility Financing (DHFF) Funds Flow and

Stakeholders Relationship in Tanzania (Kapologwe. et al., 2020).

2.2.2.2 Weaknesses in Planning and Budgeting of Activities

This was the common challenge that was presented at the PPHF as they were relying

on the CHMTs for the financial resources to implement their plans and budgets

(Boex et al., 2015b, 2015a; URT, 2016). This made implementation of budget and

plans very difficult. Some prior studies revealed poor accountability in terms of

budget transparency during implementation by the community and involved

individuals, with Tanzania scoring 15 out of 100 points on accountability

50

(International Budget Partnership, 2018). Other reports went further to explain the

minimal engagement of community in decision making especially in terms of the

governance structures like health facility governance committees that endorse

resource use and provide checks and balances (Frumence et al., 2013; Odd-Helge

Fjeldstad, 2010; World Bank, 2015). Community involvement, planning, and

budgeting are considered very important pillars of the DHFF implementation.

2.2.2.3 Weak Supportive Supervision and Mentorship

Supportive supervision from higher to the lower levels is part of health system

governance, as well as supervision within each level, is very important for ensuring

that there is constant provision of quality health services at the primary health care

facility. According to a report from the United Republic of Tanzania - MoHCDGEC

(2016) there was a slight drop of quality of cascade supervision from 64% in 2012/13

to 54% in 2014/15 among LGAs despite the fact that it was noted decrease in the

cost per supportive supervision. Through this cascade supervision superior health

facility is enabled to supervise other facilities in their catchment areas while adhered

to the standards and guidelines. Within the Tanzanian health system, it is expected

that health centers should be able to supervise the satellite dispensaries that are

within the locus or catchment area. With this approach capacities of people at the

health centers to supervise others are built, but, at the same time, it saves costs as the

distance between the supervisor and supervisee is reduce significantly (Olafsdottir et

al., 2014).

2.2.2.4 Failure to Maintain Buildings and Supply Chain of Medicines and

Supplies

Despite funds being disbursed to the district councils, there were a number of minor

activities like maintenance and renovations of infrastructure that were left

unattended. Also there was an erratic supply of medicines and other supplies due to

the inadequate funds being disbursed to the frontline facilities due to delays or

inadequate funding from the district level (Kapologwe et al., 2014; Wiedenmayer et

al., 2015; Wiedenmayer et al., 2019). This problem was also evident in Kenya prior

to the introduction of DHFF (Opwora et al., 2009).

51

2.2.3 HSR in PPHF

Health system responsiveness implies the existence of favourable environment that

promotes health seeking behaviours among people, improves transparency and

communication between heath care providers and users, and ensures adaptation to

quality health information (Njeru, Blystad, Nyamongo, & Fylkesnes, 2009).

This research adapted the domains of HSR as proposed by WHO and various studies

including those done by Mohammadi & Koorosh (2014), Murray & Frenk (2000),

and Njeru, et al., (2009) in measuring responsiveness in PPHF (Kapologwe et al.,

2020). Following a baseline study on HSR, it was found out that several domains did

not performed well some of them are access to care which had a percentage mean

scores of 50% while the overall baseline HSR was 69.1% (Kapologwe et al., 2020).

2.2.4 Structural Quality of MHS

Structural quality is an important aspect of the quality delivery of health services in

any health care system. Donabedian identified and proposed three quality

components for provision of quality of health services in any settings (Donabedian,

1973). Those components areprocesses, structures and outcome (Rourke, 1957;

Westaway, Rheeder, Van Zyl, & Seager, 2003). In the Donabedian model, the

process component examines the process of health care delivery which is an

important component during process evaluation in order to open up the black box

(Gilson & Mills, n.d.; Moore et al., 2014; Westaway et al., 2003). While the outcome

component helps to measure the health care outcomes, some researchers have

criticized this element as it does note independently reveal/explain the processes

which happen between the quality of care delivery and the outcomes (Vesel et al.,

2013; Waiswa et al., 2017). Looking into structure as the third component of the

Donabedian model, we see this component used to assess the infrastructure, medical

equipment, staffing level, and other health commodities. However, Many scholars

considers it, although it has some deficiencies in the sense that it cannot give an

attribution between the quality due to structure and the outcome in the certain

settings (Donabedian, 1973; Khamis & Njau, 2014; Westaway et al., 2003).

52

Tanzania like any other country in sub–Saharan Africa has been trying to improve

the quality of health service delivery to its people through implementation of number

of reforms like star rating program and big results now initiative. In 2012, Tanzania

through the MoHSW came up with the Quality Improvement Framework (2011 -

2016). The aim of this framework is to ensure that there is quality of health services

rendered to the people at each and every level of health care system in the country

(MOHSW, 2009). The publication of the quality improvement framework was

followed by the Service. Availability and Readiness Assessment (SARA) that was

conducted in the 2012. However, some of the shortfalls of this exercise dwelled on

just the assessment of physical status of infrastructure (structure) and never went into

assessing the relationship between the structural quality and processes and the health

outcome (Tanzania MoHSW, 2013). This was rectified by the Service Provision

Assessment (SPA) survey in 2015 which measured the effect of structural quality

and the processes (Kapologwe et al., 2020; MoHSW, 2015b). Both studies revealed

that there was poor quality of services rendered to the people who visit the health

facilities. In Tanzania more than 95% of people reside within 10 kilometres of public

primary health care facilities and more than 90% of people used primary health care

facilities as their first choice when entering the health care system (MoHSW, 2011b;

NIMR, 2010).

Furthermore, the majority of quality improvement interventions were done at the

national, zonal, and regional referral hospitals and less was done to address the

quality challenges at the primary health care levels (MOHSW, 2009; Shoo, Mboera,

Ndeki, & Munishi, 2017). As a result, it became evident that investment at the

primary health care level is vital for the better health outcomes of the majority of

people. Studying the effect of DHFF on the structural quality of MHS is important

because a study in five countries in SSA showed that quality of MHS was found to

be poor in PHC facilities regardless of the efforts which has been put by both

international and country’s effort to improve the MHS (Kruk et al., 2016).

53

2.2.5 Maternal Health Service Utilization

Health service utilization is referring to extent to which people or their

accompanying significant others are accessing and making use of health services that

are available in their communities within and outside their catchment population.

This can be either through admission or through use of the outpatient services.

Service utilization is one of the pre-requisites of any highly working and performing

health care system in any given setting. It can be considered as an important indicator

of performance of system or proxy indicator for quality of health services provision

and HSR by patients. There have been several studies conducted with the aim to

assess health services after the introduction of a certain interventions with some

notable successes (Ojemeni et al., 2017; Ramsey et al., 2013).

There have been several efforts put in place with the aim to improve the service

utilization in several countries around the globe and in Africa at large (Akowuah,

Agyei-Baffour, & Awunyo-Vitor, 2018; Basinga et al., 2011). Tanzania is not

exceptional to other African countries; it has also been trying initiatives that are

geared towards improving the health service utilization, particularly at the PHC level.

One of the programs that were initiated in Tanzania is the payment for performance

(P4P) scheme that started in Pwani region as a pilot in 2011 with the aim to improve

health service delivery by offering incentive packages to health service providers and

also improvement of some structural components of the health care system

(Chimhutu, Tjomsland, Songstad, Mrisho, & Moland, 2015). The scheme never went

into the full scale because in 2015 Tanzania embarked on the RBF program, among

other things, was to improve service utilization of certain health facility services

upon provision of financial incentives (MoHSW, 2015). Most efforts are usually

driven by financial incentives but have been inadequately evaluated for their

sustainability. Hence, studying the DHFF’s effect on health service utilization is

important as it is a systemic approach to addressing issues which may lead to

improvement of health service delivery and increase health service utilization as

hypothesized by Kapologwe et al. (2019) (Figure 7).

54

Some indicators that MHS utilization are not performing well are: post-natal care

within three to seven days; access to modern family planning methods, fourth

antenatal clinic visits; and attendance at ANCs within 12 weeks of (Samky,

2019)gestation (Borghi et al., 2015; Kanyangarara et al., 2017; Kibusi et al., 2015;

Martin et al., 2019; Saxena et al., 2013).

In another qualitative study, which was done in Dar es Salaam, health providers’

disrespect to pregnant women was the main barriers to their booking and attending

ANC visits (Afulani et al., 2020; Bishanga et al., 2019; Mgata & Maluka, 2019;

Sando et al., 2016).

2.2.6 Conceptual Frameworks Guiding the Study

In 2018/2018, DHFF conceptual framework was developed, just before baseline data

collection that was done in January, 2018 after two consultative meetings with

different stakeholders at different levels (Kapologwe et al., 2019; Samky, 2019). It

contains four dependent variables namely; health utilization of health services, HSR,

governance and accountability and structural quality on maternal health services.

Whereas the independent variables were training on DHFF, Assistant accountant

employment, DHFF coordinator, supportive supervision and facility bank account.

The ToC was adopted and modified from other studies conducted in Kenya in 2010

(Goodman et al., 2010; Opwora et al., 2010) and in Tanzania in 2019 (Kapologwe et

al., 2019).

2.2.7 Theory of Change (ToC)

This is essentially a comprehensive description and illustration of how and why a

desired change is expected to happen in a particular context. In the Tanzanian

context we expected that DHFF would induce change in the series of immediate and

intermediate outcomes, ultimately leading to an improvement of the outcomes of

interest (service utilization, HSR, structural quality and governance and

accountability). The theory was also used to establish a causal link between the

DHFF inputs and the expected outcomes. In this study the causal pathways or change

mechanism that will be brought about by DHFF implementation was studied as has

55

been outlined by MRC guidelines on process evaluation for complex interventions

which suggest studying implementations, mechanism of change and context

moderators (Moore et al., 2014).

The conceptual framework for the ToC on DHFF was used as a reference material

for the implementation of the DHFF in Tanzania and elsewhere. Figure 12 below

shows the anticipated changes that occurred as a result of implementation of DHFF

and helps to open the black box. This ToC was conceptualized in two stakeholder

meetings. The first stakeholder meeting was conducted in July 2017, when

participants spelled out the processes of change they anticipated as a result of DHFF

implementation.

A second stakeholder meeting was conducted in August 2017 followed this, in order

to fine-tune the pathways of DHFF implementation (Figure 12).

Looking at Figure 12, the DHFF components column contains the necessary inputs

required for the implementation of the DHFF initiative in Tanzania. The process

column outlines expected program activities to be implemented and monitored for

level of implementation. Intermediate effects are the immediate results/outcomes

that occur after a successful process of implementation of the DHFF, while the

outcomes are the anticipated ultimate results after eighteen months of DHFF

implementation.

56

Training and raising awareness

about the DHFF and related

financial software (FFARS &

EPICOR)

Greater engagement

of HFGC

Appointment of

assistant accountants

Opening of bank

accounts for Health

facilities

Financial planning and

reporting

Receipt of funds

and proper book

keeping

Appointment of district

DHIFF coordinator and

engagement of CHMT

Increased provider

knowledge of DHFF

Strengthened supportive

supervision and mentorship

Investment in

facility

infrastructure, drugs,

supplies

Improved structural

quality of maternal

health services

Greater provider

autonomy over

resources

Greater health system

responsiveness by

patients/clients

Increased Maternal

services utilization by

communities

Increased provider

practice towards

DHFF

Planning and Budgeting

guidelines of health

facilities

DHFF Components P rocess Intermediate effects Outcome

Training on Governance

and accountability to

HFGCs

Figure 12: Theory of Change of DHFF Implementation in Tanzania

Source: Kapologwe et al., 2019.

In this study the FoI framework was embedded along the ToC for evaluation of the

DHFF initiative implementation in the Tanzanian context (Figure 13).

Fidelity of Implementation of DHFF initiative in Tanzania

Figure 13: Conceptual Framework for Fidelity of Implementation

Source: Kapologwe et al (2019), originally from Carroll et al., 2007).

57

From Figure 13, the most important part of the FoI is the potential moderating factor

of participants’ responsiveness which simply explains how DHFF implementers

responds to the implementation. The participant’s responsiveness was assessed

through indepth interviews of the health service providers and members of the health

facility governing committee with regard to the DHFF initiate implementation (or

their level of acceptability towards DHFF initiative). Participant responsiveness can

be defined as ‘the extent to which participants are engaged by and involved in the

activities and content of the program” (Mihalic, 2002). The content in this study is

defined as knowledge or skills obtained following the training on DHFF and

availability of the operational manuals and guidelines for DHFF implementation. The

three-day training using a cascade approach was conducted to health service

providers and members of HFGC on issues relating to DHFF implementation.

Cascade training was done by having national master trainers who trained the

regional master trainers who in turn trained district councils while financial

management using the software tool called Facility Financial Accounting and Report

System (FFARS).

2.2.8 Governance and Accountability

As discussed earlier, some studies, such as Boex et al. (2015b), sought to determine

the resource allocation and flow to expenditure unit proposed that insufficient

financial resources were reaching service delivery units such as dispensaries. Also

Dutta, (2015) and the World Bank (2015a) argued that health financing, especially

at PPHFPPHF, is challenged with late disbursement of funds associated with

violation of existing guidelines. On the part of accountability, studies prove existence

of unrealistic forecasts and weak internal controls (PWC, 2016; Umarji, 2015 ; Sun

et al., 2017). The WHO estimated about 20 - 40% of health expenditures were

inefficiently allocated and therefore had little to contribute towards peoplaaaae’s

health (Sun et al., 2017).

According to the World Bank (2015a) and Waweru et al. (2013), the functionality of

HFGCs is often poor as their members lack the knowledge of their roles and

responsibilities. Moreover, Kamuzora et al (2013a) showed that despite the

58

significant role of HFGCs in managing and supervising service provision at health

facilities, several hindrances affected their performance such that lack of financial

motivation, insufficient time for making decisions, short duration for consultations,

and lack of feedback from the approving authorities (Kamuzora, Maluka, Ndawi,

Byskov, & Hurtig, 2013b; Mayumana et al., 2017).

In the study by Soto et al (2011) looked into the effects of fiscal decentralization on

health care. The study provided empirical evidence to substantiate the importance of

D-by-D policy implementation; moreover, the study acknowledges that the role of

lower level leadership and governance is of great importance for medical

interventions (Soto, 2011).

A critical review that was conducted in Tanzania in 2017 on the decentralization

revealed that for the viability of CHF as a scheme, there is a need for empowerment

of local institutions and health financing decentralization. While local people are

fully aware of local problems, centrally organized power prevented them from using

their initiative in finding solutions. Findings state that CHMT in collaboration with

Council Health Service Board (CHSB), HFGCs and the people at the grassroots can

make CHF succeed by having a clear power relationship among CHMT, CHSB and

HFGCs, while effectively cooperating in achieving CHF laid down foundation for

other programs to be implemented (Mpambije, 2017).

Ongoing interventions in the study area

Since 2013, Tanzania has been implementing a number of programs geared towards

improving health system performance and MHS. One of the programs is RBF which

was implemented in eight regions of the country in which financing of health system

was guided by the pre-defined qualitative and quantitative indicators (Chimhutu et

al., 2015; Kapologwe et al., 2020; MoHSW, 2014a). Another program which has

been implemented with great success is national development and upgrading of PHC

facilities with essential surgical services (Kapologwe et al., 2020). These initiatives

are similar to DHFF initiative as they use a similar modality to disburse funds to the

PHC facilities. The only difference is RBF have a few selected indicators the

59

majority of which are maternal and child health related whereas DHFF included all

indicators needed in the health care system perfomance.

2.2.9 Research or Knowledge Gap

Empirical and theoretical reviews have shown that many countries face challenges of

health financing decentralization. As a result, there are delays in procurement and

delivery of health services in the public primary health facilities.

The effect of DHFF in the decentralized system on the health system performance

remains unknown in Tanzania and elsewhere (Kapologwe et al., 2019; Opwora et al.,

2009, 2015). Therefore, this research and evaluation study helped to add knowledge

on how these types of initiatives can be implemented. The following are gaps of this

study addressed are:

The first gap addressed was lack of information on the effect of DHFF on HSR,

governance and accountability, structural MHS and service utilization. Moreover, the

second gap addressed was the fidelity of implementation towards DHFF addressing

the challenge how projects or initiatives should be implemented under ideal

condition. Most previous studies were done outside Tanzania in the countries like

Kenya and mainly looked at consequences on service delivery and involvement of

community members and health facility management teams; these studies did not

examine FoI (Waweru, Goodman, Kedenge, Tsofa, & Molyneux, 2016).

Other gaps that exists include:

• Limited real-life knowledge on how initiatives like DHFF produce desired

effects in the health care system. This is a new inititiative in Tanzania.

• Limited understanding on how initiatives like DHFF are implemented in real

life settings. Therefore, after carrying out this study helped to understand how

best practices like the DHFF can be implemented in real life.

• Limited literature on governance and accountability in the health care system.

Therefore, this study helped in obtaining information about governance and

accountability in the health care system in Tanzania particularly in the PHC

facilities.

60

CHAPTER THREE

METHODOLOGY

3.0 Overview

This chapter is comprised of the following subsections: study design, research

approaches, study setting (s), study population, sample size determination and

sampling technique. It also covers data collection methods and tools, data analysis,

validity and reliability and ethical issues. The data for this study were collected

through a geographically represented sample of PPHF across the seven zones of

Tanzania mainland.

This thesis follows an implementation science approach and uses a combination of

methods (i.e., crossectional study design, Focus Group Discussion and Indepth

Interviews, before and after non controlled study) so as to understand the

implementation challenges and note the successes (Peters et al., 2013). Therefore,

triangulation of methods was used to get a better understanding of DHFF initiative

from different perspectives. This is important as the effects of DHFF have been

studied against a complex health systems hence research needs flexible methods and

design to produce sound results (Adam & Savigny, 2012).

3.1 Study Settings

This study was conducted in fourteen (14) councils of seven (7) regions (Mbeya,

Shinyanga, Dodoma, Pwani, Mtwara, Manyara and Katavi) from seven (7)

geographical zones (Figure 14). The reason for selecting seven zones was to seek the

country’s geographical representation. The similar approach has been done in major

Tanzanian health studies (Kapologwe et al., 2019; MoHCDGEC, MoH [Zanzibar],

2016). The forteen regions comprised of 42% of the Tanzanian population. The study

was conducted in 42 PPHF that are in 14 district councils namely: Dodoma City

Council and Bahi District Council in Dodoma Region; Kibaha Town Council and

Kisarawe District Council in Pwani Region; Mtwara Municipal Council and

Nanyumbu District Council in Mtwara Region; Mbeya City Council and Rungwe

City Council in Mbeya Region; Shinyanga Municipal Council and Ushetu District

Council in Shinyaga Region; Babati Town Council and Hanang District Council in

61

Manyara Region; and Mpanda Municipal Council and Mlele District Council in

Kitavi Region (Figure 14).

Key:

Dispensary

Δ Health Center

Figure 14: Map of Tanzania that depicts the distribution of District Councils

and PPHF that participated in the study (Kapologwe et al., 2019).

Sources: Shape files were obtained from Tanzania National Bureau of Statistics

(NBS) that was updated in 2016.

The used software was Arc Geographical Information System (Arc GIS) version

10.3. Device used to collect spatial data was Global Positioning System (GPS).

62

Organization of Health System of Tanzania

The health system of Tanzania is organized in a pyramid pattern (Figure 15). At the

base of the pyramid is the community health care services, followed by dispensary

and health centres; these levels constitute the primary health care. These are

followed by district hospitals or designated district hospitals which then are followed

by regional referral hospitals, zonal hospitals, specialized hospitals and finally,

national hospitals. Meanwhile, there are a total of 6,640 dispensaries of which 4,554

(68.6%) are government owned. There are a total of 695 health centres (15.7%) of

which 518 (11.6%) are government owned (Kapologwe et al., 2020; URT, 2016).

The formal distinction between dispensaries and health centres is that while

dispensaries exclusively provide out-patient care, a health centre should be able to

provide around the clock care to patients (URT, 2007).

Figure 15: The Organization of Health System of Tanzania

Source: Researcher’s Construct 2018, modified from MoHCDGEC (2016).

63

3.2 Research Approaches

This research used a mixed method approach employing both qualitative and

quantitative research design. According to Gravetter (2012) and Cresswell (2014),

quantitative approach is for producing quantifiable data that can be analysed using

statistical tests and use of both approaches is aimed at increasing accuracy of

information (Creswell, 2014; Gravetter, 2012). Mixed methods were important

because this study was conducted for the first time in Tanzania and components that

are present in the conceptual frameworks are answered by both qualitative and

quantititave approaches.

3.3 Study Design

This study adopted before and after non-controlled mixed methods (concurrent study

design) employing both qualitative and quantitative with embedded process

evaluation at the midoint of the study. The reason why non-controlled study was

chosen relates to the initiative being implemented nation wide thereby not leaving

any region or district to serve as a control study site.

The pre and post component means that there was one study conducted prior to the

implementation of DHFF initiative and another study was conducted 18 months after

initial implementation of the DHFF initiative, whereas, process evaluation study was

done midway between the baseline and endeline studies. This approach is considered

appropriate for the implementation and evaluation studies (Ary et al., 2010;

Creswell, 2014; Guba, 1981; Peters, Adam, Alonge, Agyepong, & Tran, 2014;

Stuart, Maynard, & Rouncefield, 2017; Wang et al., 2015).

Individual baseline, midline and endline studies employed a cross sectional study

design since the aim was to get a snapshot of the situation and it was less expensive

as compared to other study designs. The baseline study was conducted in February

2018 prior to starting disbursement and use of funds in the primary health facilities,

the midline study in August 2018 (six months after the start of DHFF

implementation) and the endline study was conducted in August 2019. The study

64

units were PPHF, patients exiting facilities, health workers, and members of HFGFs

within their catchment areas.

The baseline and endline studies were used to collect data for the outcome evaluation

study. The midline study was conducted six months (i.e. August 2018) after DHFF

initiative inception and the data obtained was used to appraise the implementation

processes (process evaluation) of the intervention. The study units for process

evaluation were health service providers and members of HFGC.

Outcome Evaluation of Study Components

This was done before and after the implementation of DHFF based on the following

indicators: HSR as perceived by patients attending at the PPHF, structural quality of

MHS, and maternal health service utilization.

Process Evaluation of the Study Components

The process evaluationwas done by measuring the FoI of DHFF initiative at the

midline (Six months after initiation). This approach enabled a consideration of how

the implementation of DHFF initiative was achieved. Moreover, there was an

exploration on how governance and accountability of DHFF initiative were exercised

in the primary health facilities. This was important in opening up the black box

which is usually skipped in most evaluation studies especially those that are health

reform related.

3.4 Target Population

This study population included; health managers, HFGCs chairpersons, patients and

public primary healthcare facilities in the seven regions of Tanzania mainland.

3.5 Study Population

The study population was divided into facilities, in-charges of health facilities,

HFGC members, and exiting patients. Others were members of CHMT from 14

district councils.

65

3.5.1 Inclusion Criteria

A) Patients

1. All exiting outpatients who voluntarily consent from registered PPHF.

2. All exiting outpatients with history of more than 2 times visit at the same

PPHF.

B) Health Care Workers and Members of HFGC

1. All health care workers with more than six months employment in PPHF who

consented. Six months experience implied the experience in budget cycle and

understanding of financial management in PPHF.

2. All HFGC active members within a catchment area with atleast one year

3. HFGCs active members with an experience of more than one year within

PPHF. One year implies experience in the public budget cycle.

3.5.2 Exclusion Criteria

A) Patient

All patients who had less than two visits in the particular health care facility.

B) Health Care Workers and Members of HFGCs.

Health care workers who have worked less than 6 months.

C) Members of HFGC.

HFGCs with members who are not actively involved in the health facilities matters

and those with less than one year of their tenureship.

3.6 Unit of Analysis

The study assessed HSR by examining perceived experience of the exiting patients

after medical consultations in the PPHF. The structural quality of maternal healh

services was assessed by looking into domains of structural elements of PPHF. The

study assessed fidelity of implementation (adherence) of DHFF initiative and level of

content (knowledge) towards DHFF among health care workers working in the

primary health facilities. Further, aspects of governance and accountability were

explored among health workers and HFGC members within their catchment areas.

66

3.7 Sampling Method and Sample Size Estimation

3.7.1 Sampling Methods for Quantitative Data

This study employed a multistage sampling technique for selection of the study units.

The starting point was at the country zones, regions, district and health facilities

(dispensary and health centers). The sampling was predominantly probability

sampling but at some point we also used non probability sampling. The technique is

convenient for studying large and diverse population size (Fawler, 1993;

Molenberghs, 2010). The sampling stages were zones, regions, district councils,

PPHF, health care workers and patients.

First Stage (Zones): All zones were purposively selected from the geographical

zones.

Second Stage (Regions): Randomly selected seven regions out of 26 regions in

mainland Tanzania through table of random numbers. One region from each zone

was selected with each zone having between three and four regions.

Third Stage (District Councils): Two district councils were selected from each

selected region. The sampling frame per region was district councils in the respective

regions. One urban district council was purposefully selected and one rural district

council were randomly selected by using a table of random numbers.

Fourth Stage (Public Primary Health Facilities [PPHF]): The selection of health

facilities employed simple random sampling through use a table of random numbers

from two strata (urban versus rural). Stratification of health facilities based on their

location and homogeneity of the study units (i.e. rural versus urban and health centre

versus dispensary) was done. The sampling frame included the list of PPHF that

obtained from the existing Health Facility Registry (HFR) that is an online platform

used for registration of the public health facilities across the country (www.hfr.go.tz).

At the health centre strata, one health centre was randomly selected and and to the

dispensary strata, two dispensaries were randomly selected by using a table of

random numbers resulting in three facilities in each district council being selected.

This included one health centre and two dispensaries in each district in order to

67

represent urban and rural areas. In this case, 21 (i.e. seven health centres and 14

dispensaries) primary health facilities represented rural areas, and 21 urban facilities

represented urban areas (Figure 16). From each selected public health facilities

service providers were conviniently selected (all those who were present on the day

of an interview) while exiting patients were systematically sampled after medical

consultations based on their gender stratification (Figure 16).

68

Figure 16: Schematic Presentation of Sampling Technique

Random

selection

of Regions

First Stage

One region per

zone

Second

Stage

Stratification of

14 Councils (1-

urban&1 rural

council)

14 Health

centres

28 Dispensaries

Random selection of

both HCs and

Dispensaries by using

a table of random

numbers.

Systematic selection

of patients by their

sex

Systematic selection of 844

exiting patients basing on their

gender on Health System

Responsiveness

14 In-depth interviews to

DMOs, DHFFco, HFi/c and

HFGCs chair.7 FDGs to CHSB

chair, 2 HF i/c, 2 Matrons,

Assistant Accountant and 2

members of HFGCs. 42 Health

Facilities on Structural Quality and Maternal Health Service

utilization

Measuring FOI to 238

participants who were

conveniently selected.

Third

Stage

Stratification

of HFs into

urban and rural

Gender based

stratification

of patients by

their sex

Fourth

Stage

69

3.7.2 Quantitative Sample Size Determination

The study sample size calculation was based on the purpose of this study and the

nature of the population under scrutiny. Thus, referring to Cohen & Crabtree (2008)

sample size of thirty (30) can be considered to be the minimum number of cases if

some form of statistical analysis is to be done (Ary et al., 2010). For this study a total

of 42 PPHF were selected of which 14 were health centres and 28 were dispensaries.

3.7.3 Quantitative Data Collection

The database was developed by using Open Data Kit software (ODK) in which all

the data collected, entered, and then analysed. This study used the mobile data

collection (MDC) approach for quantitative data gathering. Under this approach all

data were collected via mobile phone and uploaded to the central server. Fourteen

(14) research assistants underwent four days of training on mobile data collection

skills and techniques before taking part in the pre-testing of their acquired skills in

health facilities outside the study area. The data collected were sent to the researcher

by using the ODK platform (Figure 17). In Tanzania all primary health facilities

have GPS coordinates. Therefore, all selected study primary facilities had GPS

coordinates and all data enumerators used tablets with GPS sensors to enhance data

integrity.

70

Figure 17: Schematic Presentation of Data Collection Processes

Source: Reseacher’s construct (2017).

3.7.4 Sampling Methods for Qualitative Data

In qualitative research, there are no standardized rules for sample sizes, although six

to eight participants often suffice for a homogeneous sample as they share similar

characteristics in terms of their work shedules and assignments (Cohen & Crabtree,

2008). The technique of data saturation was used to determine when additional

interviews were required, as no additional interviews upon saturation due to

repetition of ideas and concepts.

A total of seven focus group discussions (FGDs) were purposefully selected which

included the following participants namely: CHSB chairperson, HFGC member,

health facility in charges (one from dispensary and one from health centre), matrons

(one from dispensary and one health centre) and account assistants. Each FGDs had a

total of seven people who explored the level of how governance and accountability is

71

exercised in relation to DHFF initiative implementation at the PPHF. Interviews

lasted from one to one and half hours.

To understand the participant’s responsiveness towards DHFF initiative

implementation, twenty-eight in-depth interviews were planned to be conducted

however only 14 were conducted after having reached a saturation point with

representatives from the following participant groups: DMO, DHFF coordinator, and

health facility in charge, and HFGC chairperson to explore their responsiveness and

acceptance to the DHFF initiative.

3.7.5 Data Collection Methods and Tools for Qualitative

The principal investigator and research assistants used semi - structured interview

guide to conduct indepth interviews, qualifications of other data enumarators were

degree holders with more than five years of field experience, currently are working at

the University of Dodoma (UDOM) as tutors. Intervew was done by the principal

investigator. Prior to the interview there wa as a pre-test study that was done on how

best to collect qualitative data while ensuring data quality and integrity interview

guide collected qualitative data from the study sites. The qualifications of other data

enumarators were degree holders with more than five years of field experience,

currently are working at the University of Dodoma as tutors. The indepth interviews

were done in three district councils.

Whereas for the FGDs, principal investigator accompanied by two research assistants

with qualitative data collection skills conducted FGDs in seven district councils. The

moderation was done by principal investigator as he has a solid understanding of the

topic under discussion, the research assistants were responsible for taking field notes

and also recording the conversations.

Prior to the FGDs and IDS a pre-test done on how best to collect qualitative data

while ensuring data quality and integrity. The indepth interview guide was comprised

of 21 questions derived from conceptual frameworks (ToC and FoI) and categorised

into areas that measured effects of DHFF initiative (i.e., participants’ responsiveness

towards DHFF implementation and adherence).

72

The guide for the FGDs included 16 semi structured questions that were designed to

measure two aspects namely governance of DHFF initiative and accountability of

DHFF initiative in the PPHF and was also derived from conceptual frameworks and

literature reviews. The interviews were conducted at the public primary health care

facilities. During the FGD, field notes were taken simultenously with recording of

conversations. One member of the research team was designated to monitor

recording and a second member took responsibility for taking field notes.

Objective 1: To assess the level of HSR as perceived by patients attending the select

PPHF before and eighteen months after DHFF implementation.

3.8 Sample Size Determination

Sample size (n) for patient’s HSR was obtained from the following formula

(Cochran, 1977).

Where:

n = Minimum sample size

Z = Standard normal deviation of 1.96 corresponding to 95% confidence interval.

P= Is an Estimated proportion of positive health system responsiveness as perceived

by paients attending at PPHF. Since there is no published data from previous studies

available for similar circumstances a proportion of 50% was adopted.

= Degree of accuracy of the results, was 0.05.

n=384.

73

The sample size was 384 patients. The 10% of sample size was added (n= 38) to

cover for refusal to participate, making a total of 422 patients.

3.9 Data Collection Tools

A closed-ended Likert scale questionnaire was adopted and modified from the health

systems responsiveness questionnaires used in WHO multi-country studies to study

patients’ experiences during their encounters with the health care system (Appendix

1) (WHO, 2015; Naghavi et al., 2015; Robone, Rice, & Smith, 2011). There were 37

responsiveness questions that were grouped under the seven domains which have

ordinal response categories.

The 37 questions were divided among the seven domains of responsiveness: prompt

to attention (7), respect to dignity (3), clear communication (7), quality of basic

amenities (11), confidentiality (3), access to care (4) and autonomy (2) (Table 3).

The internal consistency of the overall responsiveness scale (37 items) was measured

for Cronbach’s alpha (Nunnally & Bernstein, 1979) which had an average of 0.827

for all eight domains. For each primary health facility, 10 exit interviews were

conducted with the patients who had already received curative health services and

who had made two or more than two visits to the respective facility. In consideration

of gender balance, patients were stratified basing on their gender, the calculated

sample was divided into two (50% male and 50% female patients).

Systematic sampling was used to select exiting patient in which every 3rd exiting

patient was requested to consent to participate in the study, an average of 10 patients

participated per primary health care facility.

74

Table 3: Scoring Criteria/Cut-offs for HSR Performances

Res

pon

se D

om

ain

(n

o.

of

qu

esti

on

s)

Res

pon

ses

Cate

gori

zed

Neg

ati

ve

Min

imu

m P

oss

ible

S

core

(A)

Min

imu

m %

Sco

re

(A/C

)*100

Cu

t –off

Sco

res

for

Neg

ati

ve

HS

R (

b)

%

cut-

off

fo

r N

egati

ve

HS

R

(B/C

)*100

Res

pon

ses

Cate

gori

zed

Posi

tive

HS

R

Maxim

um

sco

re (

C)

Attention (7) Never (0)

Sometimes (1)

Often (2)

0 0 14 66.7 Very Often (3) 21

Dignity (3) Never (0)

Sometimes (1)

Often (2)

0 0

6 66.7 Very Often (3) 9

75

Communication (7) Never happens

(0),

Slightly Often (1),

Often (2)

0 0 14 66.7 Very Often (3) 21

Autonomy (3) Very big (0),

Big (1), Average

(2)

0 0 6 66.7 No problem

(3)

9

Access to care(4) Above 30 minutes

(1),

Within 30 minutes

(2)

4 25 8 50.0 Few minutes

(3),

Instantly (4)

16

Confidentiality (3) Never happens

(0),

Slightly Often (1),

Often (2)

0 0 6 66.7 Very Often (3) 9

76

Basic amenities (10) Strongly disagree

(1),

Disagree (2),

Agree (3)

10 25 30 75.0 Strongly

Agree (4)

40

Total Health System

Responsiveness (37)

14 11.2 84 67.2 125

77

Objective 2

To assess the structural quality of MHS, this study adopted tools from the Results

Based Financing (RBF) program (Appendix X) that has been implemented in eight

regions of Tanzania and in more than 48 African countries including Tanzania

(World Bank, 2013). MHS data were collected in 42 selected PPHF in seven regions

of Tanzania mainland.

Data Collection tools: used an adapted and modified observational checklist

adopted from RBF program for data collection. The RBF program is implemented in

eight regions of Tanzania mainland and more than 48 countries of Africa continent.

The checklist on the tool were selected based on the existing challenges particularly

structural quality of MHS in Tanzania. (Mapunda et al., 2016; MoHSW, 2014;

Ojemeni et al., 2017; Wiedenmayer et al., 2015). The checklist was pilot tested and

modified prior to data collection exercise to enhance for rigor and validity of the tool.

Objective 3

Using nine select indicators (Appendix XI), the study assessed the maternal health

service utilization in 42 PPHF. These indicators were extracted from DHIS-2

platform. The following indicators were assessed:

o Institutional/ health facility deliveries.

o Intermittent presumptive treatment for malaria second dose (IPT-2)

o ANC mothers on mebendazole (De-worming).

o Number of women attending the fourth ANC visit.

o Number of HIV infected pregnant women who are receiving antiretroviral

treatment (ART).

o New users of modern family planning methods.

o Proportion of PPHF with availability of all 30-tracer medicines.

o Number of mothers receiving post natal care within three to seven days after

delivery

o Pregnant mothers attending their first ANC before 12 weeks.

78

All of the above indicators are present in the health management information system

(HMIS), which is a standard tool for collecting health records across all public health

facilities in Tanzania. The selected indicators are considered to be important in the

safe motherhood initiative (MOHCDGEC, 2016) and also are part of health system

indicators. The HMIS is divided into paper-based tools (MTUHA books) which

consist of 15 registers, tally and summary sheets, and an electronic base referred to

as the District Health Information Software (DHIS-2) (Musau et al., 2011). The

DHIS-2 and health facility HMIS registers were used to assess maternal health

service utilization indicators in the selected 42 PPHFin seven regions of Tanzania

mainland.

Data Collection Tools: After collection of data from the DHIS-2 platform,

compilation was completed on a pre-developed Microsoft Excel™ sheet

(https://hmisportal.moh.go.tz/hmisportal/#/ ).

Objective 4 (Midline)

This objective aimed at measuring the level of FoI towards DHFF initiative. To

achieve this objective, the tool was developed to assess fidelity of implementation

(Appendix XIII A). Reliability test was done, and the tool was found to have a

reliability of 0.812. Six domains with a total of 42 questions were assessed for their

adherence (fidelity): 1) Training on DHFF and FFARS; 2) Coordination of DHFF

activities; 3) Governance and Accountability; 4) Financial management; 5) Planning

and Guidelines; and 6) Supportive Supervision and Mentorship (Appendix 5a). Other

components assessed werecontent (knowledge) and participants’ responsiveness.

Sample size estimation for quantitative data

Sample size (n) for content (Knowledge), and FoI (adherence) of DHFF initiative

was obtained by administering the questionnaire to 238 health care workers who

were conveniently selected on the day of the data collection at 42 PPHF.

79

Sample Size Estimation for Qualitative Data

For the participants’ responsiveness towards DHFF initiative, a total of 14 in-depth

interviews out of 28 planned indepth interviews were conducted to purposefully

selected participants by virtual of their management and leadership roles in the

primary health care system (CHMTs, Health Facility Incharges and HFGC

members). The intent was to get the understanding of participant’s responsiveness

towards DHFF initiative implementation. Initially 17 participants were contacted for

an interview; however, only 14 consented with three declining due to previous

commitments. The saturation level was reached on the 11th participant; however, we

carried out 3 more interviews for a total of 14 interviews. The interviews were

conducted face-to-face and the interviewees were contacted via phone and others on

face-to-face prior to the interview. The interviews were conducted with DMOs,

district DHFF coordinators, primary health facility in charges, and chairs of HFGCs

for each of the seven purposefully selected district councils (Appendix XIV). The

selection of participants based on their implementation roles and position in the

health system. Each nterview lasted between 30 to 60 minutes.

Data Collection Tool to Measure FoI

General information about DHFF initiative implementation, knowledge (content) and

adherence was obtained after administering questionnaires to health service providers

and members of HFGC. The implementation fidelity was measured through

adherence and its sub-categories, which are content (knowledge) and its moderating

factors namely, participant’s responsiveness towards DHFF initiative, training, and

status of HFGC, supportive supervision, assistant accountants, and inclusiveness in

decision making. Moreover, there was assessment of contextual factors.

Fidelity of Implementation (FoI): A total of 40 questions were asked of health

workers to determine whether the health service providers implemented DHFF

initiative as planned. The adherence with DHFF implementation was used to decide

the FoI framework through the cut-off points (high or low).

80

Knowledge (Content): Health care providers were assessed for content by checking

their level of knowledge on DHFF and whether they received training and any

supportive supervision and mentorship. A total of 14 questions for health workers

and five questions for HFGC related to the processes of DHFF were asked. The

questions had weighted points. The mean scores were computed and those who

scored above the mean value (>9.286) were graded as having adequate knowledge

(content) while those scoring below the mean score were categorized as having

inadequate knowledge (content) of DHFF initiative.

Moderating factors: The FoI (adherence) was then assessed for association with

some of the moderating factors, namely participant’s responsiveness, training on

DHFF, supportive supervision, and HFGC existence and presence of assistant

accountants. Levels of fidelity were interpreted as previously reported in the

literature, with ≥ 80% adherence interpreted as ‘high’ fidelity, and <80% as ‘low’

fidelity (Bellg et al., 2004; Borrelli, 2012; Nurjono et al., 2019; Toomey, Mathews,

& Hurley, 2017).

Qualitative Data Analysis

A total of 14 in-depth interviews were audiotaped, transcribed for verbatim, and

anonymized for analysis. Thematic analysis was used to analyse the data of DHFF

implementers leading into thematic analysis. The analysis began with data

familiarization through listening/re-listening to audiotapes as well as reading/re-

reading of transcripts. Three researcher assistants independently coded the material

using NVivo QSR™ version 12 which was used to help to organize the codes, and a

fourth researcher reviewed the coded material, categories, sub-themes and themes to

establish key findings. The Nvivo QSR™ software developed a coding tree by

following all steps from setting up a project in the software then importing files of

interview transcripts, followed by assigning codes to the imported data. These codes

were grouped together to formulate categories, sub - themes and then themes. For

simplification and easiness of readability, processes from formation of codes,

categories, sub-themes and themes were summarized in the Coding Table (Appendix

XI).

81

The final steps were to link all codes into categories and categories into pre

determined themes. Truthworthness was maintained through requesting another

persons (supervisors) to review the entire qualitative research process (audit trial)

and the data analysis so that to ensure that the obtained findings were consistent and

could be dependable and transferable elsewhere.

Process Evaluation of DHFF Implementation

Apart from the research conducted, the process evaluation of selected indicators was

done by using ToC and FoI to measure performance of DHFF initiative 18 months’

post-implementation. The purpose of the process evaluation was to open the ‘black

box’ by understanding how the DHFF initiative was implemented and how its effects

came into being. The schematic presentation of the process evaluation, which was

carried at the midline, as shown in Figure 18.

Study duration -18months

16/02/20 18

End line

(August, 2019)

Baseline

(Feb, 2018)

Process Evaluation:

Six (6) months

Figure 18: Schematic Presentation of Process and Outcome Evaluation of the

DHFF Implementation

82

The process evaluation undertook an on-going descriptive and mixed methods

assessment of the process of DHFF initiative implementation at the public primary

health facilities, documenting the role and perspectives of key stakeholders at each

stage and at each level of the implementation to understand and open the black box

to better comprehend the ‘how’ (Dunn et al., 2015; Moore et al., 2014). Moreover,

this evaluation intended to examine the degree to which implementation was carried

out by health service providers, if they adhered to design and theory of change and

fidelity of implementation frameworks. Process evaluation data collection was done

at the midline, six months after DHFF launch in the 14 district councils of Tanzania

mainland. A FoI framework with the moderating factors was used to conduct the

process evaluation.

Outcome Evaluation of DHFF Implementation

The outcome evaluation measured the effect of DHFF after 18 months of the DHFF

initiative implementation. It involved data collection at baseline and endline. The

variables assessed were structural quality of MHS and health service utilization. Also

it assessed HSR as perceived by patient’s experience after medical consultations. The

study used the ToC framework with pathways that were hypothesised at the baseline

as stipulated in the Figure 17 and also in the protocol paper (Kapologwe et al.,

2019).

Fidelity of Implementation for DHFF initiative

According to Carroll et al. (2007b), adherence includes knowledge (content),

coverage, frequency and duration, while moderators includes intervention

complexity, facilitation strategies, quality of delivery and participants’

responsiveness. In the current intervention, the adherence was measured against how

health service providers in the PPHF implemented the intervention according to the

context. The context of this intervention includes availability of skilled professionals

(assistant accountants), operational guidelines, and HFGC, along with training and

existence of support systems like Facility Financial Accounting and Reporting

System (FFARS). These factors allowed for effective and efficient implementation of

the DHFF intervention. The degree to which the intervention is implemented defines

83

the degree of FoI. However, moderators of the FoI, such as complexity of the

intervention, strategies kept in place to facilitate it, and participants’ responsiveness

to the initiative (in this case health service providers) influence the overall

performance of the intervention (Carroll et al., 2007).

In measuring coverage, the current intervention focused on the extent of the

availability of assistant accountants in primary health facilities especially the health

centers; availability of operational guidelines and financial management tools and

use of FFARS; health care providers trained on the DHFF initiative process; and

availability and activeness of HFGCs in assuming their roles and responsibilities.

Moreover, the frequency and duration of the DHFF implementation was assessed

based on the adherence to DHFF initiative and reporting of both financial and

technical perfomances.

Not every component of adherence may be mandatory in an implementation as an

intervention can be successful when core or essential components are implemented

(Carroll et al., 2007). The current research was intended to identify the essential

components for successful implementation of DHFF iniative in order to improve the

effectiveness and efficiency of implementation of DHFF iniative.

In the course of implementation of the DHFF initiative, the potential moderators for

successful implementation were assessed. This research adopted the potential

moderators from Carroll et al. (2007b) as indicated in Figure 19. Intervention

complexity was assessed through availability and understanding of operational

guidelines by health service providers as proposed by (Century, Rudnick, &

Freeman, 2010). The evidence shows that detailed guidelines are more easily adopted

and implemented than vague ones. Also there is a higher chance of achieving fidelity

in simple than complex interventions (Carroll et al., 2007; Pérez, Stuyft, Zabala,

Castro, & Lefèvre, 2016).

The GoT has maintained facilitation strategies to ensure there is sufficient

implementation of DHFF in PPHF. These strategies include development and

dissemination of operational guidelines and financial management tools, training of

health service providers, supportive supervision and mentorship, employment of

84

account assistants in health facilities, appointment of DHFF coordinators at council

level, and use of FFARS as a tool for reporting both financial and performance

reports (Kapologwe et al., 2019).

These strategies may potentially moderate the degree of FoI performance; hence, the

more done to help implementation, through monitoring, feedback, and training, the

higher the potential FoI (Carroll et al., 2007). However, of note, the facilitation

strategies do not guarantee better FoI implementation, but rather reflection of

adequacy in quality of delivery. One of the major factors that necessitated this

research study was to establish the potential moderators that escalated the

performance of FoI in the current DHFF intervention.

Finally, participant’s responsiveness as viewed in the context of intervention

relevance to implementers contributes to low or high levels of FoI (Carroll et al.,

2007). The acceptability of the benefits of DHFF to health services providers bares

significant influence on the FoI performance. When health service providers

perceived high benefits from an intervention during their service provision there is

likelihood that adherence to the standards set for implementation will be high.

Therefore, in establishing the level of participants’ responsiveness, the current study

assessed the acceptability of health service providers towards DHFF implementation.

However, this moderator is not only assessed on the individuals receiving the

intervention but also on those responsible such as the DHFF coordinators at council

level. This study was conducted as a part of process evaluation in order to measure

the FoI amongst health service providers and HFGC committee members as a part of

community representation towards DHFF initiative implementation after it started.

Following the midline study on FoI, there were modifications/additions of some

moderating factors that were not in the hypothesis of the FoI framework (Figure 19).

Figure 19 shows that potential moderators are DHFF training, availability, and

functionality of the public financial management tools, participants’ sex, availability

of assistant accountants, inclusiveness in decision making, attitude of health care

workers, enabling environment and location of PPHF are important factors for FoI

towards DHFF initiative.

85

Levels of fidelity were interpreted as previously reported in the literature, with ≥

80% adherence interpreted as ‘high’ fidelity, and < 80% as ‘low’ fidelity (Bellg et

al., 2004; Borrelli, 2012; Nurjono et al., 2019; Toomey et al., 2017).

Figure 19: Modified Conceptual FoI Framework after the Midline Study

(Modified from Original Framework by Kapologwe et al., 2019).

Potential Moderators

• DHFF Training

• Supportive supervision

• Active HFGC

• Availability of operational

guidelines

• Availability and functionality of

PFM tools

• Participants sex

• Availability of account assistant

• Inclusive decision making

• Location of health facility

• Enabling environment

• Planning and Budgeting

• Attitude of Health workers

Adherence

• Context

• Coverage

• Frequency

• Duration

• Dose

Outcome: FoI

towards DHFF

Intervention:

DHFF

86

Objective 5 (End line)

Data collection

Qualitative: To explore how the Governance and Accountability of DHFF were

exercised in the PPHFs, we used an internally created interview guide (Appendix

XII) to conduct FGDs to explore the participants’ insights. The participants were

selected based on their roles and responsibilities in regard to the implementation of

DHFF initative. The seven FDGs were conducted with groups of seven participants

from each region. Each group was composed of the CHSB chairperson, one member

of HFGC, two-health facility in charges (one from dispensary and one health centre),

matrons (one from dispensary and one health center) and account assistants. Each

FGD lasted for an average of one to one and half hours. There was one person who

was designated for recording and another for taking field notes. The FGDs were

conducted by principal investigator accompanied with 2 research assistants with

qualitative data collection skills. The moderator was one of the research assistants

who was conversant with moderation skills and the topic under discussion.

Saturation was reached when there were no new ideas emerging from the discussion

or conversations.

3.10 Pre-testing of Tools

Pre-testing of tools was done in Kongwa – Dodoma region at the selected health

facilities where the study did not take place. This helped in modification of tools by

droping and adding some questions so that to improve data collection tools.

3.11 Data Collection Procedure

The researcher collected data daily with 14 trained research assistants using the

mobile data collection (MDC) system. The interview guide for qualitative data

collection and structured questionnaires were both translated into Kiswahili

language, as it is the national language and widely used (Appendix 1, 2, 3, 4, 5a &

b). The collected data was then sent to the researcher where it was checked and

verified for its relevance to enhance the quality of the collected data and internal

consistency. Some of the data were collected directly from the DHIS-2 platform.

87

Two research assistants accompanied the researcher to all regions for collection of

qualitative data.

3.12 Data analysis

3.12.1 Quantitative data analysis

3.12.1.1 Variables and their measures

3.12.1.1.1 Dependent Variables

i. HSR

ii. Structural quality of MHS

iii. MHS utilization

iv. Implementation fidelity of DHFF initiative

v. Governance and accountability

3.12.1.1.2 Independent Variables

i. Socio-demographic characteristics

ii. Content (knowledge) of DHFF

iii. Participants’ responsiveness of DHFF

iv. Training on DHFF

v. Supportive supervision

vi. Existence of HFGCs

vii. Location of health facilities (urban vs rural)

viii. Staffing level per health facility

ix. Inclusiveness in decision making

x. Enabling environment

The above variables are summarized in Figure 20 below:

88

Figure 20: Summary of Study Variables

3.12.1.2 Categorization of Variables

3.12.1.2.1 Categorization of Health System Responsiveness

Health system responsiveness as perceived by patient experience was assessed and

scores were computed. The health system responsiveness was analysed basing on the

primary health facilities user’s experiences as shown in the four-point Likert scale

(Table 4). Each point of the Likert scale was in percentage and the answers were

then dichotomized for further analysis; for example, ‘good’ and ‘very good’ were

categorized as ‘positive health system responsiveness (HSR)’ whereas ‘bad’ and

‘very bad’ were categorized as ‘negative health system responsiveness (HSR)’. The

Likert scale rating for each domain was matched with the responsiveness

performance categories as ‘negative HSR’ and ‘positive HSR’ (Table 4). For

instance, the corresponding code for response for basic amenities domain was four,

which was multiplied by ten (the number of questions in the domain) to produce a

89

cut-off score of 40 and yielded a maximum score “positive” for an individual for this

domain (Table 3). One multiplied by 10 (number of questions) and cut off points for

negative HSR was 30 (Table 3). This approach is similar to one that was used in

another study conducted in Ethiopia in 2017 (Yakob & Ncama, 2017).

A total of 37 questions were included to assess health system responsiveness in the

primary health care facilities in Tanzania. The assessement was done by using likert

scale questions. The likert questions had weighted scores that were increasing from

negative to more positive statements and they were converted into numerical values.

The questionnaire had four-point Likert scale question items ranged from zero to

three for five domains (attention, dignity, communication, autonomy and

confidentiality). Zero represented absence of the assessed feature of HSR and three

denoted the highest level of its availability. Any score below three was considered to

be a low score (Negative HSR) and it was hence a cut-off point for the respective

domains (Table 4). On the other hand, one to four points were used for two domains

(access to care and quality of basic amenities) with a score of four indicating the least

performance of the assessed HSR for access to care and 10 for basic amenities

(Table 3). In total, a minimum of HSR score was computed as 14 out of the

maximum score of 125 for all 37 questions (Table 3). Total score for each domain

was computed in percentage by taking the actual score (Column (B) obtained from

each respondent divided by the maximum possible score (Column C) multiplied by

100% (Table 3). Similarly, the overall HSR score was computed by dividing the

overall total cut-off (84) scores over the maximum possible value (125) multiplied by

100%.

The score of the respondents was used to categorize the perception towards HSR into

two major categories [Category I= Positive HSR and Category II= Negative HSR]

(Appendix 1).

90

Table 4: Health Care Responsiveness Performance Criteria and their

Categorization

Domain Number of

questions

Min-max

score(s)

Negative HSR Positive

HSR

Prompt attention 7 0-21 0.0-66.7 66.8-100

Respect for dignity 3 0-9 0.0-66.7 66.8-100

Clear communication 7 0-21 0.0-66.7 66.8-100

Respect of autonomy 3 0-9 0.0-66.7 66.8-100

Access to care 4 4-16 25.0-50.0 50.1-100

Respect for

confidentiality

3 0-9 0.0-66.7 66.8-100

Quality of basic

amenities

10 10-40 25.0-75.0 75.1-100

Overall responsiveness 37 14-125 11.2-67.2 67.3-100

Upon reliability testing, each domain of the health system responsiveness scored a

Cronbach’s alpha of greater than 0.7 (Table 5) indicating the tool was reliable.

Table 5: Reliabilities of each Domain of HSR

Items Number of items Cronbach’s Alpha

Prompt to attention 7 0.892

Respect to dignity 3 0.818

Clear communication 7 0.825

Autonomy 2 0.821

Access to care 4 0.822

Confidentiality 3 0.895

Basic amenities 10 0.817

91

The second step was to conduct the inferential statistics. The basic assumptions for

normality tests were conducted. Visual inspections of histogram Q-Q and box plots

(graphs) were done. Visual inspection of histograms indicated that dependent data

distribution was along the straight line for Q-Q plots for both dispensary and health

centre, whereas symmetry was observed on box plots for both dispensary and health

centre. In addition, skewness and kurtosis z-values and Shapiro-Wilk test for

dependent variables were conducted. Skewness and kurtosis z-values were within the

range of -1.96 to +1.96 for both dispensary and health centre (dispensary =.0599,

.543 and health centre =.245, .543). Shapiro Wilk test showed p ≥ .005 (p= .694 and

.828 for dispensary and health centre, respectively). These findings indicated that

data were approximately reasonably normally distributed. Therefore, parametric tests

for inferential statistics were considered relevant for performance comparisons. A t-

test was used to compare the means of the two groups (before and after).

A multivariable logistic regression analysis model was used to explore the

predictive power of each independent variable on the dependent variables,

specifically demographic information, and perceived responsiveness by the patients.

Multiple regressions were also used to assess the power of predictors for institutional

factors (e.g. staffing level, number of beds) with perceived responsiveness.

3.12.1.2.2 Categorization of Structural Quality of MHS

Using seven domains, the study assessed health facilities’ structural quality of MHS.

The cut-off point that was used was a mean score percentage of 60. This percent has

also been used by Gilson et al (1995) and in the star rating of health facilities in

Tanzania (Gilson, Magomi, & Mkangaa, 1995; Yahya & Mohamed, 2018). Mean

score was computed and those facilities that scored above the mean value were

graded as having high structural quality, while those facilities that scored below the

mean score were categorised as having low structural quality.

Scores of the respondents was used to categorize the respondents’ levels of structural

quality of maternal health services into two major categories [Category I= high

structural quality and Category II= low structural quality].

92

3.12.1.2.3 Categorization of Health Service Utilization

1. Institutional Deliveries

2. Use of IPT2

3. ANC mothers on Mebendazole (De-worming)

4. Number of women attending the fourth ANC visit

5. Number of HIV infected pregnant women receiving ART

6. New users of modern family planning methods

7. Proportion of PPHF with all 30 tracer medicines

8. Number of mothers receiving post-natal care services within three to seven

days

9. Pregnant mothers attending their first ANC below 12 weeks

This study used nine indicators relating to maternal health service utilization to

assess health service utilization at the PPHFs level. Mean score was computed and

those facilities that scored above the mean value were graded as having high health

service utilization, while those facilities that scored below the mean score were

categorised as having low health service utilization.

The scores of the respondents were used to categorize the respondents’ level of

knowledge into two major categories [Category I= high maternal health service

utilization and Category II= low maternal health service utilization].

3.12.1.2.4 Categorization of Fidelity of Implementation (FoI)

Fidelity of implementation was measured by using adherence towards DHFF

initiative. A total of 40 questions for health workers were asked to know whether the

health service providers implemented DHFF initiative. The adherence towards DHFF

implementation was used to decide the FoIframework through the cut-off points

(high or low). Levels of fidelity were interpreted as previously reported in the

literature, with ≥ 80% adherence interpreted as ‘high’ fidelity and <80% as ‘low’

fidelity (Bellg et al., 2004; Borrelli, 2012; Nurjono et al., 2019; Toomey et al., 2017).

The score of the respondents was used to categorize the respondent’s level of FoI

into two major categories [Category I= high FoI and Category II= low FoI].

93

Knowledge (Content): Health care providers were assessed for content by checking

their level of knowledge on DHFF and whether they received training and any

supportive supervision and mentorship. A total of 14 questions for health workers

and five questions for HFGC relating to what is done during DHFF were asked. The

questions had weighted points. The mean score was computed and those who scored

above the mean value of ≥ 9.286 points were graded as having adequate knowledge

while those scoring below < 9.2 points were categorized as having inadequate

knowledge of DHFF. Scores of respondents were categorized into two major groups

of knowledge categories: Category I= adequate content level and Category II=

inadequate content level.

3.13 Data Processing and Statistical Analysis

Data Processing and Quality Control

Data were collected and rechecked for completeness and consistency. Data from the

Open Data Kit (ODK) were extracted by using computer excel file, cleaned, and then

entered into the Statistical Package for Social Science ™ (SPSS) version 25 for

revised coding and analysis (for quantitative data).

3.13.1 Statistical analysis

Statistical analyses in this study used the following approaches: 1) univariate analysis

to describe the sample and distribution of the outcome and explanatory variables; 2)

bivariate analysis (logistic regression analysis) to establish the relationship and

association between variables; and 3) multiple regression analysis to establish

whether the explanatory variables could predict the outcome variables and to control

for confounding variables.

Descriptive statistics were used to generate frequency distribution, and cross

tabulation was used to describe the characteristics of the study participants. The chi

square test (χ2) was used to compare groups (categorical data) and odds ratio (OR)

with 95% confidence interval were computed and used to determine the strength of

association among variables. The statistical significance level was made at p= 0.05

(2-tail) as a cut-off point.

94

Forward stepwise logistic regressions analysis was done to establish whether the

predictor variables were independently associated with outcomes (dependent

variable) of interest. Also, it was done to control/adjust for any confounders. In the

logistic (binary) regression models the association between predictors and outcomes

was measured by odds ratios (ORs) and adjusted odds ratio (AOR) with the 95%

confidence intervals (CIs).

Multiple regression models were used to determine the variables that could predict

the four variables namely HSR, maternal health service utilization, structural quality

of MHS, and FoI. The comparisons within groups (Baseline/Endline) were estimated

using independent t-test for HSR, whereas paired t-test was used for structural

quality of MHS and Wilcoxon ranked signed test was used for maternal health

service utilization.

3.13.2 Governance and Accountability

In alignment with the objective number 7, thematic analysis was used to analyse data

that came from audio-recorded conversations and field notes that were transcribed

into transcripts. After familiarization, the transcripts were then imported into NVivo

QSR™ version 12 where themes and categories were developed.

3.14 Qualitative Data Analysis

For qualitative data, the extracted data from digital recorders were transcribed

verbatim and then thematically analysed using NVivo QSR™ version 12 and similar

codes and categories were grouped to form themes which was then grouped together

for meaningful interpretation and reporting. Quotations were used to support findings

of each theme.

Thematic Analysis

Two independent researchers carried out analyses of data from FGDs and IDIs based

on the original transcripts. The analyses utilized the thematic analysis method (Gale,

Heath, Cameron, Rashid, & Redwood, 2013) and were assisted by NVivo™

software (QSR-international) version 12. This method was selected due to its

95

flexibility and its step-by-step interconnected stages that guide the analytical process.

Previous research (Gale et al., 2013; Smith & Firth, 2011) emphasised the rigor and

transparent nature of this approach as demonstrated by its use of the matrix structure.

The approach is not aligned to any epistemological view point or theoretical

approach so it can be adapted for use in inductive or deductive analysis or a

combination of the two (e.g., using pre-existing theoretical constructs, deductively

then revisiting the theory with inductive aspects and vice versa). In addition, our

research approach was mainly descriptive and aligns itself to the pragmatic paradigm

makes the framework method a suitable approach for analysis (Smith & Firth, 2011).

Analysis followed the following steps: 1) verbatim transcription of audio recordings;

2) familiarization of recordings and field notes with the interview/FGDs; 3) labelling

and coding by underlining the key information; 4) forming categories; 5) formulating

sub-themes and themes; and 6) interpreting the themes. These steps are summarized

in the coding table that shows processes from coding, categories and finally themes

(Appendix XII). Other themes which emerged upon analysis were also incorporated

as part of the findings.

3.15 Integration of Quantitative and Qualitative methods

As this study used a mixed methods approach, it was therefore very important to

integrate findings from the two data sets. As quantitative findings had more weight

than the qualitative findings, we explored the convergence point in the objectives

(number one of the midline study) and the objective number two. Good Reporting of

A Mixed Methods Study (GRAMMS) was used to guide intergration process for

qualitative and quantitative processes (Kaur, Vedel, El Sherif, & Pluye, 2019;

O’Cathain, Murphy, & Nicholl, 2008). We looked if there were any complementarity

of information or any contradictions to each other. Integration allowed us to develop

a composite, holistic, and cross-validated picture of the reality based on results from

both quantitative and qualitative datasets. This process of integration took place after

completion of data analysis and it entailed identifying similarities and differences,

merging the results, and discussing the meaning of the integrated results. Figure 21

provides the integration steps of both.

96

Figure 21: Schematic Presentation of Integration of Mixed Methods

Triangulation Design for Measuring DHFF Fidelity of Implementation (FoI)

3.16 Trustworthiness and Credibility

3.16.1 Credibility: interpretation was achieved through data triangulation (FGDs

and indepth interviews), and supervisors checking repeated interviews. Also,

participation of research assistants’ familiar with the community and culture

potentially minimized threats to the truth-value and contributed to an accurate

presentation of the findings explored within the cultural context. The supervisors

helped in the quotes selection and making sure all protocols observed so that to

enable audit trail/confirmability audit.

Quantitative data

collection

Qualitative data

Collection

Quantitative data

analysis

Qualitative data

analysis

Integration of the Quantitative and

Qualitative Results

97

3.16.2 Dependability: in this study, was achieved through making sure that the

findings were consistent with methods used and repeatable. This was achieved

through cheking findings gathered by principal investigator by another reader

(supervisors) and all procedures were followed accordingly.

3.16.3 Transferability: is the degree to which the results of qualitative research can

be generalized or transferred to other contexts. As for this study, qualitative data

were collected from seven different regional settings making its findings possible to

be used and applied in another context.

3.17 Validity and Reliability

3.17.1 Validity

Validation of instruments was ensured through a researcher by consulting to the

supervisors and other researchers who have used the tools like that before.

3.17.2 Reliability

The pre-test was done in one of the primary health facilities under Dodoma City

Council that was not included in the actual study data collection exercise. The aim of

the pre-test was to identify how long the questions were taking to complete, to

identify the difficulties in questions, and to establish whether the instructions for the

questionnaire were understandable. Findings of this study were used to modify the

tool. Ambiguous questions were clarified and corrected to make the questions easily

understandable to the research participants.

98

Table 6: Data Analysis Plan and Measurement of Variables for Baseline and Endline Study

Specific

objectives

Dependent

variables

Independent variables Measurement Analysis technique

Objective

one

Health system

responsiveness

as perceived by

patient

experiences

Socio-demographic

characteristics

Training on DHFF

1. Supportive

supervision

2. Number of staff

per facility

3. Location of the

health facility

4. Health service

utilization

5. Fidelity of

implementation

The 37 questions were divided among the

seven domains of responsiveness: prompt to

attention (7), respect to dignity (3), clear

communication (7), quality of basic amenities

(11), confidentiality (3), access to care (4)

and autonomy (2). The internal consistency of

the overall responsiveness scale (37 items)

was measured for Cronbach’s alpha

(Nunnally & Bernstein, 1979) and had an

average of 0.827.

The questions in each domain were divided

into four groups:

Ordinal variables range 0 (never) to 3 (very

often) Ordinal variable ranging from 0 (very

big problem) to 3 (no problem)

Ordinal variable ranging from 1 (Waited for

long time) to 4 (serviced instantly) Ordinal

ranging from 1(strongly disagree) to 4

Descriptive statistics;

frequency of scores

distribution and mean &

SD

Cross tabulation for

checking differences of

scores among the

participants.

Independent t test

Regression analysis for

checking associations

among the variables.

99

(strongly agree)

Objective

two

Structural

quality of

maternal health

1. Socio-

demographic

characteristics

2. Training on

DHFF

3. Supportive

supervision

4. Number of staff

per facility

5. Location of the

health facility

7-domain observational checklist.

The domains were: 1) hygiene and sanitation,

2) waste management, 3) maternal audit, 4)

privacy, 5) sterilization equipment, 6)

obstetric emergency medicines, supplies and

equipment, 7) delivery room

Descriptive statistics

(mean, SD)

Cross tabulation to

compare groups’

performances

Paired Sample t Test

Regression analysis for

checking statistical

significances among and

between the participants

Objective

Three

Health Service

Utilization

1. Socio-

demographic

characteristics

2. Training on

DHFF

3. Supportive

supervision

4. Status of HFGCs

5. Number of staff

per facility

6. Location of the

1. Institutional deliveries

2. Consumption of IPT2

3. ANC mothers on Mebendazole

(Deworming)

4. Number of Women attending the 4th

ANC visit

5. Number of HIV infected pregnant

women receiving ARTs

6. New users of modern family planning

Descriptive statistics

(mean, SD)

Cross tabulation to

compare groups’

performances

Wilcoxon signed rank

test analysis for checking

statistical significances

among and between the

participants

100

health facility

methods

7. Proportion of PPHF with all 30 tracer

medicines

8. Number of mothers receiving post

natal care services within 3-7 days

9. Pregnant mothers attending their 1st

ANC below 12 weeks

Paired sample t-test for

comparing means

between the baseline and

end line surveys

Objective

Five

Governance and

accountability

10. FGDs interview guide Thematic analysis

101

Table 7: Objective number 4

Specific Objective Dependent

variable

Independent

variables

Measurement Analysis technique

Measuring fidelity

DHof implementation

for FF

FoI Adherence towards

DHFF

40 items questionnaires for health

service workers

Questions were divided into six

domains: 1) Training on DHFF

and FFARS, 2) coordination of

DHFF affairs, 3) Financial

management, 4) planning and

budgeting, 5) supportive

supervision, and 6) mentorship.

- Descriptive statistics on frequency,

mean & SD

- Cross tabulation to compare

adherence of the participants.

Logistic regression analysis

Participant’s

responsiveness

(acceptability)

towards DHFF

- 21 items semi - structured

interview questions

-Thematic analysis

102

3.18 Ethical Consideration

Ethical approval for this study has been granted by the University of Dodoma

(UDOM) ethical clearance committee and endorsed by National Institute of Medical

Research (NIMR) (Ref. No. NIMR/HQ/R.8a/Vol.IX/2740) (Appendix XVI).

Permission to conduct the study in the respective regions and districts was granted by

the President Office - Regional Administration and Local Government Authority

(PORALG), Regional Administrative Secretariat (RAS) and District Executive

Directors (DEDs). Oral informed consent was undertaken before the administration

of the questionnaires to the study participants (Appendix XV). Participants were able

to withdrawal from the study at any time. Confidentiality was guaranteed for the

participants, and only identification numbers (ID) were used to identify participants,

not names.

103

CHAPTER FOUR

RESULTS

4.0 Introduction

This chapter reports the main findings of the study. The report is arranged into sub-

sections to facilitate easy capture of the main findings in alignment with each

objective of the study.

The first sub-section is descriptive statistics of the study participants. The sub-section

contains the socio-demographic characteristics of health service providers, members

of health facility governance committee, DMOs, District DHFF Coordinators,

Council Health Service Board members and patients.

The second sub-section contains descriptive statistics results for each objective

covering the frequencies and percentages of the response from the study participants.

The third sub-section presents the actual performance/level/status of the objective

results. These includes level of health system responsiveness, level of

implementation fidelity, level of structural quality of maternal health service and

level of maternal health service utilization in PPHF.

The fourth sub-section presents the inferential statistics from each objective. This

part portrays factors attributing to the change in perfomance of the studied objectives

including change in health system responsiveness, structural quality of MHS and

maternal health service utilization before and after implementation of DHFF

initiative in PPHF.

The results are variably presented in the forms of percentages, graphs, figures, and

tables for simple identification/highlighting and ease of understanding of the

employed quantitative data analysis procedures.

104

4.1 Perception of Patients on Health System Responsiveness (HSR) in PPHF

4.1.1 Socio-Demographic Characteristics of Patients in PPHFs

This study included a total sample population of 844 patients who participated in an

exit interviews before (422) and after (422) study respecting HSR study. Of 844

patients (100% response rate) surveyed, 422 were surveyed before and the remaining

patients were surveyed after eighteen months of the study. Table 8 shows the

distribution of socio-demographic characteristics of respondents. The age of patients

ranged between 19 and 88 (mean, 37; SD ± 14) years. Most (36%) of the respondents

were in the age group of between 25 and 35 years; both sexes were equally

represented. Nearly two-thirds had primary education (63%) and 68% resided within

five kilometres of a PPHF. The majority (71%) had visited primary health facilities

more than five times for the past five years (Table 8).

Table 8: Social Demographic Characteristics of the Patients in PPHF (PPHFs)

before and after implementation of Direct Health Facility Financing (DHFF)

(n=844)

Before DHFF (n=422) After DHFF (n=422)

Variable Frequency Percentage Frequency Percentage

Region

Dodoma 62 14.69 59 13.98

Pwani 60 14.22 61 14.45

Mbeya 60 14.22 59 13.98

Katavi 60 14.22 64 15.17

Manyara 58 13.74 63 14.93

Shinyanga 62 14.69 55 13.03

Mtwara 60 14.22 61 14.45

105

Facility type

Dispensary 290 68.72 263 62.32

Health center 132 31.28 159 37.68

Location of the facility

Urban 219 51.90 219 51.90

Rural 203 48.10 203 48.10

Sex

Female 211 50.00 211 50.00

Male 211 50.00 211 50.00

Age group (Mean)(SD)

(Range)

37(14, 15-88)

≤24 80 18.96 81 19.19

25-35 150 35.55 155 36.73

36-44 91 21.56 68 16.11

≥45 101 23.93 118 27.96

Marital status

Married 308 72.99 281 66.59

Cohabiting 28 6.64 36 8.53

Single 65 15.40 75 17.77

Divorced/Separated 12 2.84 14 3.32

Widow/Widowed 9 2.13 16 3.79

Highest level of education

Never went to school 57 13.51 48 11.37

Primary 266 63.03 264 62.56

106

Secondary 69 16.35 79 18.72

Tertiary education 30 7.11 31 7.35

House hold size

(mean)(SD) (range)

5(2, 1-20)

<5 299 70.85 170 40.28

>5 123 29.15 252 59.72

Distance to the health

facility

Within 5 kilometres 284 67.30 288 68.25

5 - 10 kilometres 89 21.09 117 27.73

Above 10 kilometres 49 11.61 17 4.03

Frequency of attending

service delivery on the

same health facility

(Median and IQR)

Twice 51 12.09 30 7.11

Thrice 41 9.72 19 4.50

Four times 33 7.82 34 8.06

Five times 28 6.64 39 9.24

More than five times 269 63.74 300 71.09

4.1.2 Status of Health System Responsiveness (HSR) as perceived by patients in

PPHF (PPHFs) before and after implementation of Direct Health Facility

Financing (DHFF)

The perception of HSR among patients in PPHF was measured by seven domains

namely; prompt to attention, respect to dignity, clear communication, respect of

autonomy, access to care, respect for confidentiality, and quality of basic amenities

107

(Appendix I). These domains were assessed on a four - point likert scale ranging

from zero to three in five domains and one to four points in two domains as

explained in the methodology section (Table 4). The independent t-test was used to

determine significant change in perceived HSR among patients before (422) and after

(422) implementation of DHFF in PPHFs. The descriptive analysis revealed that the

overall mean score of HSR before implementation of DHFF was 62.84, SD and

range were 64.21, 10.05 respectively. Subsequently, upon implementation of DHFF,

the percentage mean score on the perceived HSR was 71.20, SD was 8.05 and range

stood at 48.87 to 99.25. Furthermore, the results portray that the percentage mean

difference among domains experienced a positive shift except for clear

communication and respect for confidentiality. The domain of respect of autonomy

had a higher mean difference (8.86) followed by quality of basic amenities (8.17) and

access to care (6.83) (Table 9). The overall responsiveness showed a significant

change from 62.84 to 71.20 (p<0.0001, 95% CI) percentage mean score upon the

implementation of DHFF.

Almost all regions showed change on HSR upon the implementation of DHFF except

for the Katavi region which showed no shift after the introduction of the DHFF

initiative as seen in Figure 22.

108

Figure 22: Regional pefromance on structural quality of MHS before and after

implementation of Direct Health Facility Financing (n=42).

109

Table 9: Two Sample Independent T-test for Perception of Patients towards Health System Responsiveness before and after

implementation of Direct Health facility Financing (DHFF) in PPHF (PPHF) (n=844, p=0.05)

Domain Patients’ perceptions

before DHFF

implementation

(Mean [95% CI])

Patientss

perceptions after

DHFF

implementation

(Mean [95% CI])

Mean

difference

T-value P-value

Respect of autonomy 76.74[74.18, 79.30] 85.60[83.89, 87.30] 8.86 5.67 <.0001

Quality of basic amenities 62.30[61.16, 63.45] 69.07[68.26, 69.88] 8.17 9.51 <.0001

Access to care 48.61[46.63, 50.59] 55.44[53.85, 57.02] 6.83 5.30 <.0001

Respect for dignity 78.95[76.82, 81.07] 82.02[80.28, 83.75] 3.07 2.20 0.0282

Prompt attention 79.40[77.48, 81.32] 81.01[79.47, 82.55] 1.61 1.28 0.2000

Clear communication 74.64[72.31, 76.98] 61.22[59.74, 62.69] 13.43 -9.58 <.0001

Respect for confidentiality 87.64[85.01, 90.26] 85.70[84.05, 87.36] 1.93 -1.22 0.2210

Overall responsiveness 62.84[61.88, 63.80] 71.20[70.42, 71.97] 8.36 13.33 <.0001

110

The descriptive results on the domain of respect for dignity has shown a significant

percentage change of patients who were shown courtesy and affection during service

provision by health care providers. Although clear communication had negative

change (Appendix III).

Furthermore, the current study has revealed that respect of autonomy had a

significant performance change upon the implementation of DHFF initiative

(p<.0001, 95%CI). The indicators in this domain showed that there existed a

potential change on the freedom of choice in health facilities from 47.87% to 72%.

Also, 73% of patients acknowledged the existence of adequate access to physicians

upon the implementation of DHFF as compared to the previous period (67%).

However, more than half (58%) of the patients interviewed in this research indicated

that there was limited freedom to choose services they prefer from physicians

(Appendix V).

Access to care was perceived to have lower change than other domains with

percentage mean difference of 6.83 (p<.0001). Given the overall perception of

patients the indicators have shown the potential improvement from the baseline data

towards end line, since the number of patients awaiting medical consultation

spending more than 30 minutes dropped from 49% to 34.6%. Also, patients in the

waiting roomfor more than 30 minutes decreased from 32% to 18%. However,

patients who spent more than 30 minutes for laboratory services rose from 10% to

18.9% (Appendix VI).

4.1.3 Factors Associated with change in overall perception of patients on Health

System Responsiveness in PPHF before and after implementation of Direct

Health Facility Financing.

In determining factors influenced perception patients towards health system

responsiveness, the cut off points for positive perception and negative perception for

overall health system responsiveness was first determined by assigning minimum

and maximum mean score. Negative perceptions of patients ranged between the

score of 11.2 – 67.2, while the positive perceptions ranged between 67.3 – 100 scores

(Table 4). The reliability was attested by a Cronbach’s alpha greater than 0.8 (Table

111

5). Moreover, there was an introduction of one variable namely time that was

categorized into two groups (i.e., before and after implementation of DHFF). This

variable has been created based on the time of the data collection, whereby those

patients from baseline were coded as 0 (Patients before DHFF) and those patients

from end line were coded as 1 (Patients after DHFF).

Using chi-squire test and binary logistic regression analysis assessed factors

contributing to the performance of HSR in PPHF. The logistic model in cooperated;

time (before and after DHFF introduction) and socio-demographic characteristics of

the patients (Table 10).

The findings showed that implementation of DHFF had almost nine times odds

(AOR = 8.919; CI = 4.953 - 16.060; p< 0.001) towards perception of positive health

system responsiveness among patients compared to before the implementation of this

initiative in the PPHF.

Moreover, household size and type of health facilities had significant association

with perception of patients on HSR but they were not statistically significant upon

adjusting with some other factors through AOR.

Other factors like implementation of RBF initiatives was also significantly associated

with patient’s perception on health system responsiveness (AOR = 5.984; CI = 3.416

- 10.480; p <0.001, nevertheless, renovation of PPHFs independently was not

statistically significant towards HSR.

The results from the multiple regression analysis supported the findings from the

indepenent t-test where before and after implementation of DHFF mean scores of

HSR were compared. Therefore, this denotes that DHFF has a significant effect on

improving HSR as repoted by the patients attending services at primary health

facilities.

112

Table 10: Chi-squire test for factors associated with change in overall

perception of Health System Responsiveness among patients who attended in

the public primary health care facilities before and after DHFF implementation

(n=844, p=0.05).

Variable Overall HSR Chi-square P-value

Negative N

(%)

Positive N

(%)

Time 87.4686 <. 0001

Before DHFF 137(30.90) 285(69.10)

After DHFF 29(6.87) 393(93.13)

Type of Health

Facility

9.4250 0.0021

Dispensary 60(15.19) 335(84.81)

Health center 106(23.61) 343(76.39)

Sex 1.8734 0.1711

Female 83(17.97) 379(82.03)

Male 83(21.73) 299(78.27)

Age 2.2017 0.5316

15-24 33(20.50) 128(79.50)

25-35 58(19.02) 247(80.98)

36-44 37(23.27) 122(76.73)

45+ 38(17.35) 181(82.65)

Marital status 1.2645 0.2608

Married and

cohabitates

123(18.84) 530(81.16)

Single, divorced,

separated, widow

43(22.51) 148(77.49)

113

Variable Overall HSR Chi-square P-value

Negative N

(%)

Positive N

(%)

and widowed

Level of education 2.1755 0.5368

No formal

education

17(16.19) 88(83.81)

Primary education 111(20.94) 419(79.06)

Secondary

education

25(16.89) 123(83.11)

Tertiary education 13(21.31) 48(78.69)

Household size 47.5625 <. 0001

≤5 152(22.62) 520(77.38)

>5 14(8.14) 158(91.86)

Distance to health

facility (km)

2.9501 0.2288

Within 5 120(20.98) 452(79.02)

5 - 10 32(15.53) 174(84.47)

Above 10 14(21.21) 52(78.79)

Health Facility

Visits

6.1990 0.1848

Twice 16(19.75) 65(80.25)

Thrice 16(26.67) 44(73.33)

Four times 17(25.37) 50(74.63)

Five times 17(25.37) 50(74.63)

More than five 100(17.57) 469(82.43)

114

Variable Overall HSR Chi-square P-value

Negative N

(%)

Positive N

(%)

Renovation and

RBF

41.5314 <. 0001

Both 2(9.52) 19(90.48)

Innovation

only

8(17.39) 38(82.61)

RBF only 16(6.64) 225(93.36)

None 140(26.12) 396(73.88)

115

Table 11: Logistic regression model for factors associated with change in overall perception of Health System Responsiveness

before and after implementation of Direct Health Facility Financing in PPHF (p<0.05, 95% CI).

Unadjusted logistic model Adjusted logistic model

Variable OR [95%CI] p-value AOR [95%CI] p-value

Time <. 0001 <. 0001

Before DHFF Reference Reference

After DHFF 6.514[4.243, 10.001] 8.919[4.953, 16.060]

Type of the Health facility 0.0023 0.7712

Dispensary Reference Reference

Health center 0.580[0.408, 0.823] 1.063[0.705, 1.603]

Sex 0.1716

Female Reference

Male 0.789[0.562, 1.108]

Age 0.5334

116

Unadjusted logistic model Adjusted logistic model

Variable OR [95%CI] p-value AOR [95%CI] p-value

<30 Reference

31-44 1.098[0.681, 1.770]

45-64 0.850[0.500, 1.445]

65+ 1.228[0.731, 2.062]

Marital status 0.2614

Married Reference

Single 0.799[0.540, 1.182]

Level of education 0.5389

No formal education Reference

Primary 0.729[0.417, 1.276]

Secondary 0.951[0.484, 1.865]

Tertiary education 0.713[0.320, 1.593]

Household size (Mean)(SD) <.0001 0.3070

117

Unadjusted logistic model Adjusted logistic model

Variable OR [95%CI] p-value AOR [95%CI] p-value

≤5 Reference Reference

>5 3.298[1.855, 5.864] 0.671[0.313, 1.442]

Distance to the health facility (km) 0.2317

Within 5 Reference

5 - 10 1.443[0.941, 2.213]

Above 10 0.986[0.529, 1.840]

Number of Facility visit in the past 0.1898

Twice Reference

Thrice 0.677[0.307, 1.494]

Four times 0.724[0.333, 1.573]

Five times 0.724[0.333, 1.573]

More than five times 1.154[0.641, 2.079]

Renovation and RBF <. 0001 <. 0001

118

Unadjusted logistic model Adjusted logistic model

Variable OR [95%CI] p-value AOR [95%CI] p-value

Both 3.359[0.772, 14.603] 0.1062 4.168[0.902, 19.256] 0.0675

Renovation only 1.679[0.765, 3.687] 0.1964 1.838[0.798, 4.230] 0.1525

RBF only 4.972[2.890, 8.553] <. 0001 5.984[3.416, 10.480] <. 0001

None of them Reference Reference

No other factor could explain the change in reported health system responsiveness except time of the intervention and whether facilities

were implementing RBF and renovation interventions at the time of this study.

119

4.2 Level of Structural Quality of MHS in the PPHF before and after Direct

Health Facility Financing Implementation

4.2.1 Demographic Characteristics of PPHF before and after implementation of

Direct Health Facility Financing

A total of 42 (100%) public primary health facilitates were assessed on structural

quality of MHS. Of all the facilities assessed 14 (33.3%) were health centers,

whereas 28 (66.7%) were dispensaries. The furthest primary health facilities from the

district head office were 140 kms and the nearest 2 kms.

One third of health facilities was located in the rural areas (Table 12). The majority

(60%) of health centers had less than 39 skilled staff and majority (88.89%) of

dispensaries had less than 15 staff (Table 12). Some health facilities experienced

critical shortage of staffs; for instance, five (17.89%) dispensaries had 2 staff only

while another 16 (57.14%) dispensaries had 3 staff.

The average annual deliveries per dispensary before DHFF were 120 while that of

health center was 1,203 after introduction of DHFF initiative. Upon the introduction

of DHFF the average deliveries at dispensary level increased to 230 where as that of

health center plunged to 1,313.

Majority (37.04%) of dispensaries had a catchment population of less than 5,000

people while the majority of health centers (40%) had a catchment population of

more than 30,001.

Of all 42 studied primary health facilities, 25 had piped water, 11 they were using a

rainwater harvesting system, 5 had bore holes and 1 health facility did not have a

reliable water source. Of all the studied health facilities, 14 had ambulances and, of

those ambulances, only 8 were functional.

120

Table 12: Characteristics of the Public Primary Health Facilities before and

after implementation of Direct Health Facility Financing (n=42).

Variable Before DHFF After DHFF

Number

of HFs

Percentage Number of

HFs

Percentage

Region

Dodoma 6 14.29 6 14.29

Pwani 6 14.29 6 14.29

Mbeya 6 14.29 6 14.29

Katavi 6 14.29 6 14.29

Manyara 6 14.29 6 14.29

Shinyanga 6 14.29 6 14.29

Mtwara 6 14.29 6 14.29

Location of health

facility

Urban 21 50.00 21 50.00

Rural 21 50.00 21 50.00

Facility type

Dispensary 28 66.67 28 66.67

Health centre 14 33.33 14 33.33

Catchment population

(Mean) (SD) (Range)

Dispensary

8735(6629, 1723-22039) 8735(6629, 1723-22039)

<5,000 10 37.04 10 37.04

5,000 – 10,000 8 29.63 8 29.63

>10,000 9 33.33 9 33.33

121

Catchment population

(Mean) (SD) (Range)

Health centre

25831(14442, 9105-

52781)

25831(14442, 9105- 52781)

<20,000 6 40.00 6 40.00

20,0000 – 30,000 3 20.00 3 20.00

>30,001 6 40.00 6 40.00

Number of beds (Mean)

(SD) (Range)

Dispensary

5(5,1-20) 6(3,1-16)

<10 26 92.86 24 88.89

≥10 2 7.14 3 11.11

Number of beds (Mean)

(SD) (Range) Health

centre

25(8,13-38) 33(15,12-56)

≤20 5 35.71 4 26.67

>20 9 64.29 11 73.33

Average annual

deliveries

Dispensary

≤50 12 44.44 7 25.93

51 – 100 4 14.81 7 25.93

101 – 150 5 18.52 5 18.52

151 – 200 1 3.70 2 07.41

>200 5 18.52 6 22.23

Health centre

≤150 1 6.67 3 20.00

122

151– 300 2 13.33 4 26.67

301 – 450 1 6.67 1 06.67

451 – 600 3 20.00 1 06.67

>600 8 53.33 6 40.00

Number of staffs (Mean,

SD, Range) Dispensary

6(6,2-34) 6(4,2-18)

0 – 14 26 92.86 24 88.89

≥15 2 7.14 3 11.11

Number of staffs (Mean,

SD, Range) Heath centre

39(33,10-110)

42(35,13-120)

0 – 38 10 71.43 9 60.00

≥39 4 28.57 6 40.00

Availability of

ambulance

Not available 28 66.67 28 66.67

Available 14 33.33 14 33.33

Available & source of

energy in health facility

No Electricity 6 14.29 6 14.29

National Grid 20 47.62 20 47.62

Solar Panels 12 28.57 12 28.57

Generator 2 4.76 2 4.76

Others 2 4.76 2 4.76

Available & source of

water in health facility

123

No sources 1 2.38 1 2.38

Piped Water into Health

Facility

21 50.00 21 50.00

Rainwater Harvesting 11 26.19 11 26.19

Public Tap or Standpipe 4 9.52 4 9.52

Tube Well or Borehole 2 4.76 2 4.76

Others 3 7.14 3 7.14

Status of HFGCs

Dispensary

Active 167 95.98 112 82.35

Inactive 7 4.02 24 17.65

Health center

Active 168 96.55 96 94.12

Inactive 6 3.45 6 5.88

New project introduced/

implemented during

study period

Dispensary 0 0 0 0

Health center 0 0 5 28.57

The descriptive results from the individual domains showed that there were

percentage mean differences existed before and after implementation of DHFF

initiative. There was slight percentage mean change in the structural quality domains

after DHFF implementation compared to when DHFF was not introduced (Table

13).

124

Table 13: Descriptive Results on Change of Structural quality of MHS in

Tanzania before and after DHFF implementation (n= 42).

Domain Before

DHFF

After

DHFF

N % N %

Privacy

All service delivery rooms have doors that

close

36 85.71 41 97.62

All service delivery rooms with screen to

partition the examination area/bed

24 57.14 33 78.57

Windows with curtains or painted or with

frosted glass

28 66.67 35 83.33

Divider (screen/curtain) if the service delivery

room is shared.

24 57.14 34 80.95

Hygiene

Functioning toilet [VIP] latrine which is not

full OR a flushing toilet with work

33 78.57 37 88.10

Toilets clean inside and out with no stagnant

water and no foul smells

18 42.86 35 83.33

Presence of toilet paper or water. In case of

flushing toilet, a dust bin

17 40.48 30 71.43

Hand-washing facilities just outside the toilet

or with basin inside toilet [soap]

13 30.95 23 54.76

No organic waste 30 71.43 38 90.48

No used syringes, needles 19 45.24 37 88.10

125

Domain Before

DHFF

After

DHFF

N % N %

No used bandages or dangerous products on

the ground of the facility that area

6 14.26 37 88.10

Do the burning chamber fenced in and ash pit

available?

6 14.26 20 47.62

Do the placenta pit with slab and cover

available?

6 14.26 33 78.57

Labour ward

Delivery bed functional, clean, adjustable,

with a footstool

23 56.10 39 92.86

One functional gouse neck lamp / light source 17 41.46 24 57.14

One functional new-born weighing scale 36 87.80 40 95.24

One drum with Sterile gauze (with date and

closed)

18 43.90 34 80.95

One drum with cotton wool (with date and

closed),

16 38.10 32 76.19

One Suction machine 16 39.02 36 85.71

Resuscitation kit (Ambu-bag, tubes) 27 65.85 38 90.48

One Mackintosh for each delivery bed 28 68.29 40 95.24

At least 2 sterilized delivery trays 28 68.29 38 90.48

Two Plastic aprons 28 68.29 37 88.10

Two pairs Gumboots/closed shoes 28 68.29 30 71.43

Two Masks 28 68.29 24 57.14

Two Goggles 28 68.29 24 57.14

126

Domain Before

DHFF

After

DHFF

N % N %

At least one full or nearly full box of Clean

(50 pairs)

28 68.29 34 80.95

One full or nearly full Sterile gloves (50 pairs) 28 68.29 32 76.19

Obstetric Care

Sterilized manual removal aspiration kit

available [MVA kits2]

7 50.00 13 86.67

Blood transfusion facilities available [Blood

bank with 5 units different groups

3 21.43 6 40.00

Presence of Vacuum extractor 8 57.14 9 60.00

Presence of Gun let gloves for manual

removal of placenta

8 57.14 10 66.67

Suction machine 13 92.86 15 100.00

Resuscitation kit [Ambubags different sizes,

Sodium bicarbonate, Vit. K]

11 78.57 13 86.67

Three bottles of Ringer lactate, Three bottles

of Normal saline 1000mls

11 78.57 13 86.67

Sets of cannula, Gauge 14 & 18 13 92.86 11 73.33

Giving sets 3 11 78.57 14 93.33

Syringes 2cc, 5cc, 10cc [5,5,5] 10 71.43 15 100.00

Magnesium Sulphate 14 100.00 15 100.00

Nifedipine/Hydralazine 9 64.29 12 80.00

Metronidazole Inj. 8 57.14 11 73.33

Ampicillin inj. OR Gentamycin inj. 9 64.29 13 86.67

127

Domain Before

DHFF

After

DHFF

N % N %

Ceftriaxone inj. 7 50.00 12 80.00

Oxytocin 7 50.00 15 100.00

Sedatives (E.g. diazepam) 11 78.57 13 86.67

Waste management

Labor ward and dressing room 4 9.52 38 90.48

Three buckets, each bucket clearly labeled

with todays date, 1 bucket with chloro

2 4.76 30 71.43

Inclient wards (Including labor ward,

laboratory and immunization/injection room)

8 19.05 38 90.48

At least 1 safety box with sharps not

exceeding ¾ full, and no sharps sitting on

6 14.29 40 95.24

Proper waste segregation using Red, Yellow

and Black/Blue bins with color coded

8 19.05 36 85.71

Liners labeled bin liners ok in lieu of colored 11 26.19 30 71.43

Sterilization

Existence of proper means / methods of

sterilizing instruments: Steam sterilization

36 85.71 39 92.86

SOPs for sterilization displayed on the wall by

the equipment

16 38.10 17 40.48

Each pack has an indicator for control of

sterility (litmus or date of sterilization

15 35.71 21 50.00

Maternal death audit

Select one audited case and check if it is

completely

6 14.29 10 23.81

128

Domain Before

DHFF

After

DHFF

N % N %

Check the selected audited case if it is

correctly filled

6 14.29 9 69.23

Check the selected audited case if it is action

plan in place

6 14.29 8 61.54

Assess if the strategies are in place with

emphasis on ANC clinic

9 32 76.19

Assess if the strategies are in place with

emphasis on labour and delivery

9 21.43 32 76.19

Assess if the strategies are in place with

emphasis on post-natal care

9 21.43 30 71.43

Assess if the strategies are in place with

emphasis on Client/Community factors

9 21.43 30 71.43

4.2.2 Level of Structural Quality of Maternal Health Service in PPHF before

and after implementation of Direct Health Facility Financing.

The level of structural quality of MHS was assessed by using seven domains of

maternal related indicators (Table 14). The overall structural quality was determined

by using mean score percentage of 60 cut-off point adopted from previous studies

(Gilson et al., 1995; Yahya & Mohamed, 2018). The descriptive results portays that

mean, median, and standard deviation before implementation of DHFF were 29.77,

27.20 and 16.40 respectively. Upon the implementation of DHFF mean, median and

standard deviation changed to 44.10, 40.50 and 19.53 correspondingly. The mean

difference before and after implementation of DHFF was 14.33, while change in

median and standard deviation were 17.43 and 18.56 respectively. Normality of data

was tested by using Kolmogorov–Smirnov test and was found to be normally

distributed, therefore the change of structural quality of MHS in PPHF was

determined by using paired t-test.

129

The performance of structural quality of maternal health service from paired T-test

portrayed overall structural quality had a significant change (p<0.001) with mean

difference of 19.39 from the baseline to endline. Moreover, 30.95% of PPHFs scored

above 60% mean score which was a cutoff point for high structural quality of MHS

post-DHFF implementation as compared to 2.38% of PPHFs which scored below

cutoff point pre-DHFF introduction (Figure 23). This significant change was also

observed in the specific domains of structural quality except sterilization of medical

equipment that had no significant change before and after, while availability of

health commodities for obstetric services showed a negative change (Table 14).

The results show that maternal audit had a significantly higher (p<0.001) mean

difference (7.62), followed by the labour ward which had mean difference of 3.55

(p<0.001), hygiene and sanitation and privacy had mean difference of 1.26 (p=0.001)

and 0.74 (p=0.001) respectively (Table 14).

However, the overall change of structural quality of MHS showed that before the

implementation of DHHF only 2.38% (Figure 23) of primary health facilities had

higher structural quality of maternal health service but after eighteen months of

DHFF implementation there were a hike in the number of primary health facilities

with higher structural quality of MHS to 30.95% as shown in Figure 23 below.

.

130

Table 14: Paired T-test Results for Change in Structural Quality of MHS in PPHFs before and after implementation of DHFF

(n=42, p=0.05, CI = 95%).

Domain Mean (95%CI) Mean

deference

T-value P-value

Privacy

Before implementation of DHFF 2.67 [2.29,3.04] 0.74 3.56 0.0010

After implementation of DHFF 3.41 [3.10,3.71]

Hygiene and Sanitation

Before implementation of DHFF 3.83 [3.14,4.53] 1.26 3.60 0.0009

After implementation of DHFF 5.10 [4.59,5.60]

Labour room

Before implementation of DHFF 6.62 [5.51,7.73] 3.55 6.57 <. 0001

After implementation of DHFF 10.17 [9.39,10.94]

131

Domain Mean (95%CI) Mean

deference

T-value P-value

Waste management

Before implementation of DHFF 3.83 [3.14,4.53] 0.86 2.26 0.0290

After implementation of DHFF 4.69 [4.19,5.19]

Availability of Obstetric emergences medicines

Before implementation of DHFF 13.02 [10.84,15.21] -3.76 -2.79 0.0080

After implementation of DHFF 9.26 [5.25,13.27]

Sterilization of medical equipment

Before implementation of DHFF 1.60 [1.32,1.87] 0.24 1.50 0.1423

After implementation of DHFF 1.83 [1.56,2.11]

Maternal death Audit

Before implementation of DHFF 4.64 [2.62,6.66] 7.62 4.51 <. 0001

After implementation of DHFF 12.26 [9.55,14.98]

132

Domain Mean (95%CI) Mean

deference

T-value P-value

Overall Structural quality

Before implementation of DHFF 24.71[20.46,28.95] 19.39 7.10 <.0001

After implementation of DHFF 44.10[38.01,5018]

133

Figure 23: Percentage Change of Structural Quality of Maternal Health service

in PPHF before and after implementation of Direct Health Facility Financing

(n=42).

4.2.3 Other factors that influenced change in structural quality of MHS in

PPHF before and after implementation of Direct Health Facility Financing.

The determination of other factors that attributed to change in structural quality of

MHS before and after implementation of DHFF suggested that all primary health

facilities that implemented RBF increased inhigh structural quality of maternal health

service in PPHFs (Table 15). However, the majority of renovated primary health

facilities had low structural quality of maternal health health services (Table 15).

134

Table 15: Fisher’s exact test results for other factors that attributed to change in

overall structural quality of MHS in PPHF before and after implementation of

Direct Health Facility Financing (n= 42).

Variable Low

Structural

quality

High

Structural

quality

P-value

Neither Renovation nor RBF 0.0485*

No 10(66.67) 5(33.33)

Yes 10(37.04) 17(62.96)

RBF 0.0457*

No 20(52.63) 18(47.37)

Yes 0(0.00) 4(100.00)

Renovation 0.0024*

No 10(32.26) 21(67.74)

Yes 10(90.91) 1(9.09)

On the same note the results from cell analysis portrayed that observed cases for

PPHFs without implementation of RBF and Renovation had significant influence on

high structural quality, also all facilities implementing RBF had higher influence on

high structural quality while renovation alonehad no significant influence on

structural quality (Table 16).

135

Table 16: Cell analysis results for for other factors that attributed to change in

overall structural quality of MHS in PPHF before and after implementation of

Direct Health Facility Financing (n= 42)

Variable Low Structural

Quality

High

Structural

Quality

Neither Renovation nor RBF

No Observed 10 5

Expected 7.14 7.86

Yes Observed 10 17

Expected 12.86 14.14

RBF

No Observed 20 18

Expected 18.10 19.91

Yes Observed 0 4

Expected 1.90 2.10

Renovation

No Observed 10 21

Expected 14.76 16.24

Yes Observed 10 1

Expected 5.24 5.76

136

4.3 Level of maternal health service utilization in PPHF before and after

implementation of Direct Health Facility Financing

In this study, maternal health service utilization was assessed by using selected nine

indicators of health service utilization namely number of institutional deliveries,

number of women attending 4th ANC visits, number of ANC mothers given IPT2,

number of ANC mother initiated mebandazole for deworming, availability of 30

tracer medicines, and use of modern family planning methods.

Also, the study incorporated indicators for number of HIV positive pregnant women

receiving ARVs, number of new users on modern family planning methods and

number of pregnant mothers attending first ANC visit with first 12 weeks of

pregnancy. Moreover, the study covered postnatal services by assessing the number

of mothers receiving postnatal services within three to seven days after delivery.

Furthermore, this study established the status of the availability of all 30-tracer

medicine before and after the implementation of DHFF in PPHFs.

Data for assessing these indicators were extracted from Health Management

Information System (HMIS) using District Health Information Sofware (DHIS2).

The data covered the period of two years namely (January – December) 2017

reflecting the period before the implementation of DHFF and two years later in

(January – December) 2019 after the implementation of DHFF. Moreover, results

from the descriptive analysis shows that mean, median, and standard deviation before

implementation of DHFF were 3652, 1897 and 4476 while after the implementation

of DHFF they changed to 4792, 3089 and 4942 respectively.

The results from the current study showed that, the overall maternal health service

utilization had significant positive mean difference (mean difference = 515; p=0.05),

however, service utilization at the dispensary level had higher significant positive

change as compared to the health center level (Table 17). Moreover, the dispensary

had significant positive mean difference in all nine indicators compared to only three

indicators (i.e., number of women attending four ANC visits, attending first ANC

before 12th week, and HIV positive pregnant women receiving ARV) that were

statistically significant in health centers.

137

The results of this study showed change on the user of modern family planning

methods was higher than other indicators with a mean difference of 262 (p<0.001).

Attendance at the 4th ANC visit and ANC visit within 12 weeks were significantly

positively increased upon the introduction of DHFF with mean difference of 236

(p<0.001) and 140 (p<0.001) respectively.

Further significant change was observed on institutional deliveries, use of IPTp2 and

post-natal services within three to seven days. The proportion of availability of tracer

medicines in PPHFs had a mean difference of 3.64 (p=0.02) pre and post DHFF.

Upon running into the Wilcoxon signed ranged test it was found out that; All

domains of maternal health service utilization were significant after the introduction

of DHFF except for the use of IPT2 (Table 17) and number of HIV infected women

who had significant negative change between the baseline and endline.

4.3.1 Other factors associated with Change of Maternal Health Service

Utilization in PPHF before and after implementation of Direct Health Facility

Financing.

As Table 17 signifies the implementation of DHFF as one of the factors that

attributed to the change in health service utilization at PPHFs, the current study

further explores other factors that attributed to this change. The Poisson regression

analysis was used to determine other factors for the change of maternal health service

utilization in PPHFs since the dependent variable was discrete and assumed

occurrence of events were independent of the outcome, also mean and standard

deviation of overall health service utilization were 1142.52 and 1511.35 respectively

which are close and attained poisson regression model assumptions.

The results from the poisson regression analysis indicated that health centers, rural

health facilities, staffing level, structural quality of MHS and other health program

such that RBF and Renovation of PPHF were significantly associated with change of

maternal health service utilization in primary health facilities (Table 18).

138

Table 17: Wilcoxon signed rank test for Change in Maternal Health Service

Utilization before and after implementation of Direct Health Facility Financing

in PPHF (n=42, p= 0.05).

Indicator Before

DHFF

(2017)

After

DHFF

(2019)

Difference

Mean

Student's

t-test

p-value

Institutional deliveries

Dispensary 170 194 72 4.39 0.0003*

Health center 1203 1701 526 2.55 0.0244

Overall 537 765 249 2.85 0.0074*

Use of IPT2

Dispensary 197 465 271 3.59 0.0016*

Health center 1004 1091 4.92 0.01 0.9885

Overall 492 686 175 1.36 0.1835

Number of ANC mother

initiated on Mebandazole

(De-Worming)

Dispensary 302 428 127 3.56 0.0016*

Health center 974 1172 325 1.64 0.1295

Overall 537 676 191 2.78 0.0085*

Number of Women

attending 4th ANC Visit

Dispensary 137 263 181 4.50 0.0001*

Health center 363 831 573 3.47 0.0046*

Overall 220 452 315 4.62 <

.0001*

Number of HIV positive

139

pregnant on ARVs

Dispensary 17 3 -19 -3.44 0.0262

Health center 39 14 -35 -34.67 0.0061*

Overall 24 7 -24 -24.62 0.0009*

Number of new users on

modern Family Planning

methods

Dispensary 826 1225 595 4.04 0.0005*

Health center 2961 2838 408 1.04 0.3154

Overall 1589 1819 262 3.114 0.0036*

Public health facility with

all 30-tracer medicines

Dispensary 91.23 95.61 4.36 4.31 0.0002*

Health center 92.72 95.13 2.41 0.92 0.3717

Overall 91.77 95.44 3.6439 3.20 0.0027*

Number of mothers

receiving Post Natal

Services within 3-7 days

after delivery

Dispensary 63 119 52 2.69 0.0126*

Health center 591 653 62 1.12 0.2824

Overall 256 310 56 2.39 0.0219*

ANC before 12 weeks of

Gestation

Dispensary 50 145 95 3.54 0.0022*

Health center 202 431 216 3.99 0.0021*

140

Overall 107 246 140 5.04 <.

0001*

Overall Utilization

Dispensary 1657 2386 1149 4.39 0.0002*

Health center 7243 8374 1130 1.76 0.0998

Overall 3652 4794 1143 4.09 0.0002*

Table 18: Poisson Regression Model for other factors associated with change of

maternal health service utilization in Public Primary Health Facilities before

and after implementation of DHFF.

Parameter Estimate Standard

Error

Wald 95%

Confidence

Limits

Wald

Chi-

Square

Pr > ChiSq

Intercept 6.9542 0.0115 6.9316 6.9767 365921 <.0001

Facility type

Health center -0.3256 0.0162 -

0.3572

-

0.2939

405.90 <.0001

Dispensary Ref

Location

Rural 0.1567 0.0108 0.1355 0.1780 209.67 <.0001

Urban Ref

Structural quality

of ,MHS

0.0162 0.0004 0.0155 0.0170 1907.89 <.0001

Renovation and

RBF

Done 0.7092 0.0135 0.6828 0.7356 2768.64 <.0001

Not done Ref

141

4.4 Level of Direct Health Facility Financing (DHFF) Fidelity of

Implementation (FoI) and its potential moderators

The FoI of DHFF was assessed in two categories namely the knowledge of health

care workers to implement the iniative and the adherence of the health care workers

towards the guiding principles for implementation of DHFF in PPHFs.

4.4.1 Socio-Demographic Characteristics of Health Service Providers in

PPHFs

A total of 238 of health service providers participated in the study for fidelity of

implementation. Of 238 health service providers participated, more than two-third

were (70.59%) were male. Most of health service providers were between 31 and 44

years old. In terms of education, half of health service providers were certificate

holders, followed by diploma holders (30.67%) and university degree holders

(8.40%). More than half of the participants had employment experience less than 10

years, while 21% had work experience of between 10 and 20 years. Similarly, 50%

of health service providers were nurses, while clinicians were 11.76% of the sample

and assistant accountants were 3.78%. More than two-thirds of the health service

providers were frontline health service providers, while 4% were health managers

(Table 19).

142

Table 19: Social Demographic characteristics of Health Care Workers (HCWs)

in PPHF implementing DHFF.

End line

Variable Frequency Percentage

Region

Dodoma 43 18.07

Pwani 57 23.95

Mbeya 25 10.50

Katavi 26 10.92

Manyara 26 10.92

Shinyanga 33 13.87

Mtwara 28 11.76

Facility type

Dispensary 102 42.86

Health center 136 57.14

Location of the facility

Urban 124 52.10

Rural 114 47.90

Sex

Female 70 29.41

Male 168 70.59

Age group Mean [SD, Range] 37.9 [10.2,22-59]

≤ 35years 123 51.68

>35 115 48.32

Marital status

143

Married, cohabiting 164 68.91

Single, divorced, separated, widow and

widowed

74 31.09

Highest level of education

Certificate 125 52.52

Diploma 68 28.57

Advanced diploma 5 2.10

University degree 20 8.40

Masters (MMED) 3 1.26

Other 17 7.14

Years of experience (Median, IQR) (8.0, 1-41)

≤ 10 144 60.50

>10 94 39.50

Cadre

Nurse 120 50.42

Clinicians 28 11.76

Assistant Accountant 9 3.78

Medical Attendant 48 20.17

Pharmacist 13 5.46

Laboratory personnel 20 8.40

Position Distribution

Health service provider 180 78.60

Matron/Patron 7 3.06

In charge 3 1.31

Assistant Account 9 3.93

Supporting staffs 30 13.10

144

4.4.2 Knowledge of Health Service Providers on the implementation of Direct

Health Facility Financing in PPHF.

Out of 238 health service providers, 75.63% (180) had adequate knowledge in

relation to DHFF (mean scored >9.286 points) while 24.37% (58) had inadequate

knowledge. The mean score was 9.286 (SD =4.1409; range 0 -16) points indicated

that, on average, health service providers had adequate content on DHFF.

The participant’s awareness was 88.24% however, descriptive results on the

assessment of knowledge had shown that, 95.71% of HSPs acknowledge that DHFF

is different from the previous method of sending money to PPHFs; while 30% can

explain properly what DHFF meant, and 22.86% were trained or oriented about

DHFF prior to its implementation.

Table 20: Descriptive results of knowledge among health care workers in PPHF

during the implementation of Direct Health Facility Financing

Variable Frequency Percentage

Have you ever heard about DHFF program?

No 28 11.76

Yes 210 88.24

Where did you hear about DHFF program?

On the training 22 10.48

From In charge of HF 133 63.33

From other staffs 42 20.00

Somewhere else 13 6.19

What does DHFF Really mean?

Direct submission of funds to the primary Health

Facilities from treasurer.

144 68.75

145

Direct submission of funds from whatever source to

the primary health facilities

107 31.43

Does DHFF differ from the previous disbursement

mechanism?

No 9 4.29

Yes 201 95.71

DHFF must start from the treasurer

No 47 23.38

Yes 154 76.62

DHFF select few funding sources

No 107 53.23

Yes 94 46.77

It should be disbursed on a monthly basis

No 150 74.63

Yes 51 25.37

DHFF does not need presence of HFGC?

No 106 52.74

Yes 95 47.26

Do the health workers abide to financial regulations

when implementing DHFF?

No 9 4.29

Yes 201 95.71

Timely reporting on income and expenditures

No 15 7.46

Yes 186 92.54

Use of receipts whatever the transaction is done

No 22 10.95

146

Yes 179 89.05

Abide to procurement regulations

No 1 0.50

Yes 200 99.50

Preparation of performance report

No 1 0.50

Yes 200 99.50

Where you trained on FFARS and DHFF initiative

No 162 77.14

Yes 48 22.86

Looking into level of FoI and the knowledge of health care workers. It was found out

that of 238 health workers who participated in the study, 75.63% had adequate

knowledge and 27.73% had high FoI towards DHFF (Table 21).

Table 21: Level of FoI among Health Service Providers in PPHFs

Variable Frequency (Percent)

Knowledge

Adequate knowledge (>9.9878049) 180 (75.63)

In adequate knowledge (≤9.9878049) 58(24.37)

Fidelity/ Adherence

High fidelity (≥80%) 66(27.73)

Low fidelity (<80%) 172(72.27)

147

The current research posits that most of the socio- demographic factors analysed

were not associated with knowledge on DHFF, except respecting the location of the

health facilities (Rural) (p=0.004). The results from the logistic regression model

found out that health service providers from PPHFs located in the rural settings were

7.7 times more likely to have DHFF knowledge as compared to their urban

counterparts [AOR 7.7 (2.093,28.356)]. Additionally, inclusive decision-making

process had association with knowledge, but it was not statistically significant

(Table 22).

Table 22: Logistic regression model for factors associated knowledge (p=0.005)

Variable Unadjusted logistic model Adjusted logistic model

OR (95%CI) p-value AOR (95%CI) p-value

Health facility type 0.7274

Dispensary Reference

Health center 1.112[0.613, 2.018]

Facility location 0.0005

Urban Reference Reference 0.0021

Rural 3.137[1.644, 5.987] 7.704[2.093,

28.356]

Sex 0.4957

Female Reference

Male 0.793[0.407, 1.546]

Age (years) 0.1343

≤ 35 Reference

148

Variable Unadjusted logistic model Adjusted logistic model

OR (95%CI) p-value AOR (95%CI) p-value

> 35 0.634[0.349, 1.151]

Marital status 0.7361

Married Reference

Single 1.118[0.585, 2.134]

Level of education 0.2465

Certificate Reference

Diploma 0.788[0.392, 1.584]

Degree and above 0.525[0.247, 1.118]

Inclusive decision-

making process

0.0231 0.0080

No Reference Reference

Yes 3.000[1.162, 7.743] 3.899[1.426,

10.663]

Years of experience

of working at PPHF

0.3403

≤ 10 Reference

>10 0.747[0.411, 1.360]

149

4.4.3 Fidelity of Implementation of DHFF by Health Service Providers

Of 238 health service providers, only 66 (27.73%) reported to have high level of

fidelity on DHFF implementation in their health care facilities (Table 21). The

results of chi-squire for determination of factors associated with FoI revealed that,

cadre of health service providers (p<0.001), position held by the health service

providers (p=0.002) and level of knowledge among health service providers on the

implementation of DHFF (p<0.001) influenced high level of DHFF FoI in PPHFs

(Table 22).

For cadre of health service providers, it was noted that, all assistant account had

higher level of FoI as compared to other cadres such that nurses (28.33%) and

clinicians (21.10%). On the same aspect, all health service providers who were

assigned in managerial task (other staffs) had higher chances of FoI (43.10%)

compared to frontline service providers (22.78%). Moreover, health services

providers with adequate knowledge on DHFF were more likely to possess high level

of FoI compared to those with inadequate knowledge on DHFF (Table 23).

Table 23: Chi-squire test for factors associated with FoI

Variable Level of DHFF Fidelity of

Implementation

Chi-

square

P-value

Low FoI

N (%)

High FoI

N (%)

Health facility type 1.5724 0.2099

Dispensary 78(76.47) 24(23.53)

Health center 94(69.12) 42(30.88)

Facility Location 0.5738 0.4488

Urban 87(70.16) 37(29.84)

Rural 85(74.56) 29(25.44)

Sex 0.0171 0.8959

Female 51(72.86) 19(27.14)

Male 121(72.02) 47(27.98)

Age(years) 0.8116 0.3676

≤ 35years 92(74.80) 31(25.20)

150

> 35 80(69.57) 35(30.43)

Marital status 0.0225 0.8809

Married 119(72.56) 45(27.44)

Single 53(71.62) 21(28.38)

Level of education 0.6356 0.7277

Certificate 93(74.40) 32(25.60)

Diploma 48(70.59) 20(29.41)

Degree and above 31(68.89) 14(31.11)

Cadre 25.8671 <.0001*

Nurses 86(71.67) 34(28.33)

Clinicians 86(78.90) 23(21.10)

Assistant accountant 0(0.00) 9(100.00)

Position in the health

facility

9.0426 0.0026*

Health service provider 139(77.22) 41(22.78)

Other staff 33(56.90) 25(43.10)

Years of experience 0.0004 0.9841

≤ 10 104(72.22) 40(27.78)

>35 68(72.34) 26(27.66)

Knowledge 16.6105 <.0001

Inadequate 54(93.10) 4(6.90)

Adequate 118(65.56) 62(34.44)

Acceptability 0.0014 0.9698

Acceptable 63(72.41) 24(27.59)

Unacceptable 109(72.19) 42(27.81)

The results from the logistic regression model portrays that knowledge on DHFF

implementation among health care workers had higher influence on fidelity of DHFF

implementation.

Health care workers with adequate knowledge were eight times more likely to

implement DHFF with higher FoI compared to HCWs with inadequate knowledge

(AOR = 8.14, CI = 2.755-24.099, p=0.0001) (Table 24). Likewise, health care

workers with managerial positions were three times more likely to implement DHFF

with higher FoI as compared to frontline health service providers (AOR = 3.049, CI

=1.557-5.972, p=0.0012) (Table 24).

151

Table 24: Logistic regression results for factors associated with FoI in PPHF

before and after implementation of DHFF (n=234).

Variable Unadjusted logistic model Adjusted logistic model

OR (95%CI) p-

value

AOR (95%CI) p-

value

Position of HCWs

in the health

facility

0.0031 0.0012

Health service

provider

Reference Reference

Other staff 2.568[1.374- 4.801] 3.049[1.557-5.972]

Knowledge 0.0003 0.0001

Inadequate

knowledge

Reference Reference

Adequate

knowledge

7.093[2.455- 20.496] 8.148[2.755-

24.099]

4.5 The association between Fidelity of Implementation (FoI) and overall

Structural quality of maternal health service in PPHF.

Looking into the association between changes of structural quality by FoI it was

found out that, there were positive structural changes among facilities with high FoI

as compared to those with low FoI.

Upon running in the logistic regression model to look for changes on structural

quality by FoI it was found out that, facility with high FoI were almost twice high in

structural quality of MHS compared with low FoI facilities (AOR=1.821, CI = 0.994

- 3.334, p= 0.00523) (Table 25).

152

Table 25: Association between Changes of Structural Quality by Fidelity of

Implementation

Variable Overall Structural Quality of MHS Chi-

square

P-value

Low Structural

Quality (%)

High Structural

Quality (%)

FoI 3.8199 0.0506

Low fidelity 76(44.19) 96(55.81)

High fidelity 20(30.30) 46(69.70)

Table 26: Logistic regression results for association between FOI and overall

structural quality of HS and in PPHFs before and after implementation of

DHFF (n= 42).

Variable Unadjusted Overal Change of Structural

Quality

OR (95% CI) P-value

Fidelity of Implementation 0.00523

Low fidelity Reference

High fidelity 1.821[0.994,3.334]

153

4.6 The association between Fidelity of Implementation (FoI) and overall

Maternal health service utilization in PPHF before and after implementation

of Direct Health Facility Financing.

Looking into the association between changes of maternal health service utilization

by FoI it was found out that, there was high maternal health service utilization among

facilities with low FoI as compared to those with high FoI (Table 27).

Table 27: Association between overall changes of health service utilization and

Fidelity of Implementation of Direct Health facility financing in PPHFs.

Variable Utilization Chi-square P-value

Low maternal

health service

utilization (%)

High maternal

health service

utilization (%)

FoI 16.3856 <. 0001

Low fidelity 16(9.30) 156(90.70)

High fidelity 20(30.30) 46(69.70)

The results of logistic regression analysis on the association between FoI and

utilization of maternal health service portrayed that, where there was existence of

high FoI of DHFF among health service providers, there was less utilization of MHS

(OR = 0.236, CI = 0.113 - 0.492, p<0.001) compared to high FoI health facilities

(Table 28).

154

Table 28: Logistic regression results for the association between FoI and overall

maternal health service utilization in PPHFs before and after implementation of

DHFF.

Variable Unadjusted OR for overall change of

maternal health service utilization

OR (95% CI) P-value

Fidelity of Implementation 0.0001

Low fidelity Reference

High fidelity 0.236[0.113,0.492]

4.6.1 Health Care Providers’ experience towards Direct Health Facility

Financing

Participants’ responsiveness towards DHFF was explored through 14 in-depth

interviews with key informants including DMOs, District DHFF coordinator, Health

Facility in charges and Chair of HFGCs. A total of three themes and eight sub –

thematic areas were obtained after thematic analysis.

A total of fourteen people from three district councils participated in 14 in-depth

interviews with the following socio-demographic characteristics (Table 29).

155

Table 29: Socio-demographic Characteristics of Interviewees

DMOs DDHFF Co In-Charges HFGC

Total

Interviewed

3 3 4 4

Age ≤ 35 2 1 2 2

>35 1 2 2 2

Gender

Male

2

1

1

2

Female 1 2 3 2

4.6.2 Thematic Findings for Qualitative Research

Table 30: Results from Indepth interviews by thematic and sub thematic areas

Theme Sub – themes

Health Care Providers’

experience toward DHFF

implementation

Training

Knowledge

Coordination

Health Care Providers’

Responsiveness towards

DHFF implementation

Attitude

Panctuality

- Issues

- Availability

- Utilization/Provision

Enabling Environment - Capacity building for DHFF implementers

- Adherence to the Public Financial management

protocols

- Motivation of DHFF implementers

156

Thematic Area Number 1: Health Care Providers’ experience towards DHFF

4.6.2.1 Training on DHFF, FFARS, Planning and Budgeting

Exploring whether participants had training on DHFF and FFARS it was revealed

that, majority attended the training, the challenge; however, they also had to

subsequently train others at their health facilities.

“Capacity building through training was another thing which

was implemented before the start of the program and therefore

one of the staff had to attend a special training on how this

program operates and shared the feedback with rest of the

workers at the facility”. (IDIs – DDHFF Co).

It was also revealed that, the approach that was used for the training was cascading,

as it was echoed by one of the participants who said:

“Training has only been provided to the health facility

in-charge and not to all healthcare providers and this

brings difficulties when discussing various issues related

to DHFF as most of the staffs seem to be not aware on most

of the issues”. (IDIs – Health Facility in Charge).

Other participants acknowledeged a training they have had on Planning and

Budgeting, as it was testified by a participant who said:

“Having knowledge and skills on planning and budgeting offers

a great opportunity to have a realistic budget that reflects

the community needs”. (IDIs – Health Facility in Charge).

157

4.6.2.2 Knowledge (Content) on DHFF initative

Assessment of knowledge of health service providers in relation to the DHFF

inititative, it was found out that, majority of health facility incharges were

knowledgable.

“This is a program which involves sending fund directly

to the health facilities as opposed to the old way which

involved sending money to the health facilities via district

medical officer”. (IDI in Charge-Health Facility).

Another testimonial was given by another participant who said:

“This program requires involvement of HFGC before you

commit any transaction and also whatever decision is made must

be participator”.

He added by saying:

“This is one of the financial decentralization approaches that

are geared towards improving the quality of health services

at the primary health care level”. (IDI – Health Facility in charge).

4.6.2.3 Coordination of DHFF activities

Exploring the coordination of DHFF activities it was found that the majority of

health service providers were aware of the of the coordination issues around DHFF

implementation. This was testified by one of the participants who said: -

“Once the fund is sent to the facilities, they have to acknowledge

the receipt of the fund by revealing the exact amount sent to

158

them. The next step is that I will ask them to submit their plan

before they start spending the money. Therefore, I can monitor

the fund through their plan; no money can be authorized

without submitting the action plan. The action plan also helps

to trace if what is planned to be spent on is what is in the budget?”

Then he paused, and started giving an example by saying:

“For instance, if you receive five million, you have to submit an

action plan for verification showing how you are going to spend

the money and if it happens to be not in the annual budget, we don’t

approve it. So for an activity to be approved, it has to be passed

through all he recommended steps and once it is implemented,one

has to submit a receipt and a report. So I do monitor the program

through all these mentioned steps”. (IDI – DDHFF Co).

Whereas the other participant testified by saying that:

“Before submitting their budgets, we usually go and meet the

health facilities guardians to discuss the identified challenges

and advise them the best process of planning and budgeting”.

He then added by saying;“This goes hand in hand with conducting supportive

supervision and mentorship so that to ensure if they have adhered to he guidelines

and protocols”. (IDI – DMO).

159

Thematic Area Number 2: Health Care Providers’ Responsivenes towards

DHFF implementation

4.6.2.4 Attitude of Health care providers towards DHFF

Looking at the attitude of the health service providers towards DHFF

implementation. They admitted to having had a good experience with the outcomes

of DHFF.

“This is one of the best programs I have ever seen, almost everyone

loves it at the Health Facility”. (IDI HFGC Chairperson).

Another participant appreciated the feeling of belongingness and ownership:

“This program makes us feel we are part of the system as we

can make decisions on our own and things happens right away

and things gives us impetus to thrive more”. (IDI Health Facility in

Charge).

4.6.2.5 Timeliness in addressing issues due to DHFF presence

Respondents admitted that there has been timely response to the issues they raise

with regards to the DHFF due to its web in nature.

“To me DHFF is the best because it has made the process

of planning easy as compared to the previous system. The web

based has helped us to work with the specific health facility

as compared to the old stand-alone plan Rep that forced us

to manually prepare the action plan for each individual facility”.

(IDI– DMO).

160

Then added by saying that:

“The old stand-alone plan Rep involved a lot of paper works and

books preparation as compared to the current system where one

can anytime access the system online, wherever someone is, and

for each specific facility. To me this program is the best even in

the preparation of quarterly reports, I can trace the performance

of any facility online by looking at their reports and their status.

So this is the best program”. (IDI– DMO).

4.6.2.6 Timely availability of health service provider’s benefits, tools and other

resources

Majority of health care workers acknowledged that, the introduction of DHFF initive

has brough some relief in terms of getting their benefits, working tools and other

resources as echoed by some of them who said:

“Nowdays we are getting our uniform allowances timely and also

we have all ledgers that are helping us to keep inventory of all

what is needed in our facilities”. (IDI - Health Facility in Charge).

Other participants testified that there is constant supply of medicines since

introduction of DHFF by saying that:

‘Since the introduction of DHFF initiative we have been having

constant supply of essential medicines and people are no longer

complaining about medicine shortage”. (IDI – HFGC Chairperson).

4.6.2.7 Timely fund utilization and service provision

Majority of respondents echoed on the importance of having DHFF initiative being

implemented as it helps to improve services provision close to their locality.

“What I can say is that this program has brought healthcare

161

services close to the health facilities as compared to the previous

years. In the previous years, planning was done by the heath facility

workers but the fund was retained by the DMO’s office, so after

planning each facility had to apply for the fund from the DMO’s

office”. (IDI-1; In charge of Health Facility).

The other participant also echoed this in the following statement:

“During previous arrangement very few people were deciding how

much to give us and when to give us that caused a lot of delays and

also hampered service provision of our health facility”. (IDI – HFGC

Chairperson).

4.6.2.8 Capacity of health service providers on management of DHFF initiative

Exploring the capacity on management of DHFF initiative it was found out that, the

lower level health facilities had adequate capacity for the management of DHFF.

“Yes, there were some challenges as we were worried if the lower

level could manage to implement the program and if they could

manage to timely utilize the whole fund but we thank God that it

was well perceived and its implementation is well execute”.

(IDIs–DDHFF Coordinator).

Other participant spoke on the importance of having knowledge and skills for

leading:

“Leadership skills are important for successful DHFF

162

implementation, we at the district level we cannot teach them

everything”.(IDIs – DMO).

4.6.2.9 Timeleness in disbursement of funds

The majority of participants indicated that the DHFF initiative has solved the issue of

fund disbursement delays as compared to the previous arrangement, which was

reflected in the following comment:

“Nowdays we receive funds timely as compared to previous time,

however, the challenge is you may find out that sometimes funds

have gone to our facilities without our knowledge”. (IDI –DMO).

Another participant echoed the same:

“We really love this arrangement; it is up to us to make most of this

System”. (IDI – HFGC Chairperson).

4.6.2.10 Improvement in working environment

Many participantsshared that:

“In the meantime, we can conduct minor renovations and also take

tea or coffee during working hours centrally to previous

arrangement were we have to ask everything from the DMO’s

office”. (IDI – In Charge).

Another participant also echod this experience by saying:

“Nowdays our facility undergoes minor renovations now and then

and makes facilities attractive to patients and make the health care

workers happy”. (IDI –HFGC Chairperson).

4.7 Thematic Number 1: Governance

A total of 49 people participated in seven FGDs which composed of seven people

whose socio-Demographic characteristics are as follows: - Council Health Service

Board, Health Facility Governing Committee, two matrons (one from Dispensary

and one from Health centre), two Health facility in charges (one from Health Centre

and One from Dispensary) and Assistant accountants (Table 30).

163

The themes and sub themes obtained from FGDs were the ones from the theoretical

frameworks and also the others emerged as per discussion. The transcripts for the

indepth interview and FGDs are found in Appendix 9.

Table 31: Characteristics of Interviewees

CHSB HFGC In Charges Matrons Assistant

Accountants

Total

Interviewed

7 7 14 14 7

Age ≤ 35 1 2 4 4 6

> 35 6 4 10 10 1

Gender

Male

6

2

12

1

3

Female 1 4 2 13 4

4.7.1 Governance of DHFF

Participants in the FGDs pointed out to their roles in the DHFF initiative and pointed

out that their roles are mainly to supervise and also to endose various decisions at the

primary health care facilities, as pointed out by one of participants of the FGDs.

“Our role is to authorize the allocated fund towards the main

objectives of the health facility. Dispensary has some priorities that

are shared to us for the review”. (FGD 1).

Moreover, they highlighted the challenges they face in implementing the DHFF

initiative processes that include the following: -

“A challenge is that, FFARS system involves a series of processes

and hence time consuming”.(FGD 2).

164

4.7.2 Funds Management

Participants expressed the importance of having funds management knowledge and

skills so that to enable smooth implementation of DHFF initiative. One of the

participants echoed by saying that:

“We were taught on financial management, however there is a

need to be also tought on bookkeeping”. (FGD 4).

The other participants expressed importance of participatory monitoring,

and endorsement of all transactions by all key player. One of the

participants said:

“All members should always participate in decision making

regarding endorsement of funds, its use and also expenditure

report”. (FGD 7).

4.7.3 Capacity of Health Facility Governing Committee (HFGC) Members

Majority of participants echoed on the importance of capacitating the HFGC

members on their roles and reponsibilities in relation to the implementation of DHFF

initiative. One of the partipants in the discussion said:

“We have been just inspecting medicines and supplies that are

brought at the facility and we do not have much skills on the

financial management, and budgeting”. (FGD 6).

Other participant expressed the importance of using mobile technology in helping the

HFGC to discharge their duties accordingly.

“Nowdays it is important to be flexible so that to build the

capacity of the HFGC so that they can discharge their duties

accordingly, itis not feasible to call for a meeting of HFGC

members throughout the country”. (FGD 5).

4.7.4 Planning and Budgeting

In this aspect the participants explained their involvement in the process and also

spelled out their roles and responsibilities. For example, one of the FGD participants

said:

165

“They have to prepare their own budget then share it with us for

verification so see if they have followed the rules and guidelines”. (FGD 3).

Other participants acknowledged the importance of having a well-prepared budget

and plan for successful implementation of DHFF initiative, one of the participants

said:

“For successful implementation of DHFF initiative we as

implementers we must be conversant with planning and

budgeting skills so that we can have realist budgets for

improvement of community health”. (FGD 4).

4.8 Thematic Area Number 2: Accountability

In this theme the particpants responded to the issues around transparency and

community involvement, supportive supervision and mentorship.

4.8.1 Transparency and Community Involvement

For this thematic area, participants were able to explain how they were engaged in

various discusions around transparency and also their participation in the decision-

making processes.

“All issues discussed and agreed through various committee

meetingsre posted on the notes board for community

consumption”. (FGD 4).

The other participant echoed similar sentiments with slight difference. She said: -

“There are other issues we must get a go ahead from the general

village assembly where we have representatives from different

hamlets”. (FGD 6).

166

4.8.2 Financial and performance report review and approval

The financial management and performance report also surfaced in the FGDs

regarding DHFF implementation.

One of the participants said:“With introduction of FFARS it has simplified how to

manage finances and how to report our perfomances”. (FGD 7).

4.8.3 Oversight

The issue of the oversight was also discussed in FGD as in important component for

effective implementation of DHFF initiative.

In one of the FDG one participants echoed on the importance by saying that:

“The oversight to the public primary health facilities has

resulted into the change of so many things regarding DHFF

implementation”. (FGD 6).

4.9 Thematic Number 3: Supportive Supervision and Mentorship

With regard to supportive supervision and mentorship of DHFF initiative related

activity, it was found out that the quality of supportive supervision was in question,

as facilities didn't receive regular supportive supervision and mentorship as testified

in the group discussion. One participant echoed on the importance of supportive

supervision and mentorship by saying: -

“Our main role is to do supervision and to regularly visit the

system (FFARS) to see if there are issues not moving smoothly

as the system provides a clear picture on what is being

implemented and what is not being implemented. E.g. if the

received fund is just sitting in the account without being

utilized, that signifies that things are not being attended and

167

we have to find the responsible person for more clarification

and insist him/her to move things forward”. (FGD 5).

Other participant testsfies that they do not regullary receive supportive supervision

and mentorship. While participatory approches decision-making was aconsidered as

one of the key prerequisites in having successful and impactiful supportive

supervision as echoed by one of the focus group discussions:

“After submitting their action plans, we do organize a two or

three days joint meeting (depending on the work load) to

review the action plans by agreeing together which should

be included and which should not. Sometimes one can include

a certain challenge in the action plan but after a joint

discussion and deep analysis you can find it as not a challenge”.

He continued by saying that:

“After getting the consensus, the next step is to enter the action

plan into the system after this it has to go through several steps

like assessment and securitization. After being approved, they

have to be given some copies of the budget and they have to

give feedback to the committee and their colleagues at the

health facility”. (FGD 6).

Other participants, stressed that through supportive supervision CHMT offers clarity

to the circulars and guidelines to the primary health facilities, as testified by one of

the participants:

168

“Our role is to ensure that guidelines, standard operating procedures

and circulars are availed to all facilities in the district council and

also to offer clarity whenever is needed”. (FGD 7).

4.9.1 Regular monitoring of primary health facilities, Consistency provision of

guidance (mentorship)

This was stressed by majority of participants as they felt it was a new initiative that

needed guidance. This was testified by one of the participants who said:

“Being supervised by people who knows very well DHFF

initiative helps us to implement the initiative”. (FGD 6).

4.9.2 Clarification/ interpretation of policies, strategies and operational

guidelines.

This was echoed by some participants, that operational guidelines are important for

DHFF implementation. This was testified by one of the participants who said: -

“To have a unified performance throughout the region we

need to have a standardized operational guideline”. (FGD 4).

4.9.3 Integration of Quantitative and Qualitative Results

The integrated results provide a holistic picture of the DHFF initiative

implementation processes and the effects of DHFF on the health system performance

of the PPHF. The intergration used Good Reporting of A Mixed Methods Study

(GRAMMS) to guide intergration process for qualitative and quantintative results

(Kaur et al., 2019; O’Cathain et al., 2008).

4.9.3.1 Framework for Integration

In order to achieve intergration, this study used a side-by-side comparison approach

(Creswell & Plano Clark, 2011), quantitative and qualitative results were compared

169

according to the main topics of the study in order to describe the similarities and

difference presented by the two strands.

4.9.3.2 Interpretation of the Integrated Results

Both quantitative and qualitative results indicated differences in the level of

knowledge of health care providers towards DHFF initiative. The majority

participants were aware of the DHFF initiative; however, about 76% had adequate

knowledge on DHFF. During FGDs very few participants articulated well what was

the DHFF initiative all about. There was also a general lack of understanding of

DHFF knowledge, specifically the disbursement modality from Ministry of Finance

and Planning to thePPHF.

The quantitative and qualitative findings indicated the variation of implementation of

fidelity towards DHFF among health care workers who were residing in rural areas

versus those who were residing in the urban areas.

The results on implementation fidelity from both strands of data indicate that there

were variations of implementation fidelity across the facilities in implementing the

DHFF initiative.

170

CHAPTER FIVE

DISCUSSION

5.0 Introduction

This chapter discusses results from the research on the DHFF’s effects onHSR,

governance and accountablity, health services utilization, and structural quality of

maternal health services. The other results that were discussed and measured were

FoI towards DHFF initiative in PPHF. Qualitative results on participant’s responses

towards DHFF implementation and how Governance and Accountability were

exercised in the PPHF were also discussed. The discussed results are compared to the

studies done inside and outside Tanzania.

The discussed findings from this study were obtained from four study sample units,

namely, patients exiting facilities after their medical consultations, PPHF, health

managers, health service providers and members of HFGCs from seven select

regions across all zones of the Tanzanian mainland.

Globally, governments have been trying to implement new health sector reforms and

policies that are geared towards improving the health system performance of

respective countries (Berman & Bitran, 2011b; OECD, 2017; Roberts et al., 2002).

Reforms in various components of the health sectors are necessary and need to be

sustainable to be responsive to the changing demands and evolution of health

systems.

5.1 Health System’s Responsiveness

Responsiveness of health services as perceived by patients has emerged as a strong

aspect of measuring the legitimate expectations of populations for the non-health

aspects of health care systems. In Tanzania, there has been little or no evidence from

studies conducted on HSR that measures the non-health aspects of the system;

however, there are many studies conducted to measure patient satisfaction

(Binyaruka et al., 2015; Juma & Manongi, 2009; Khamis & Njau, 2014; Mwansisya

& Mwansisya, 2015). However, the results from this study present mixed findings in

comparison to other studies (Bazzaz et al., 2015; Ebrahimipour et al., 2013; Gupt,

171

Kaur, Kamraj, & Murthy, 2016; Liabsuetrakul, Petmanee, Sanguanchua, &

Oumudee, 2012; Mohammadi & Koorosh, 2014; Yakob & Ncama, 2017).

In the current study, domains of access to care, respect to dignity and clear

communication were rated as the three most important domains of HSR given their

high-performance rate. The overall performance was positive since more than half

(55.9%) of the patients/clients were satisfied with HSR at PPHFs, illustrating a

beneficial DHFF initiative effect. These results are similar to the study conducted by

Mohammadi & Koorosh (2014) in Iran with inpatient clients at Zanjan University

Hospital who reported that responsiveness satisfaction level was 58.4% .

In a study conducted in Thailand (Liabsuetrakul et al. ,2012) that sought to discover

the reasons women chose to deliver in a hospital, the authors found that respect to

dignity, clear communication, and autonomy had higher scores than other domains.

Likewise, for this current study, both women and men showed high scores with

respect to these same three measures. This can also be substantiated by the Tanzania

Demographic Health Survey of 2015/16 that showed an upward increase in the

institutional deliveries across all primary health facilities in Tanzania (MoHCDGEC,

MoH [Zanzibar], 2016). This has been demonstrated by an increase in deliveries in

institutions that proxycally show that there is an increase in satisfaction with respect

to dignity and communication as this was one of the complaints in the previous

studies as pointed out by Bishanga et al. (2019) in the study about respective

maternity care (Bishanga et al., 2019).

In contrast, studies conducted in Iran and Ethiopia (Mohammadi & Koorosh, 2014;

Yakob & Ncama, 2017) found that that the domain of respect to dignity was low.

However, similar studies had higher percentage scores on the domain of clear

communication. The reasons for the differences could be due to the fact that a study

that was conducted in Iran (who?) was done amongst inpatients while the study in

Ethiopia by Yakob & Ncama (2017) was done solely with people living among HIV

and AIDS enrolled in the ART programme. This study design therefore denied the

opportunity to study other clients or patients who were not enrolled in care and

treatment programmes for people living with HIV and AIDS. However, the current

172

study incorporated opinions from clients in different health departments and units in

primary health care facilities, including people living with HIV and AIDS while

making sure that both genders were equally represented.

This study reports a significance increase in positive health system responsiveness

from 62.84% at the baseline to 71.20% at the endline, this shows that the DHFF

initiative have had some effects on HSR. There were confounding effects from

programs like RBF in the facilities where DHFF initiative has been implemented,

showing some effect on increasing positive HSR this could be due to the fact that the

RBF program uses a similar implementation approaches and indicators to the DHFF

initiative, hence having a multiplier effect on the outcomes of interest.

5.2 Structural Quality of MHS

The quality of health services can be measured through different approaches and the

toolkit for its measurement is replete with components of structure, process, and

outcome. These components were introduced by Donabedian and have been used and

tested by many other scholars (Donabedian, 1973; Gunawardena, Bishwajit, & Yaya,

2018; Yahya & Mohamed, 2018). However, this study looked into the structural

quality of PPHF especially on MHS delivery. Some of the structural quality were

grouped into seven domains of maternal health related services and staffing level. In

this study the majority of health facilities had high structural quality, a finding that

was similar to previous studies done in five African countries (Kruk et al., 2016;

Langer et al., 2014). In the current study high structural quality was associated with

an increase in maternal health service utilization, but this finding was contrary to that

of the Ethiopian demographic health survey. In that context, one study showed that

infrastructure improvement was not enough to attract mothers to deliver at the health

facilities (Central Statistical Agency [Ethiopia] & ICF International, 2012). Poor

quality of health services has been always associated with poor maternal health

outcomes, as shown in the study that was done in resources limited countries (Van

Den Broek & Graham, 2009). In this study 60% of health centers had less than 39

required health providers while 89% had less than 15 required health providers,

which could have an effect on the structural quality of services rendered to the

173

people. In this study, there was significance difference between baseline and endline

studies; however, health centers were five times more likely to offer structural

quality as compared to dispensaries. This could be due to fact that health centers are

more equipped and better staffed than the dispensaries. It could be due to some

diagnostic services that are offered at the health centers like X – ray and ultra Sound

services that cannot be offered at the dispensary level, thereby making the health

center a superior point of care than the dispensaries. In this study, structural quality

of MHS was found to be an important determinant of the health system

responsiveness as facilities with high structural quality has three times the odds to

have positive HSR as compared to those facilities with low structural quality. This

might reflect how high structural quality facilities tend to have adequate staffing

levels with adequate availability of essential commodities and good infrastructure,

thereby attracting people to access services.

5.3 Maternal Health Service Utilization

Health service utilization can be considered as a proxy indicator for the improvement

of quality of services in sub-Saharan Africa, especially in relation to access to MHS

(Alam et al., 2015; Kanyangarara et al., 2017). Generally, in this study, there was

significant increase in maternal service utilization between the before and after

studies. This finding is in line with other studies that have been conducted during the

Tanzania Service Provision Assessment (TSPA) and Service Availability and

Readiness Assessment (SARA) both of which revealed similar findings (MoHSW,

2015b; Tanzania Ministry of Health and Social Welfare, 2013). However, the only

difference that exists is a sharp increase in the maternal health service utilization just

eighteen months after DHFF introduction, which is reasonably interpreted that there

is an effect due to DHFF implementation on maternal health service utilization.

Additionally, it is important to note that, prior to the DHFF introduction, there have

been other government initiatives that might have contributed to the increase in the

maternal health service utilization especially among women who visit primary health

care facilities. This finding is similar to the study conducted in Tanzania by

Kanyangarara et al. (2017) that revealed the same findings (Kanyangarara et al.,

2017).

174

Dispensaries had positive significant mean differences in all nine indicators as

compared to the health centers with only three (i.e., ANC visits, ANC attendance

before 12th week of gestation, and number of positive pregnant women who receive

ARVs. This achievement may reflect that MHS can be easily accessed at the

dispensary level, hence women finding no reason whatsoever to go to the health

centers for the said services while they are available at the dispensary level. In this

study, use of family planning, intermittent presumptive treatment for malaria and

other 30 tracer medicines were found to have significant difference between baseline

and endline studies. The DHFF initiative has brought more resources to the primary

health facilities; hence, they have more autonomy to decide in a timely manner on

how to order essential health commodities even before the stockout as some studies

from World Bank Fiduciary System Assessment and Boex, et al. (2015) showed that

there has been misappropriation of funds and less autonomy when funds were

disbursed through the district head office (Boex et al., 2015b; The World Bank,

2015).

In this study, facilities with high structural quality had 41 times the likelihood to

have high maternal health service utilization than those with low structural quality.

This might reflect that many women or other clients are more attracted to use high

quality services as shown in previous studies done in sub–Saharan Africa

(Kanyangarara et al., 2017; Kruk et al., 2016).

5.4 FoI towards DHFF initiative

In this study, of all health service providers studied, about 76% had adequate

knowledge (adequate content) on DHFF initiative implementation although 88.2%)

of them were aware of the existence of DHFF initiative. This might relate to the

training which was conducted with the health service providers which employed a

cascade approach that might have compromised the quality of training and impacted

its implementation due to the dilution effect; hence, about 26% did not report

adequate knowledge. In some facilities there were no DHFF guidelines that would

have helped them to acquire more knowledge on the DHFF initiative. Surprisingly,

health service providers, who were working in the rural primary health facilities, had

over 7 times more knowledge compared to their counterparts in urban settings, which

175

may reflect that those working in the rural settings have adequate time to learn more

about DHFF initiative, as there are fewer programs to be implemented than required

in urban settings. Another reason could be those who are working in the urban

settings have higher workloads yielding inadequate time to focus on best

implementation of the DHFF initiative.

Of all health services providers in the current study, approximately 28% were found

to have high fidelity to DHFF implementation. This is a lower level of FoI than

would have been anticipated as they received training on DHFF prior to the

implementation and had started to implement the initiative six months after its

introduction. This could be due to assistant accountants being in the health centers

only and providing assistance to satellite dispensaries, leading to potential burnout

and low FoI towards DHFF initiative. Also it could be due to shortage of human

resources for health. However, many studies around the global take note of

difficulties experienced and the challenges faced by the achievement of high fidelity

by new innovations or programmes (Durlak & DuPre, 2008). Durlak & Dupre (2008)

suggested that getting the higher FoI (85%) has been achieved in some contexts due

to early monitoring of the programme coupled with constant feedback to the

responsible people and the programme’s need to be sustained (Durlak & DuPre,

2008; Hasson, 2010). In this study, it was observed that, the training was done by

using a cascade approach that might have resulted to some health service providers to

fail to comprehend what they have been taught from their coallegues due to lack on

facilitation skills and hence failed to implement according to the description as

shown in the theory of change frameworks (Kapologwe et al., 2019). The other

reason could be some of the health facilitity which were assessed did not had

guidelines and in some facilities they did not engage the HFGCs in the decision

making process and implementation of the initiative due to various reasons including

expirely of HFGC tenureship is three years and some of members been pre occupied

by other duties when they are needed. HFGC are instrumental in admninistrative and

financial decisions at the primary health facility level as DHFF initiative depends on

the interface meetings between health facility management and the community

representatives, therefore any weak or partial involvement may have a negative

176

effect on DHFF initiative. The existing relationship has been well elaborated by

authors from Tanzania (Kessy, 2014; Maluka, 2017; Maluka et al., 2018).

In this study, high FoI was strongly associated with the adequate knowledge of the

participants and position held by health care worker in the health facility. This was

similar to the studies conducted by Nurjono et al (2019) in which findings portrayed

a similar picture (Nurjono et al., 2019). Supprisingly, Health care workers experience

of more than ten years was not associated with high FoI, which was also the case for

the study which was done in Dodoma –Tanzania on the improved community health

fund (iCHF) implementation in which it was found out that the implementers

responded well on the call to shift to the new designed CHF hence high FoI (Kalolo

et al., 2015a). In this study, health care workers holding managerial position had

three times high FoI of DHFF initiative as compared other front-line health service

providers. This is due to the fact that, health service providers with managerial

positions were trained on the implementation of DHFF initiative and they are

responsible to account for financial resources being used at the health facilities.

Moreover, it could be due to fact that their roles at the managerial positions do not

involve management of patients hence they have adequate time to concentrate with

DHFF issues as compared to the frontline workers.

In this study, it was found out that, facilities with high structural quality of MHS

were strongly associated with high FoI. This might be due to the fact that health

service providers at the facilities with low structural quality compensated their

performances with the high FoI towards DHFF initiative due to low turn up of

patients or clients hence giving them an opportunity to implement DHFF as per

design.

In this study, facilities with high maternal health service utilization were associated

with low FoI. This might be due to fact that, facilities with high FoI towards DHFF

initiative were much concentrating towards realization of better performance on

DHFF initiative implementation than service delivery. This was expressed by some

health care workers that; DHFF initiative has brought some additional tasks hence

increase in the workload as they had also to carry out some financial management

tasks.

177

5.5 Health Care Providers’ responsiveness towards implementation of Direct

Health Facility Financing in PPHF.

During the in-depth interviews, participants clearly outlined that there were some

improvements attributable to the DHFF initiative, while concomitantly addressing

previous challenges including timely disbursement of funds and health services

utilization. This was also the case for the evaluation study that was conducted in

Kenya (Goodman et al., 2010; Waweru, Goodman, et al., 2016; Waweru, Molyneux,

et al., 2016).

Participants showed that they had adequate knowledge on the DHFF initiative and

had received the training needed to help them implement the DHFF initiative. This

was acknowledged despite the shortcomings of the cascade approach used to foster

training.

Exploring attitude of participants towards DHFF implementation, it showed that,

majority said they had a good feeling and experience and also showed positive

attitude towards DHFF initiative implementation. Meaning that, they were ready to

implement the initiative without any problem as they see its importance to their

health systems. This might be due to the fact that the introduction of the DHFF

initiative has been able to address number of challenges they have experienced

before like late disbursement of funds and also having low control over their finances

but also it could be the mandatory nature of the initiative that gave no other option of

implementers apart from implementing it. In this study, it was revealed that the

health facility’s capacity on management of DHFF initiative was good due to

existence of account assistants in some facilities and FFARS that eased the process

of management and reporting of fund use. This was different from the study done in

Kenya and Rwanda in which health facilities were found not to have adequate

capacity to manage funds and the program as whole (Kemri Welcome Trust, 2013;

Tsofa et al., 2017).

178

5.6 Governance and Accountability on DHFF implementation

5.6.1 Governance

In this study, governance was assessed through FGDs that were carried out to

explore governance through a number of pre-determined themes that looked on the

functionality of the existing governance structures like the HFGC. Upon discussion,

the sense of autonomy and ownership of financial affairs at the PPHF was one of

positive hope and potential, as reflected in the quotation “Our role is to authorise the

allocated funds towards achieving the main objective of the facility”.

This role was seen as being able to expedite various activities without any challenges

or interference from the district level. This finding was echoed in a Kenyan study on

facility governing committees (Waweru, Molyneux, et al., 2016). However, in this

study, health service providers complained of being overwhelmed by many tasks due

to the introduction of DHFF and FFARS to manage finances in the primary health

facilities, contributing additional tasks without additional staff. this might due to fact

that this is considered as an addition task to their routine duties especially in the

facilities that had no assistant accountants.

In this study, planning and budgeting emerged as a key governance function as all

members of team were fully engaged from the beginning which built a sense of

ownership in their respective health facilities from the planning to implementation.

This finding was similar to a study conducted in Kenya which after introduction of

direct facility financing saw uniting of members of the community and health

facilities to fully participate in the planning and budgeting for the welfare of their

facilities (Goodman et al., 2010; Waweru, Molyneux, et al., 2016). This finding was

also the case for the study which was done among European countries that looked

into the governance functions of the primary health care facilities in which. among

other things the issue of planning, budgeting,, and financing, found to be of

importance to flourish by using a health system thinking approach (Espinosa-

gonzález, Delaney, Marti, & Darzi, 2019). The governance structures tend to play a

pivotal role in implementation through appropriate planning and budgeting at the

primary health care level (Kessy, 2014; Mpambije, 2017).

179

5.6.2 Accountability

In the current study, three major sub-thematic areas of accountability were studied

namely transparency and community, financial and performance report review and

approval, supportive supervision and mentorship. All sub themes had a significant

degree of dimension across study area and participants.

The first dimension of transparency and community involvement assessed issues

related to reporting mechanisms, inclusive decision-making processes, endorsement

procedures, and availability of income and expenditure reports. These were aligned

for adherence of external accountability frameworks in PPHFs. The FGD with

members of HFGC revealed that there existed a potential increase in trust,

confidence, and functionality of HFGCs especially in making decision over financial

use and endorsement of funds. This result signifies that the ToC for DHFF

(Kapologwe et al., 2019) stressed the importance of HFGC in the process of making

DHFF effective. A similar study that was conducted in Kenya (Goodman, Opwora,

Kabare, & Molyneux, 2011) revealed functionality of HFGC increased upon the

introduction of direct facility financing due to increased financial autonomy in the

facilities. Moreover, on the aspects of reporting mechanism, the current study

revealed adherence to means in which the community were informed of the decision-

making processes through community meetings and other modes of communication.

Study by Maluka (Maluka & Bukagile (2016) reported on the functionality of

HFGCs stating that community meetings were used to provide feedback on the

performance of health facilities. Both the current and Maluka & Bukagile (2016)

studies identified challenges that affect performance of HFGCs, including inadequate

knowledge of their roles and responsibilities, inadequate training to improve their

skills, and knowledge especially with the introduction of new indicatives like DHFF,

lack of motivation, and low interaction or linkage among the HFGC members and

other instruments of accountability at the community level.

On the second thematic area of financial and performance report review and approval

an important need for the interface meetings between the members of HFGC and

health facility management teams was identified to build consensus on expenditures

180

and performance. This is similar to study done in Kenya which revealed the same

experience for successful implementation of any initiative (Goodman et al., 2011).

The third dimension of accountability in the current study was supportive supervision

and mentorship. The ToC proposed by (Kapologwe et al., 2019) and other

stakeholders to measure the performance of DHFF initiative in PPHFs acknowledged

the use of supportive supervision and mentorship as an important input to ensure

adherence to guidelines during implementation of DHFF initiative. Supportive

supervision was assessed in the current study among health managers at the council

level and at health facility level revealing irregularities in the implementation of this

important component. The in-depth interview conducted with district DHFF

coordinators revealed the existence of document and system review that was

undertaken at the council headquarters upon the submission of health facility

comprehensive plan and fund request. All of these strategies are essential for the

oversight function of CHMT; however, the quality aligned in the supportive

supervision and mentorship is compromised by this practice. According to DHFF

operational standard operating procedures (URT, 2016), members of CHMT under

the DHFF coordinator are responsible to ensure quality planning and effective

implementation of the plans.

The assessment conducted at the council level assured the plans and budgets

submitted by primary health facilities address all the priorities as per health facility

planning and budgeting guidelines (URT, 2016). Moreover, the current practice

addressed the proposed benefits of supportive supervision and mentorship by other

studies (Mboya et al., 2016; Olafsdottir et al., 2014) which affirms that the quality of

implementation of programs and their effectiveness are assured through supportive

supervisions and mentorship and at all levels.

In the current study, in-depth interviews with health managers at the health facility

level revealed the existence of weakness in supportive supervision and mentorship as

these activities are not done in a timely manner due to lack of capacity to conduct

support supervision in the area of DHFF and financial resources constraints in some

council. Also, in councils with proper supportive supervision and mentorship, there

181

were good financial management practices compared to others. These findings

resonate with those reported in studies conducted in Kenya, Tanzania, and Pakistan

(Bossert & Mitchell, 2011; Kamuzora et al., 2013b; Kessy, 2014). However, the

difference between the current study and the previous studies is that, this has been

done under improved financial decentralization environment where health facilities

have been given the mandate to plan and use their own resources.

5.6.3 Theoretical reviews

In this study, two related models that helped to understand health system

performance and reforms are reviewed. These theories were summarized in the six

steps namely: Problem identification, as per this study, the main identified problem

was delays in timely disbursement of funds and misappropriation of funds that led to

poor health service delivery at the primary health facilities and lowered the level of

autonomy amongst frontline workers and governance structures like HFGC of the

primary health care facilities. It was clearly seen upon introduction of DHFF

initiative it has clearly solved the problem of delays in the disbursement of funds that

was witnessed before the introduction of DHFF initiative in 2017/2018. The other

one is problem diagnosis. This study had all control knobs that were used are in the

areas of: financing, payment, organization, regulation and behaviours or persuasion

(Roberts et al., 2002), which thus qualified this effort as a major country’s health

financing reform. Looking into each of the items in the control knobs: -

The financing knob determines what resources were available to run the health care

system. As for this study were funds from health basket fund, national health

insurance funds and improved community health fund (iCHF).

The payment knob determines resources that were available to providers. As for this

study payment was made by the Ministry of Finance and Planning (MoFP) directly to

the primary health facilities via exchequer without going through District’s head

office.

The organization knob determines the kinds of provider organizations that exist and

their internal structures that shape how these organizations perform. In this study, the

182

organizational structures that are responsible for implementation of DHFF are

CHMT, CHSB, Health Facility Management Teams (HFMT) and HFGC.

The regulation knob imposes constraints on desired behaviors to allow the health

system to perform. In this study, the regulations for financial disbursement were

under the Ministry of Finance and Planning who are also mandated with

implementation of Budget Acts.

Finally, the efforts to change behavior knob influences how individuals respond to

health sector organizations, which, in turn, shapes the opportunities that

organizations confront (Roberts et al., 2002).

Another important step was policy development. Under this step we diagnosed health

problems through the control knobs using this information to inform policies and

guidelines used to guide its implementation. As for this study, there was a policy

meeting that took place in December 2016 in which among other things they decided

an introduction of DHFF as a strategy to improve quality of health care services at

the primary health facility level. Political will and commitment were evident during

the introduction of DHFF initiative in Tanzania through a series of meetings from

both technical (technical working group and annual joint technical review meeting)

and political levels (annual joint policy meeting and discussion with social services

parliamentary committee). Both series of meetings aimed at ensuring that there was

political decisions and commitments to the introduction of the DHFF (MOHCDGEC,

2016). Evaluation of the implemented reform. In this study, the baseline and endline

cross-sectional studies were conducted to evaluate the effect of the DHFF on selected

health system areas. Then a midline study was done to evaluate processes and also

the FoI of DHFF in order to get a better understanding on how the initiative evolved

over a period of time (Kapologwe et al., 2019).

5.6.4 Conceptual frameworks

In this study, ToC was used as one of the conceptual frameworks to evaluate the

implementation of DHFF initiative in Tanzania. There were some modifications

needed, which was hypothesised at the baseline study by adding other variables that

were found to be of great importance to the endline study (Kapologwe et al., 2019).

183

Some variables included leadership training on implementation of the DHFF

initiative to all relevant stakeholders as there were variation in performance of

initiative among different primary health facilities, which some participants pointed

out might have been due to leadership capability factor (e.g. the issue of inclusive

decision making by engaging the members of the community structures). The

importance of strong leadership in the implementation of new programs or initiative

has been also pointed out by authors from Tanzania (Kamuzora et al., 2013a; S.

Maluka et al., 2018; Mgata & Maluka, 2019).

In the FoI framework, there were couple of variables that emerged during a midline

study that were not included in the original proposed framework as moderating

factors by Kapologwe et al (2019), that found to be of great effect in the

implementation of the DHFF initiative (Kapologwe et al., 2019). Some of variables

were functioning financial management tools like cashbooks, ledgers and software

like FFARS this was also one of the things that was observed in the study that was

done in Kenya that availability of financial management tools was important in the

implementation of direct facility financing in Kenya (Goodman et al., 2010;

Waweru, Molyneux, et al., 2016). This modification of variables from the original

concept was also something that was experienced by other authors who conducted

some other studies in Tanzania to look into FoI of programs like improved

community health fund (Kalolo et al., 2015b).

Therefore, combination of the above study objectives and conceptual framework was

important for the better understanding of the study variables.

184

CHAPTER SIX

CONCLUSION, RECOMMENDATIONS AND SUGGESTION FOR

FURTHER RESEARCH

6.1 Conclusion

The findings from this study revealed a significant increase in HSR after the

introduction of the DHFF initiative in Tanzania, meaning that the initiative have had

an effect on the health system performance in the country. Moreover, there has been

a significance change in the structural quality of MHS and maternal health service

utilization since the introduction of the DHFF initiative. It is noteworthy that

programmes such as Results Based Financing (RBF) and ongoing construction of the

primary health facilities across the country have had some effect on the health system

performance although these two variables were controlled for their potential

confounding effects in this study.

Furthermore, findings from this study revealed that governance of DHFF initiative

needs a collaborative effort from both the supply side (i.e., health service providers)

and the demand side (i.e., community side through the HFGC). Therefore,

understanding the governance structures by those who are responsible with

implementation of the DHFF initiative is of great importance as it determines the

success and ultimate sustainability of the initiative.

The study findings also revealed that internal accountability among health managers

presented mixed opinions since some of the participants admitted that there is

existence of strong accountability while others felt a weak accountability mechanism

was in place. This variability in results calls for decision makers and health managers

at the ministerial level to critically assess the DHFF initiative in order to create a

sense of ownership at the implementation level, rather than concentrating on a few

staff members who possess managerial positions having the control. It is important to

invest on ensuring that DHFF guidelines are available and achievable, and that there

is adequate training on DHFF and FFARS as health service providers described these

are potentially enhancing accountability mechanisms.

185

The findings of this study revealed that FoI towards the DHFF initiative was very

low despite majority of health service providers found to have adequate knowledge.

This finding calls for revision of training methodology used and exploration of

motivators and facilitators necessary to deliver the DHFF initiative as per design.

6.2 Recommendations

Based on the current research findings, discussions and conclusion, this study

generates the following recommendations:

6.2.1 Recommendation to Policy makers

From this study it was clear that the DHFF initiative have had an effect on the

structural quality of MHS in the PPHF in Tanzania, despite confounding effects from

existing programs like RBF and ongoing renovation and construction of primary

health care facilities, both of which they were controlled during study and analysis

period. Structural quality assessment should be part of routine data to be obtained

from registers (MTUHA Books) that are present at the PPHF, which should be

incorporated into management information system like the District Health

Management Information System – 2 (DHIS-2). This should go hand in hand with

development of the

From this study we learned of different sources of funds at the primary health facility

level like National Health Insurance Fund (NHIF), Community Health Fund (CHF),

User Fees (UF), Basket Fund (BF), and Council Own Source (COS). However, all

these funds have different resource allocation and use formulae hence confusing the

health managers and health service providers on expenditure patterns. There is a

need for a harmonization of these resources.

Tanzania has embarked into streghthening its Community Health System (CHS) by

formalizing the Community Own Resource Persons (CORPs) cadre commonly

known as Community Health Workers (CHWs). However, one of the identified

challenges has been lack of unified modality that is used for payment of CORPs.

This study calls for all CORPS to be paid their incentives or allowances through

PPHF which serves as a conduit for DHFF approach hence enhancing accountability

186

to both parties. This is important in the sense that CORPS are represented in the

HFGC.

In this study we used Open Data Kit (ODK) software enabling us to access real time

data with instant analysis. The tablets with GPS sensors were used for data collection

at all 42 primary health care facilities. We highly recommend this data collection

system to allow managers at all levels to have access to date to inform timely data-

driven decisions resulting responsive interventions. Furthermore, ODK offers an

opportunity for cost reduction as it reduces the burdensome of the current system

with about 15 registers for data collection from patients or clients.

6.2.2 Recommendation to Policy Implementers

The DHFF initiative in an important approach in making sure that the health system

responds to the legitimate requirements of the people who visit primary health

facilities. It is therefore important that all seven domains that were used to assess

HSR should always be carried out in the specified period of time to ensure that

delivery of services to meet the expectations of the clients. Therefore, we highly

recommend HSR surveys through exit interviews should be carried out in all primary

health facilities on a monthly or quarterly basis to offer feedback to the managers at

the respective facilities and also at district and regional on areas where they can be

improved. The tool used can also be integrated into the already existing health

management system like DHIS-2 that can simplify its analysis at different levels.

It was clear that the DHFF had an effect on the maternal health service utilization

with an increase in the number of women accessing the MHS after introduction of

the DHFF initiative. However, this calls for a study that will look into the socio-

demographics and social determinants of health that will help to understand various

factors that are responsible for an increase in maternal health service utilization. We

highly recommend that local data use on maternal health service utilization should be

given a priority as it will help health managers to be informed on what should be

improved in order to increase maternal health service utilization.

187

In this study, accountability in the PPHFs was an integral part of successful

implementation of DHFF initiative. Therefore, it is the role of the government and

stakeholders to ensure that, there is capacity development of health managers at

PPHFs and council level in order to ensure effective and efficient use of public

funds. This should be compared to the service delivered at the PPHF which satisfy

the requirements of the members of the community.

This study also calls for enhancing the governance and accountability functions like

supportive supervision and mentorship from CHMT to the primary health facilities

so that to improve and also to address all challenges that are experienced in the

course of DHFF implementation by both health service providers and HFGC which

are comprised of the community representatives.

The accountability can be appreciated by following the financial memorandum and

budget acts to ensured timely disbursement of funds and utilization by the primary

health facilities. Therefore, this study calls for Ministry of Finance and Planning,

who is the main funder, MoHCDGEC, PO-RALG and other stakeholders to comply

with all regulations governing public fund allocation and disbursement for efficiency

and effectiveness of primary health care system performance.

This study also recommends that all the stakeholders from both Government and

Non-State Actors (NSA) should comply with the DHFF initiative. There were some

NSA who used a direct to project fund approach as their disbursement modality

therefore by passing the DHFF initiative and posing a challenge in the accountability

of funds that are disbursed outside of the DHFF arrangement.

In addition, this is an opportune time to digitize all primary health care, reaching

beyond the FFARS health centre level software, in order to remove the dispensary

level reliance on financial service support including assistant accountants from the

health centres.

From this study it was learned that Guidelines for implementation of DHFF initiative

was not yet developed by the responsible ministries instead PPHF are using Circulars

and FFARS guidelines to implement the initiative. This study calls for responsible

188

ministries to develop and disseminate guidelines for DHFF implementation so that to

improve FoI.

6.2.3 Recommendations to the DHFF Initiative

From this study, it was clear that for successful implementation of the DHFF

initiative there is need to have strong and resilient governance structures from

national level to the community level. Despite successes observed in the governance

there is a need to capacitate all players in the implementation of DHFF initiative with

some leadership skills to enable them to discharge their duties accordingly. We

therefore strongly recommend for leadership and management training to all people

who are responsible for DHFF implementation.

From this study it was clear that the DHFF initiative has been rolled out the country

despite an observed level of FoI which was low among the majority of health care

providers. This flag calls for revision of training program and materials that were

used during training coupled with changing the approach of training from cascade to

another approach that will not dilute the content of the training.

Moreover, this study recommends that all financial management tools should be

availed to all primary health facilities so that to enable them to conduct proper book

keeping in line with the Government’s financial regulations and memorandum.

All health facilities need to have assistant accountants as those facilities who have

full time assistant accountants out performed dispensaries which relied on satellite

services from health centers.

From this study it was learnt that, there were long procurement processes for

commodities and supplies to be used at the PPHF as they had to seek for

endorsement of procurement of even small needed items. This was partly due to lack

of Guidelines and Circulars with threshold that allows frontline teams to procure

essentials at their locality instead of travelling long distance to seek for procurement

approval. I therefore, recommend that there should be a set threshold for

procurement of essentials at the PPHF that gives allowance for petty or micro

189

procurement and for the commodities that exceeds threshold should be forwarded to

District Head Office for approval and endorsement by District Tender Board.

6.3 Strengths of the Study

o This study used standards and validated tools to measure health system

responsiveness and structural quality of MHS (World Bank, 2013; Valentine

et al., 2003; Van Der Kooy et al., 2014).

o It used 9 indicators for assessing the Maternal Health Service utilization that

are in the One Plan II and III which are guiding documents for maternal and

child health in Tanzania and these information can be obtained from the

DHIS-2 (MOHCDGEC, 2016).

o This is both implementation research study and an evaluation study

o This study is supported by two theoretical reviews: the health system

performance model and the health system reform model. DHFF

implementation and two conceptual frameworks were also used to inform it

i.e., theory of change (ToC) and FoI.

o A range of the various stakeholders who are involved in the DHFF

implementation participated in this study, such as health managers (DMOs

and DHFF Coordinators), health service providers, patients/clients, and

members of the community (CHSB chairpersons and HFGC members).

o This study employed a mixed method approach, using both quantitative and

qualitative approaches.

6.4 Limitations of the Study

o The main limitation to this study is that it used a before and after design that

is non-controlled, because DHFF initiative has been implemented as a part of

national wide initiative; hence, there is no control group. In this case a case -

control study design would have been appropriate; however, it cannot be

done now as DHFF has been started already. Such a method may have been

possible by comparing Tanzania with neighboring countries, but such a study

would have resulted in high cost implications. In order to address some of the

challenges that might have emerged as a result of data contamination during

190

study period, there were some strategies which were put in place. First, the

project established a surveillance system with special arrangements to track

any events, projects, programmes, or any support in the study areas. That

helped in understanding if there were any contributions from other

interventions. Secondly, there was triangulation of data in order to

authenticate the collected information.

o The other limitation is that, at both baseline and endline, the study used a

cross-sectional survey that just offers the snapshot of opinions at that

moment, and such a survey makes it difficult to establish a causal effect

pathway.

o The other limitation is that, is on the objective to assess the maternal health

service utilization in which we relied on utilization data that are not enough

as other social demographic factors such as education level and social quintile

level were not studied.

6.5 Suggestions for Further Research

Further studies should address the existing gaps raised due to study limitations and

areas where the current study failed to address. The following are potential areas for

the future studies to be undertaken:

DHFF implementation is a lasting program that has undertaken an approach within

structures of the government; therefore, the results from the current study can be

improved with a long period study like prospective cohort study that will help to

establish behaviour change and health system improvement overtime.

The current study was conducted in some regions where other programs like RBF

were implemented along with the ongoing construction or renovation of primary

health facilities across the country. Despite the fact that we successfully managed to

control for confounders, we recommend that other interventional studies should be

undertaken to compare between other programme implementing regions and non-

programme implementing regions to get a clear picture of the performance.

191

The current study was a before and after, therefore we recommend another inter-

country study to be conducted Tanzania and any neighbouring country that is yet to

implement DHFF initiative as it will take care all the noted limitations from before

and after studies.

The findings from this study showed a majority of health service providers had low

FoI towards the DHFF initiative. This finding calls for another study to look for level

of motivation among health service providers and their governing structures like

CHSB and also HFGC.

The other area is implementation research to offer practical evidence-based solutions.

192

REFERENCES

Abel-smith, B., & Rawal, P. (1992). Can the poor afford “free” health services? a

case study of Tanzania. Health Policy and Planning, 7(4), 329–341.

https://doi.org/10.1093/heapol/7.4.329

Adam, T., & Savigny, D. De. (2012). Systems thinking for strengthening health

systems in LMICs : need for a paradigm shift, 2006–2008.

https://doi.org/10.1093/heapol/czs084

Addicott, R. (2011). Social enterprise in health care, 1–28.

Afulani, P. A., Buback, L., McNally, B., Mbuyita, S., Mwanyika-Sando, M., & Peca,

E. (2020). A rapid review of available evidence to inform indicators for routine

monitoring and evaluation of respectful maternity care. Global Health Science

and Practice, 8(1), 125–135. https://doi.org/10.9745/GHSP-D-19-00323

Akowuah, J. A., Agyei-Baffour, P., & Awunyo-Vitor, D. (2018). Determinants of

antenatal healthcare utilisation by pregnant women in third trimester in peri-

urban Ghana. Journal of Tropical Medicine, 2018(2000).

https://doi.org/10.1155/2018/1673517

Alam, N., Hajizadeh, M., Dumont, A., & Fournier, P. (2015). Inequalities in

maternal health care utilization in sub-saharan African countries: A multiyear

and multi-country analysis. PLoS ONE, 10(4), 1–16.

https://doi.org/10.1371/journal.pone.0120922

Andrew, C. (1995). Health sector reform: key issues in less developed countries.

Journal of International Development, 7(3), 329–347.

https://doi.org/10.1002/jid.3380070303

Ary, D., Jacobs, L., Sorenson, C., Razavieh, A., Jacobs, C., C, S., & Razavier, A.

(2010). Introduction to Research in Education.

Asian Development Bank. (2016). Rural Primary Health Services Delivery Project

(RRP PNG 41509), 1–4.

193

Ayala, G. X., & Elder, J. P. (2011). Qualitative methods to ensure acceptability of

behavioral and social interventions to the target population. Journal of Public

Health Dentistry, 71(SUPPL. 1), 1–17. https://doi.org/10.1111/j.1752-

7325.2011.00241.x

Basinga, P., Gertler, P. J., Binagwaho, A., Soucat, A. L., Sturdy, J., & Vermeersch,

C. M. (2011). Effect on maternal and child health services in Rwanda of

payment to primary health-care providers for performance: An impact

evaluation. The Lancet, 377(9775), 1421–1428. https://doi.org/10.1016/S0140-

6736(11)60177-3

Bazzaz, M. M., Reza, M., Taghvaee, E., Salehi, M., Bakhtiari, M., & Shaye, Z. A.

(2015). Health System ’ s Responsiveness of Inpatients : Hospitals of Iran.

Global Journal of Health Science, 7(7), 106–113.

https://doi.org/10.5539/gjhs.v7n7p106

Bellg, A. J., Borrelli, B., Resnick, B., Hecht, J., Minicucci, D. S., Ory, M., …

Czajkowski, S. (2004). Enhancing Treatment Fidelity in Health Behavior

Change Studies : Best Practices and Recommendations From the NIH Behavior

Change Consortium. Health Psychology, 23(5), 443–451.

https://doi.org/10.1037/0278-6133.23.5.443

Berman, P., & Bitran, R. (2011). Health Systems Analysis for Better Health System

Strengthening. Washington, DC 20433.

Binyaruka, P., Patouillard, E., Powell-Jackson, T., Greco, G., Maestad, O., & Borghi,

J. (2015). Effect of paying for performance on utilisation, quality, and user costs

of health services in Tanzania: A controlled before and after study. PLoS ONE,

10(8), 1–16. https://doi.org/10.1371/journal.pone.0135013

Bishanga, D. R., Massenga, J., Mwanamsangu, A. H., Kim, Y. M., Eorge, J.,

Kapologwe, N. A., … Stekelenburg, J. (2019). Women’s experience of facility-

based childbirth care and receipt of an early postnatal check for herself and her

newborn in Northwestern Tanzania. International Journal of Environmental

Research and Public Health, 16(3). https://doi.org/10.3390/ijerph16030481

194

Boex, J. (2013). PEFA Performance Measurement Framework at Sub National

Government Level – definitions and typology.

Boex, J., Fuller, L., & Malik, A. (2015). Decentralized Local Health Services in

Tanzania Are Health Resources Reaching Primary Health Facilities , or Are

They Getting. Urban Institute, (April).

https://doi.org/10.13140/RG.2.1.4208.5288

Borghi, J., Ramsey, K., Kuwawenaruwa, A., Baraka, J., Patouillard, E., Bellows, B.,

… Manzi, F. (2015). Protocol for the evaluation of a free health insurance card

scheme for poor pregnant women in Mbeya region in Tanzania: A controlled-

before and after study. BMC Health Services Research, 15(1), 15–17.

https://doi.org/10.1186/s12913-015-0905-1

Borrelli, B. P. (2012). The Assessment, Monitoring, and Enhancement of Treatment

Fidelity In Public Health Clinical Trials, 71, 1–21.

https://doi.org/10.1111/j.1752-7325.2011.00233.x.The

Bossert, T. J., & Mitchell, A. D. (2011). Health sector decentralization and local

decision-making : Decision space , institutional capacities and accountability in

Pakistan. Social Science & Medicine, 72(1), 39–48.

https://doi.org/10.1016/j.socscimed.2010.10.019

Braveman, P., & Gottlieb, L. (2014). The social determinants of health: It’s time to

consider the causes of the causes. Public Health Reports, 129(SUPPL. 2), 19–

31. https://doi.org/10.1177/00333549141291s206

Brousselle, A., & Champagne, F. (2011). Program theory evaluation: Logic analysis.

Evaluation and Program Planning, 34(1), 69–78.

https://doi.org/10.1016/j.evalprogplan.2010.04.001

Brunelli, B. (2007). Structural Adjustment Programs and the Delivery of Health Care

in the Third World. Pell Scholars & Senior Theses, 16, 1–29.

Carroll, C., Patterson, M., Wood, S., Booth, A., Rick, J., & Balain, S. (2007). A

195

conceptual framework for implementation fidelity. Implementation Science, 9,

1–9. https://doi.org/10.1186/1748-5908-2-40

Central Statistical Agency [Ethiopia], & ICF International. (2012). Ethiopia

Demographic and Health Survey 2011, 1–452.

Century, J., Rudnick, M., & Freeman, C. (2010). A Framework for Measuring

Fidelity of Implementation : A Foundation for Shared Language and

Accumulation of Knowledge, 199–218.

https://doi.org/10.1177/1098214010366173

Chakraborty, N., Islam, M. A., Chowdhury, R. I., Bari, W., & Akhter, H. H. (2003).

Determinants of the use of maternal health services in rural Bangladesh. Health

Promotion International, 18(4), 327–337.

https://doi.org/10.1093/heapro/dag414

Chang, R., Zangle, K. C., & Hunter, J. (n.d.). Social enterprise opportunities in the

healthcare sector.

Chimhutu, V., Tjomsland, M., Songstad, N. G., Mrisho, M., & Moland, K. M.

(2015). Introducing payment for performance in the health sector of Tanzania-

the policy process. Globalization and Health, 1–10.

https://doi.org/10.1186/s12992-015-0125-9

Cohen, D. J., & Crabtree, B. F. (2008). Research in Health Care : Controversies and

Recommendations. Annals Of Family Medicine, 6(4), 331–339.

https://doi.org/10.1370/afm.818.INTRODUCTION

Compaoré, R., Yameogo, M. W. E., Millogo, T., Tougri, H., & Kouanda, S. (2017).

Evaluation of the implementation fidelity of the seasonal malaria

chemoprevention intervention in Kaya health district, Burkina Faso. PLoS ONE,

12(11), 1–18. https://doi.org/10.1371/journal.pone.0187460

Creswell, J. W. (2014). Research Design: Quantitative, Qualitative and Mixed

methods Approaches (4th Edition).

196

Darby, C., Valentine, N., Murray, C. J., & de Silva, A. (2001). World Health

Organization (WHO) : Strategy on Measuring Responsiveness.

Donabedian. (1973). Evaluating the Quality of Medical Care. New England Journal

of Medicine, 288(25), 1352–1353.

https://doi.org/10.1056/NEJM197306212882509

Donnell, C., (2015). Defining , Conceptualizing , and Measuring Fidelity of

Implementation and Its Relationship to Outcomes in K – 12 Curriculum

Intervention Research, (June). https://doi.org/10.3102/0034654307313793

Dunn, S., Sprague, A. E., Grimshaw, J. M., Graham, I. D., Taljaard, M., Fell, D., …

Walker, M. (2015). A mixed methods evaluation of the maternal-newborn

dashboard in Ontario: dashboard attributes, contextual factors, and facilitators

and barriers to use: a study protocol. Implementation Science, 11(1), 59.

https://doi.org/10.1186/s13012-016-0427-1

Durlak, J. A., & DuPre, E. P. (2008). Implementation matters: A review of research

on the influence of implementation on program outcomes and the factors

affecting implementation. American Journal of Community Psychology, 41(3–

4), 327–350. https://doi.org/10.1007/s10464-008-9165-0

Dusenbury, L, Brannigan, R., Falco, F., & Hansen, W. B. (2003). A review of

research on fidelity of implementation: implications for drug abuse prevention

in school settings. Health Education Research, 18(May), 237–256.

https://doi.org/org/10.1093/her/18.2.237

Dusenbury, Linda, Brannigan, R., Falco, M., & Hansen, W. B. (2003). A review of

research on fidelity of implementation : implications for drug abuse prevention

in school settings. Health Education Research, 18(2), 237–256.

Dutta, A. (2015). Prospects for Sustainable Health Financing in Tanzania: Baseline

Report. Health Policy Project , Futures Group: Washington , (February), 1–40.

Ebrahimipour, H., Najjar, A. V., Jahani, A. K., Pourtaleb, A., Javadi, M., Rezazadeh,

197

A., … Shirdel, A. (2013). Health System Responsiveness : A Case Study of

General Hospitals in Iran. International Journal of Health Policy and

Management, 1(1), 85–90. https://doi.org/10.15171/ijhpm.2013.13

Espinosa-gonzález, A. B., Delaney, B. C., Marti, J., & Darzi, A. (2019). The impact

of governance in primary health care delivery : a systems thinking approach

with a European panel, 8, 1–16.

Fawler, A. (1993). Non-governmental organizations as agents of democratization:

An African perspective.

Fjeldstad, O. H. (2001). Fiscal decentralisation in Tanzania: For better or for worse?

Working Paper - Chr. Michelsen Institute, (10), 1–15.

Frederick J Gravetter, L.-A. B. F. (2012). Research Methods for Behavioral Sciences

(5th Editio).

Frieden, T. R. (2014). Six components necessary for effective public health program

implementation. American Journal of Public Health, 104(1), 17–22.

https://doi.org/10.2105/AJPH.2013.301608

Frumence, G., Nyamhanga, T., Mwangu, M., & Hurtig, A. K. (2013). Challenges to

the implementation of health sector decentralization in Tanzania: Experiences

from kongwa district council. Global Health Action, 6(1), 1–11.

https://doi.org/10.3402/gha.v6i0.20983

Gale, N. K., Heath, G., Cameron, E., Rashid, S., & Redwood, S. (2013). Using the

framework method for the analysis of qualitative data in multi-disciplinary

health research. BMC Medical Research Methodology, 13(1), 1.

https://doi.org/10.1186/1471-2288-13-117

Gary, P. et al. (1990). The New England Journal of Medicine Downloaded from

nejm.org on April 1, 2015. For personal use only. No other uses without

permission. Copyright © 1990 Massachusetts Medical Society. All rights

reserved. The New English Journal of Medicine, 323(16), 1120–1123.

198

Gilson, L., Magomi, M., & Mkangaa, E. (1995). The structural quality of Tanzanian

primary health facilities. Bulletin of the World Health Organization, 73(1), 105–

114.

Gilson, L., & Mills, A. (n.d.). Health sector reforms in sub-Saharan Africa: lessons of

the last 10 years. Health Policy (Amsterdam, Netherlands), 32(1–3), 215–243.

Goodman, C., Opwora, A., Kabare, M., & Molyneux, S. (2011). Health facility

committees and facility management - exploring the nature and depth of their

roles in Coast Province , Kenya. BMC Health Services Research, 11(1), 229.

https://doi.org/10.1186/1472-6963-11-229

Goodman, C., Waweru, E., Kedenge, S., Tsofa, B., & Molyneux, S. (2010). Funding

Kenyan health centres : experiences of implementing direct facility financing

and local budget management, 3.

Grittner, A. M. (2013). Results-based Financing Evidence from performance-based

financing in the health sector.

Guba, E. (1981). Criteria for assessing the trustworthiness of naturalistic inquiries.

Educational Communication & Technology, 29(2), 75–91.

https://doi.org/10.1007/BF02766777

Gunawardena, N., Bishwajit, G., & Yaya, S. (2018). Facility-Based Maternal Death

in Western Africa: A Systematic Review . Frontiers in Public Health .

Gupt, A., Kaur, P., Kamraj, P., & Murthy, B. N. (2016). Out of pocket expenditure

for hospitalization among below poverty line households in district Solan,

Himachal Pradesh, India, 2013. PLoS ONE, 11(2), 1–11.

https://doi.org/10.1371/journal.pone.0149824

Harachi, T. (1999). Opening the Black Box : Using Process Evaluation Measures to

Assess Implementation and Theory Building Opening the Black Box : Using

Process Evaluation, (1999). https://doi.org/10.1023/A

Harachi, T. W., Abbott, R. D., Catalano, R. F., Haggerty, K. P., & Fleming, C. B.

199

(1999). Opening the Black Box: Using Process Evaluation Measures to Assess

Implementation and Theory Building. American Journal of Community

Psychology, 27(5), 711–731. https://doi.org/10.1023/A:1022194005511

Hasson, H. (2010). Systematic evaluation of implementation fidelity of complex

interventions in health and social care, 1–9.

Hsiao, W. (2000). Inside the Black Box of Health Systems. Bulleting of the World

Health Organization.

Hulleman, C. S., & Cordray, D. S. (2009). Moving From the Lab to the Field : The

Role of Fidelity and Achieved Relative Intervention Strength. Journal of

Research on Educational Effectiveness, 2(November 2014), 37–41.

https://doi.org/10.1080/19345740802539325

Innocent Semali, D. D. S. & M. T. (2005). Health Sector Reform And

Decentralization In Tanzania : The Case Of The Expanded Programme On

Immunization At District Level. Journal of Health & Population in Developing

Countries.

International Budget Partnership. (2018). Open Budget Survey 2017 Partners.

Washington, DC 20002.

International Health System Group- Havard School of Public Health. (2000). A

Decade of Health Sector Reform: What Have We Learned? DDM Project, 2.

James Macinko, Barbara Starfield, T. E. (2009). The Impact of Primary Healthcare

on Population Health in Low- and, 32(2), 150–171.

Juma, D., & Manongi, R. (2009). Users’ perceptions of outpatient quality of care in

Kilosa District Hospital in central Tanzania. Tanzan J Health Res, 11(4), 196–

204.

Kalolo, A., Radermacher, R., Stoermer, M., Meshack, M., & De Allegri, M. (2015).

Factors affecting adoption, implementation fidelity, and sustainability of the

Redesigned Community Health Fund in Tanzania: A mixed methods protocol

200

for process evaluation in the Dodoma region. Global Health Action, 8(January

2018). https://doi.org/10.3402/gha.v8.29648

Kamuzora, P., & Gilson, L. (2007). Factors influencing implementation of the

Community Health Fund in Tanzania. Health Policy and Planning, 22(2), 95–

102. https://doi.org/10.1093/heapol/czm001

Kamuzora, P., Maluka, S., Ndawi, B., Byskov, J., & Hurtig, A. K. (2013). Promoting

community participation in priority setting in district health systems:

experiences from Mbarali district, Tanzania. Global Health Action, 6, 22669.

https://doi.org/10.3402/gha.v6i0.22669

Kanyangarara, M., Munos, M. K., & Walker, N. (2017). Quality of antenatal care

service provision in health facilities across sub-Saharan Africa: Evidence from

nationally representative health facility assessments. Journal of Global Health,

7(2). https://doi.org/10.7189/jogh.07.021101

Kapologwe, N.A,et al., (2020). Development and upgrading of public primary

healthcare facilities with essential surgical services infrastructure : a strategy

towards achieving Universal Health Coverage in Tanzania. BMC Health

Services Research.

Kapologwe N.A, J. S., & Msuya, S. E. (2011). Perceived barriers and attitudes of

health care providers towards Provider-Initiated HIV Testing and Counseling in

Mbeya region, southern highland zone of Tanzania. The Pan African Medical

Journal, 8, 17. https://doi.org/10.4314/pamj.v8i1.71070

Kapologwe,N.A (2011). Provider-initiated HIV testing and counseling in Mbeya

City, south-western Tanzania: knowledge and practice of health care providers.

Tanzania Journal of Health Research, Vol 13, No(ISSN: 0856-6496;), 4.

Kapologwe,N.A, Kagaruki, G. B., Kalolo, A., Ally, M., Shao, A., Meshack, M., …

Hoffman, A. (2017). Barriers and facilitators to enrollment and re-enrollment

into the community health funds/Tiba Kwa Kadi (CHF/TIKA) in Tanzania: a

cross-sectional inquiry on the effects of socio-demographic factors and social

201

marketing strategies. BMC Health Services Research, 17(1), 308.

https://doi.org/10.1186/s12913-017-2250-z

Kapologwe, N. A., Kalolo, A., Kibusi, S. M., Chaula, Z., Nswila, A., Teuscher, T.,

… Borghi, J. (2019). Understanding the implementation of Direct Health

Facility Financing and its effect on health system performance in Tanzania: a

non-controlled before and after mixed method study protocol. BMC Health

Services Research, 1–13. https://doi.org/https://doi.org/10.1186/s12961-018-

0400-3

Kapologwe, N. A., Kibusi, S. M., Borghi, J., Gwajima, D. O., & Kalolo, A. (2020).

Assessing health system responsiveness in primary health care facilities in

Tanzania. BMC Health Services Research, 20(1), 104.

https://doi.org/10.1186/s12913-020-4961-9

Kaur, N., Vedel, I., El Sherif, R., & Pluye, P. (2019). Practical mixed methods

strategies used to integrate qualitative and quantitative methods in community-

based primary health care research. Family Practice, 36(5), 666–671.

https://doi.org/10.1093/fampra/cmz010

Keith, R. E., Hopp, F. P., Subramanian, U., Wiitala, W., & Lowery, J. C. (2010).

Fidelity of implementation : development and testing of a measure.

Implementation Science, 5(99), 1–11.

Kemri Welcome Trust. (2013). Direct funding of health facilities Findings from an

evaluation of the Health Sector Services Fund in Kenya.

Kessy, F. L. (2014). Improving Health Services through Community Participation in

Health Governance Structures in Tanzania. Journal of Rural and Community

Development.

Khaleghian, P. (2004). Decentralization and public services: The case of

immunization. Social Science and Medicine, 59(1), 163–183.

https://doi.org/10.1016/j.socscimed.2003.10.013

202

Khamis, K., & Njau, B. (2014). Patients’ level of satisfaction on quality of health

care at Mwananyamala hospital in Dar es Salaam, Tanzania. BMC Health

Services Research, 14(1), 1–8. https://doi.org/10.1186/1472-6963-14-400

Kibusi, S. M., Kimunai, E., & Hines, C. S. (2015). Predictors for uptake of

intermittent preventive treatment of malaria in pregnancy (IPTp) in Tanzania.

BMC Public Health, 15(1), 1–8. https://doi.org/10.1186/s12889-015-1905-0

Kinney, M. V., Kerber, K. J., Black, R. E., Cohen, B., Nkrumah, F., Coovadia, H., …

Lawn, J. E. (2010). Sub-Saharan Africa’s mothers, newborns, and children:

Where and why do they die? PLoS Medicine, 7(6), 1–9.

https://doi.org/10.1371/journal.pmed.1000294

Kolehmainen-Aitken, R. L. (2004). Decentralization’s impact on the health

workforce: Perspectives of managers, workers and national leaders. Human

Resources for Health, 2, 1–11. https://doi.org/10.1186/1478-4491-2-5

Komba, A. A. (2012). Strategies for enhancing equity in financing primary education

in Tanzania, 3(June), 495–501.

Kress, D. H., Su, Y., Wang, H., Kress, D. H., Su, Y., & Wang, H. (2016).

Assessment of Primary Health Care System Performance in Nigeria : Using the

Primary Health Care Performance Indicator Conceptual Framework Assessment

of Primary Health Care System Performance in Nigeria : Using the Primary

Health Care Performance Indicato, 8604.

https://doi.org/10.1080/23288604.2016.1234861

Kruk, M. E., Leslie, H. H., Verguet, S., Mbaruku, G. M., Adanu, R. M. K., &

Langer, A. (2016). Quality of basic maternal care functions in health facilities of

five African countries: an analysis of national health system surveys. The

Lancet Global Health, 4(11), e845–e855. https://doi.org/10.1016/S2214-

109X(16)30180-2

Kutzin, J. (2013). Financement des soins de santé pour une couverture santé

universelle et résultats du système de santé: Concepts et implications politiques.

203

Bulletin of the World Health Organization, 91(8), 602–611.

https://doi.org/10.2471/BLT.12.113985

Langer, A., Salam, R. A., Lassi, Z. S., Link, C., Austin, A., Langer, A., … Bhutta, Z.

A. (2014). Approaches to improve the quality of maternal and newborn health

care : an overview of the evidence The Harvard community has made this article

openly available . Please share how this access benefits you . Your story matters

. Approaches to improve the q. Reproductive Health, 11(Suppl 2), S1.

https://doi.org/10.1186/1742-4755-11-S2-S1

Liabsuetrakul, T., Petmanee, P., Sanguanchua, S., & Oumudee, N. (2012). Health

system responsiveness for delivery care in Southern Thailand. International

Journal for Quality in Health Care, 24(2), 169–175.

Liaropoulos, L., & Goranitis, I. (2015). Health care financing and the sustainability

of health systems. International Journal for Equity in Health, 14(1), 5–8.

https://doi.org/10.1186/s12939-015-0208-5

Lindeman, S. (2014). “Until We Live Like They Live in Europe”: A Multilevel

Framework for Community Empowerment in Subsistence Markets. Journal of

Macromarketing, 34(2), 171–185. https://doi.org/10.1177/0276146713514753

Mainz, J. (2003). Defining and classifying clinical indicators for quality

improvement. International Journal for Quality in Health Care, 15(6), 523–

530. https://doi.org/10.1093/intqhc/mzg081

Maluka, S. (2017). Comprehensive case study from United Republic of Tanzania

PRIMARY HEALTH CARE SYSTEMS (PRIMASYS).

Maluka, S., Chitama, D., Dungumaro, E., Masawe, C., Rao, K., & Shroff, Z. (2018).

Contracting-out primary health care services in Tanzania towards UHC : how

policy processes and context influence policy design and implementation.

International Journal for Equity in Health, 17(118), 1–13.

https://doi.org/10.1186/s12939-018-0835-8

204

Maluka, S. O., & Bukagile, G. (2016). Community participation in the decentralised

district health systems in Tanzania : why do some health committees perform

better than others ?, (June 2015), 86–104.

Mapunda, O. E., Msuya, S. E., A. Kapologwe, N., John, B., Damian, D. J., &

Mahande, M. J. (2016). Assessment of Maternal Mortality and its Associated

Causes at Shinyanga Regional Hospital in Tanzania. Women’s Health Bulletin,

Inpress(Inpress). https://doi.org/10.17795/whb-38976

Martin, V., Msuya, S. E., Kapologwe, N., Damian, D. J., John, B., & Mahande, M. J.

(2019). Prevalence and Determinants of Modern Contraceptive Methods Use

among Women of Reproductive Age (15 - 49 Years) in Rural Setting: A Case of

Kishapu District, Shinyanga Region. Advances in Sexual Medicine, 09(04), 53–

66. https://doi.org/10.4236/asm.2019.94005

Massoi, L., & Norman, A. S. (2009). Decentralisation by devolution in Tanzania:

Reflections on community involvement in the planning process in Kizota Ward

in Dodoma. Journal of Public Administration and Policy Research, 1(7), 133–

140.

Mayumana, I., Borghi, J., Anselmi, L., Mamdani, M., & Lange, S. (2017). Effects of

Payment for Performance on accountability mechanisms : Evidence from Pwani

, Tanzania. Social Science & Medicine, 179, 61–73.

https://doi.org/10.1016/j.socscimed.2017.02.022

Mboya, D., Mshana, C., Kessy, F., Alba, S., Lengeler, C., Renggli, S., … Schulze, A.

(2016). Embedding systematic quality assessments in supportive supervision at

primary healthcare level : application of an electronic Tool to Improve Quality

of Healthcare in Tanzania. BMC Health Services Research, 1–15.

https://doi.org/10.1186/s12913-016-1809-4

Mgata, S., & Maluka, S. O. (2019). Factors for late initiation of antenatal care in Dar

es Salaam, Tanzania: A qualitative study. BMC Pregnancy and Childbirth,

19(1), 415. https://doi.org/10.1186/s12884-019-2576-0

205

MoHCDGEC., PORALG., (2017). Direct Health Facility Financing Guide.

Dodoma.

Mmari, V., Stephen, K., Lilian, M., & Osaki, K. (2019). The Implementation Fidelity

of Competency Based Curriculum for Nursing and Midwifery Programme in

Tanzania: A protocol for a Mixed Methods. Nursing & Primary Care, 3(2).

https://doi.org/10.33425/2639-9474.1101

Mohammadi, A., & Koorosh, K. (2014). Responsiveness in the Healthcare Settings :

A Survey of Inpatients. International Journal of Hospital Research, 3(3), 123–

132.

MoHCDGEC, MoH [Zanzibar], (2016). Tanzania Demographic and Health Survey

and Malaria Indicator Survey 2015-2016. Tanzania Demographic and Health

Survey and Malaria Indicator Survey (TDHS-MIS) 2015-16. Dar es Salaam,

Tanzania, and Rockville, Maryland, USA.

MoHCDGEC. (2015). Availability , Utilisation and Quality of Emergency Obstetric

and New-born Care ( EmONC ) Services in Tanzania Mainland.

MOHCDGEC. (2016). The National Road Map Strategic Plan to Improve

Reproductive, Maternal, Newborn, Child and Adolescent Health in Tanzania

(2016 - 2020) (One Plan II), (June), 142.

MoHSW. (2011a). The Tanzania Quality Improvement Framework in Health Care

(2011 - 2016).

MoHSW. (2011b). Comprehensive Council Health Planning Guidelines, (July), 1–

393.

MoHSW. (2014). RESULT BASED FINANCING (RBF) OPERATIONAL

MANUAL.

MoHSW. (2015). Tanzania Health Sector Strategic Plan 2015 -2020 (HSSP IV),

2020(July), 53.

206

MOHSW. (2009). National Guidelines for Improving Quality of Care, Support, and

Protection for Most Vulnerable Children in Tanzania. Ministry of Health and

Social Welfare, (September).

Molenberghs, G. (2010). Sampling Techniques Master in Statistics , Universiteit

Hasselt Master in Quantitative Methods , Katholieke Universiteit Brussel

Contents.

Moore, G., Audrey, S., Barker, M., & Bond, L. (2014). Process evaluation of

complex interventions. UK Medical Research Council (MRC) Guidance

Prepared, 19–45; 64–75.

https://doi.org/http://www.populationhealthsciences.org/MRC-PHSRN-Process-

evaluation-guidance-final-2-.pdf

Mowbray, C. T., Holter, M. C., Gregory, B., & Bybee, D. (2003). Fidelity Criteria :

Development , Measurement , and Validation. American Journal of Evaluation,

24(3), 315–340.

Mpambije, C. J. (2017). Decentralisation of Health Systems and the Fate of

Community Health Fund in Tanzania : Critical Review of High and Low

Performing Districts, 5(2), 136–144.

https://doi.org/10.11648/j.sjph.20170502.21

Murray, C. J., & Frenk, J. (2000). A framework for assessing the performance of

health systems. Bulletin of the World Health Organization, 78(6), 717–731.

https://doi.org/10.1590/S0042-96862000000600004

Musau, S., Chee, G., Patsika, R., Malangalila, E., Chitama, D., Van Praag, E., &

Schettler, G. (2011). Tanzania Health System Assessment 2010, (January), 1–

101.

Musgrove, P., Creese, A., Preker, A., Baeza, C., Anell, A., & Prentice, T. (2000).

Health Systems: Improving Perfomance. World Health Organization, 78(1), 1–

215. https://doi.org/10.1146/annurev.ecolsys.35.021103.105711

207

Mwansisya, T. E., & Mwansisya, G. C. M. (2015). Reliability and Validity of Patient

Satisfaction Questionnaire for Community Pharmacies in Dodoma Municipality

, Tanzania, 2(6), 163–170.

Naghavi, M., Wang, H., Lozano, R., Davis, A., Liang, X., Zhou, M., … Temesgen,

A. M. (2015). Global, regional, and national age-sex specific all-cause and

cause-specific mortality for 240 causes of death, 1990-2013: A systematic

analysis for the Global Burden of Disease Study 2013. The Lancet, 385(9963),

117–171. https://doi.org/10.1016/S0140-6736(14)61682-2

Nielsen, K., & Randall, R. (2013). Opening the black box : Presenting a model for

evaluating organizational-level interventions, 0643(January).

https://doi.org/10.1080/1359432X.2012.690556

NIMR. (2010). Evidence-informed Policy Making in the United Republic of

Tanzania : Setting REACH-Policy Initiative Priorities for Evidence-Informed

Policy Making in the United Republic of Tanzania : Setting REACH-Policy

Initiative Priorities for 2008-2010. Reeport, (September 2008).

Njeru, M. K., Blystad, A., Nyamongo, I. K., & Fylkesnes, K. (2009). A critical

assessment of the WHO responsiveness tool : lessons from voluntary HIV

testing and counselling services in Kenya. BMC Health Services Research, 11,

1–11. https://doi.org/10.1186/1472-6963-9-243

Nunnally, J. C., & Bernstein, I. H. (1979). Psychometric theory. PsycCRITIQUES.

https://doi.org/10.1037/018882

Nurjono, M., Shrestha, P., Yi, I., Ang, H., Shiraz, F., Yoong, J. S., … Vrijhoef, M.

(2019). Implementation fidelity of a strategy to integrate service delivery :

learnings from a transitional care program for individuals with complex needs in

Singapore. BMC Health Services Research, 19(177), 1–14.

https://doi.org/https://doi.org/10.1186/s12913-019-3980-x (2019)

O’Cathain, A., Murphy, E., & Nicholl, J. (2008). The quality of mixed methods

studies in health services research. Journal of Health Services Research and

208

Policy, 13(2), 92–98. https://doi.org/10.1258/jhsrp.2007.007074

Odd-Helge Fjeldstad, L. K. and E. N. (2010). Planning in Local Government

Authorities in Tanzania: Bottom-up Meets Top-down. Repoa Brief, (18).

OECD. (2017). Statistics by region: Africa. Development Aid at a Glance.

Ojemeni, M. T., Niles, P., Mfaume, S., Kapologwe, N. A., Deng, L., Stafford, R., …

Squires, A. (2017). A case study on building capacity to improve clinical

mentoring and maternal child health in rural Tanzania: the path to

implementation. BMC Nursing, 16(1), 57. https://doi.org/10.1186/s12912-017-

0252-0

Olafsdottir, A. E., Mayumana, I., Mashasi, I., Njau, I., Mamdani, M., Patouillard, E.,

… Borghi, J. (2014). Pay for performance : an analysis of the context of

implementation in a pilot project in Tanzania, 1–9.

Opwora, A., Kabare, M., Molyneux, S., & Goodman, C. (2009). The Implementation

and Effects of Direct Facility Funding in Kenya’s Health Centres and

Dispensaries, (April), 1–28.

Opwora, A., Kabare, M., Molyneux, S., & Goodman, C. (2010). Direct facility

funding as a response to user fee reduction: Implementation and perceived

impact among Kenyan health centres and dispensaries. Health Policy and

Planning, 25(5), 406–418. https://doi.org/10.1093/heapol/czq009

Opwora, A., Waweru, E., Toda, M., Noor, A., Edwards, T., Fegan, G., … Goodman,

C. (2015). Implementation of patient charges at primary care facilities in Kenya:

Implications of low adherence to user fee policy for users and facility revenue.

Health Policy and Planning, 30(4), 508–517.

https://doi.org/10.1093/heapol/czu026

Panda, B., & Thakur, H. P. (2016). Decentralization and health system performance

– a focused review of dimensions , difficulties , and derivatives in. BMC Health

Services Research, 16(Suppl 6), 1–14. https://doi.org/10.1186/s12913-016-

209

1784-9

Pérez, D., Stuyft, P. Van Der, Zabala, C., Castro, M., & Lefèvre, P. (2016). A

modified theoretical framework to assess implementation fidelity of adaptive

public health interventions. Implementation Science, 11(91), 1–11.

https://doi.org/10.1186/s13012-016-0457-8

Peters, D.H., Tran, N. T., & Adam, T. (2013). Implementation Research in Health: a

practical guide. Alliance for Health Policy and Systems Research, World Health

Organization. A Practical Guide, 66. https://doi.org/ISBN 978 92 4 150621 2

Peters, David H., Adam, T., Alonge, O., Agyepong, I. A., & Tran, N. (2014).

Republished research: Implementation research: What it is and how to do it.

British Journal of Sports Medicine, 48(8), 731–736.

https://doi.org/10.1136/bmj.f6753

PWC. (2016). Sub-national (Local Government) PEFA Assessment in Tanzania

Final Consolidated Report, (July).

Raine, R., Fitzpatrick, R., Barratt, H., Bevan, G., Black, N., Boaden, R., …

Zwarenstein, M. (2016). Challenges, solutions and future directions in the

evaluation of service innovations in health care and public health. Health

Services and Delivery Research, 4(16), 1–136.

https://doi.org/10.3310/hsdr04160

Ramsey, K., Hingora, A., Kante, M., Jackson, E., Exavery, A., Pemba, S., …

Phillips, J. F. (2013). The Tanzania Connect Project: A cluster-randomized trial

of the child survival impact of adding paid community health workers to an

existing facility-focused health system. BMC Health Services Research,

13(SUPPL.2), S6. https://doi.org/10.1186/1472-6963-13-S2-S6

Renggli, S. (2017). Promoting Universal Health Coverage in Tanzania : Towards

Improved Health Service Quality and Financial Protection.

Rifkin, S. B. (2018). Alma Ata after 40 years: Primary Health Care and Health for

210

All-from consensus to complexity. BMJ Global Health, 3, 1–7.

https://doi.org/10.1136/bmjgh-2018-001188

Roberts, M. J., Reich, M. R., Hsiao, W., & Berman, P. (2002). Getting Health

Reform Right (First). Oxford University Press.

Robone, S., Rice, N., & Smith, P. C. (2011). Health Systems’ Responsiveness and Its

Characteristics: A cross-Country Comparative Analysis. Health Services

Research, 46(6 PART 2), 2079–2100. https://doi.org/10.1111/j.1475-

6773.2011.01291.x

Rogers, P. J. (2008). Using programme theory to evaluate complicated and complex

aspects of interventions. Evaluation, 14(1), 29–48.

https://doi.org/10.1177/1356389007084674

ROURKE, A. J. (1957). Evaluating the quality of medical care. Hospital Progress,

38(9), 72–73. https://doi.org/10.2307/3348969

Saltman, R. B., Bankauskaite, V., & Vrangbaek, K. (2007). Decentralization in

Health Care. European Observatory on Health Systems and Policies Series, 9–

21. https://doi.org/126,00 Euro

Samky, H. (2019). The Influence of Direct Health Facility Financing on Perceived

Health System Responsiveness , Health Service Satisfaction and Accountability

Compliance Among Public Primary Health Facilities in Central Zone of

Tanzania . Master of Science in Public Health.

Sando, D., Ratcliffe, H., McDonald, K., Spiegelman, D., Lyatuu, G., Mwanyika-

Sando, M., Langer, A. (2016). The prevalence of disrespect and abuse during

facility-based childbirth in urban Tanzania. BMC Pregnancy and Childbirth,

16(1), 1–10. https://doi.org/10.1186/s12884-016-1019-4

Savedoff, W. D. (2011). Governance in the Health Sector: A Strategy for Measuring

Determinants and Performance. Corporate Governance, 1(May), 780–0810.

https://doi.org/doi:10.1596/1813-9450-5655

211

Schaap, R., Bessems, K., Otten, R., Kremers, S., & van Nassau, F. (2018).

Measuring implementation fidelity of school-based obesity prevention

programmes: A systematic review. International Journal of Behavioral

Nutrition and Physical Activity, 15(1), 1–14. https://doi.org/10.1186/s12966-

018-0709-x

Sharon Mihalic. (2002). The importance of implementation fidelity, (57), 1–16.

O’donnell, D., Donnelly, S., Davies, C., Fattori, F., & Kroll, T. (2020). “What

Bothers Me Most Is the Disparity between the Choices that People Have or

Don’t Have”: A Qualitative Study on the Health Systems Responsiveness to

Implementing the Assisted Decision-Making (Capacity) Act in Ireland.

International Journal of Environmental Research and Public Health, 17(9).

https://doi.org/10.3390/ijerph17093294

Shewade, H. D., & Aggarwal, A. K. (2012). Health sector reforms: Concepts, market

based reforms and health inequity in India. Educational Research, 3(2), 118–

125.

Shoo, R. S., Mboera, L. E. G., Ndeki, S., & Munishi, G. (2017). Stagnating maternal

mortality in Tanzania: What went wrong and what can be done. Tanzania

Journal of Health Research, 19(2), 1–12. https://doi.org/10.4314/thrb.v19i2.6

Smith & Firth, (2011). Qualitative data analysis: application of the framework

approach, 18, 52–62.

Smith, P. C. (2002). Measuring health system performance. European Journal of

Health Economics, 3(3), 145–148. https://doi.org/10.1007/s10198-002-0138-1

Smith, P. C., & Busse, R. (2008). Health policy and performance measurement.

Performance Measurement for Health System Improvement Experiences,

Challenges and Prospects.

Smith, P. C., Rice, N., Robone, S., & Smith, P. C. (2008). The measurement and

comparison of health system responsiveness. Health, Econometrics and Data

212

Group (HEDG), 08/05(March).

Stains, M., & Vickrey, T. (2017). Fidelity of Implementation : An Overlooked Yet

Critical Construct to Establish Effectiveness of Evidence-Based Instructional

Practices. American Society for Cell Biology, 1–11.

https://doi.org/10.1187/cbe.16-03-0113

Stuart, K., Maynard, L., & Rouncefield, C. (2017). Types of Evaluation. Evaluation

Practice for Projects with Young People: A Guide to Creative Research, 59–78.

https://doi.org/10.4135/9781473917811.n5

Sun, D., Ahn, H., Lievens, T., & Zeng, W. (2017). Evaluation of the performance of

national health systems in 2004-2011 : An analysis of 173 countries, 1–13.

https://doi.org/10.1371/journal.pone.0173346

Ministry of Health and Social Welfare. (2013). Tanzania Service Availability and

Readiness Assessment (SARA). World Health Organisation (WHO), (July).

Tashobya, C., da Silveira, V., Ssengooba, F., Nabyonga-Orem, J., Macq, J., & Criel,

B. (2014). Health systems performance assessment in low-income countries:

learning from international experiences. Globalization and Health, 10(1), 5.

https://doi.org/10.1186/1744-8603-10-5

The United Republic of Tanzania. (2016). Annual Health Sector Perfomance Profile

2014/2015. Dar es Salaam.

The World Bank. (2005). Financing health in low-income countries. Health

Financing Revisited, 209–248.

The World Bank. (2013). Results-Based Financing for Health. African Health

Forum, 3–6.

The World Bank. (2015). Fiduciary Systems Assessment Tanzania – Strengthening

Primary Health Care Services for Results, (March).

Tomson, G., & Biermann, O. (2015). Health Policy Reform in Low-Income and

213

Lower Middle-Income Countries in Southeast Asia BT - The Palgrave

International Handbook of Healthcare Policy and Governance. In E. Kuhlmann,

R. H. Blank, I. L. Bourgeault, & C. Wendt (Eds.) (pp. 171–187). London:

Palgrave Macmillan UK. https://doi.org/10.1057/9781137384935_11

Toomey, E., Matthews, J., & Hurley, D. A. (2017). Using mixed methods to assess

fidelity of delivery and its influencing factors in a complex self-management

intervention for people with osteoarthritis and low back pain. BMJ Global

Health, 7(e015452), 1–14. https://doi.org/10.1136/bmjopen-2016-015452

Tsofa, B., Goodman, C., Gilson, L., & Molyneux, S. (2017). Devolution and its

effects on health workforce and commodities management - early

implementation experiences in Kilifi County, Kenya. International Journal for

Equity in Health, 16(1), 169. https://doi.org/10.1186/s12939-017-0663-2

Umarji, M. (2015). Informative Note PEFA | Public Expenditure and Financial

Accountability Assessment Methodology Maputo , February 2015.

UN. (2015). Sustainable development goals - United Nations.

URT. (1999). Vision and Priorities to Achieve Middle Income Status by 2025

Contents from Ministry of Finance & Planning 1–35.

URT. (2007). Primary Health Services Development Programme - MMAM: 2007-

2017, 1–130.

Valentine, N., de Silva, A., Kawabata, K., Darby, C., Murray, C., Evans, D., & et al.

(2003). Health system responsiveness: concepts, domains, and

operationalization. Health System Responsiveness: Debate, Methods and

Empericism, (January 2003), 573–596.

Van Den Broek, N. R., & Graham, W. J. (2009). Quality of care for maternal and

newborn health: The neglected agenda. BJOG: An International Journal of

Obstetrics and Gynaecology, 116(SUPPL. 1), 18–21.

https://doi.org/10.1111/j.1471-0528.2009.02333.x

214

Van Der Kooy, J., Valentine, N. B., Birnie, E., Vujkovic, M., De Graaf, J. P.,

Denktaş, S., … Bonsel, G. J. (2014). Validity of a questionnaire measuring the

world health organization concept of health system responsiveness with respect

to perinatal services in the Dutch obstetric care system. BMC Health Services

Research, 14(1). https://doi.org/10.1186/s12913-014-0622-1

Vesel, L., Manu, A., Lohela, T. J., Gabrysch, S., Okyere, E., Ten Asbroek, A. H. A.,

… Kirkwood, B. R. (2013). Quality of newborn care: A health facility

assessment in rural Ghana using survey, vignette and surveillance data. BMJ

Open, 3(5), 1–11. https://doi.org/10.1136/bmjopen-2012-002326

Victoria, E., (2011). Fiscal Decentralisation and Infant Mortality Rates: The

Colombian Case

Waiswa, P., Manzi, F., Mbaruku, G., Rowe, A. K., Marx, M., Tomson, G., …

Hanson, C. (2017). Effects of the EQUIP quasi-experimental study testing a

collaborative quality improvement approach for maternal and newborn health

care in Tanzania and Uganda. Implementation Science, 12(1), 89.

https://doi.org/10.1186/s13012-017-0604-x

Walraven, G. (2019). The 2018 Astana Declaration on primary health care, is it

useful? Journal of Global Health, 9(1). https://doi.org/10.7189/jogh.09.010313

Wang, M., Fang, H., Bishwajit, G., Xiang, Y., Fu, H., & Feng, Z. (2015). Evaluation

of rural primary health care in Western China: A cross-sectional study.

International Journal of Environmental Research and Public Health, 12(11),

13843–13860. https://doi.org/10.3390/ijerph121113843

Waweru, E., Goodman, C., Kedenge, S., Tsofa, B., & Molyneux, S. (2016). Tracking

implementation and (un)intended consequences: A process evaluation of an

innovative peripheral health facility financing mechanism in Kenya. Health

Policy and Planning, 31(2), 137–147. https://doi.org/10.1093/heapol/czv030

Waweru, E., Molyneux, S., Goodman, C., & Tsofa, B. (2016). Direct funding of

health facilities: Findings from an evaluation of the Health Sector Services Fund

215

in Kenya. Resilient & Responsive Health System, 2010–2013.

Waweru, E., Opwora, A., Toda, M., Fegan, G., Edwards, T., & Goodman, C. (2013).

Are Health Facility Management Committees in Kenya ready to implement

financial management tasks : findings from a nationally representative survey,

1–14.

Westaway, M. S., Rheeder, P., Van Zyl, D. G., & Seager, J. R. (2003). Interpersonal

and organizational dimensions of patient satisfaction: the moderating effects of

health status. International Journal for Quality in Health Care : Journal of the

International Society for Quality in Health Care, 15(4), 337–344.

WHO, OECD, WB. (2018). Delivering quality health services.

WHO, UNICEF, UNFPA, World Bank Group, U. N. P. D. (2019). Trends in

maternal mortality: 2000 to 2017: estimates. Geneva.

WHO. (2013). Arguing for Universal Health Coverage. World Health Organization,

39.

WHO. (2016). The sustainable development goals report 2016. The Sustainable

Development Goals Report 2016.

https://doi.org/10.29171/azu_acku_pamphlet_k3240_s878_2016

Stansfield, S. K., Walsh, J., Prata, N., Evans, T., Organization, W. H., … Low-Beer,

D. (2012). Monitoring and evaluation of health systems strengthening: An

operational framework. Health (San Francisco), 2nd Editio(November), 1–19.

https://doi.org/10.3402/gha.v6i0.20001

WHO, & WORLDBANK. (2012). Trends in Maternal Mortality : 1990 to 2010.

Organization, 32(5), 1–55. https://doi.org/ISBN 978 92 4 150363 1

Wiedenmayer, K. A., Kapologwe, N.A, Charles, J., Chilunda, F., & Mapunjo, S.

(2015). The reality of task shifting in medicines management- a case study from

Tanzania. Journal of Pharmaceutical Policy and Practice, 8(1).

https://doi.org/10.1186/s40545-015-0032-8

216

Wiedenmayer, K., Mbwasi, R., Mfuko, W., Mpuya, E., Charles, J., Chilunda, F., …

Kapologwe, N. (2019). Jazia prime vendor system- A public-private partnership

to improve medicine availability in Tanzania: From pilot to scale. Journal of

Pharmaceutical Policy and Practice, 12(1), 1–10.

https://doi.org/10.1186/s40545-019-0163-4

World Health Organization. (2017). Developing a national health financing strategy:

A reference guide.

Yahya, T., & Mohamed, M. (2018). Raising a mirror to quality of care in Tanzania:

the five-star assessment. The Lancet Global Health, 6(11), e1155–e1157.

https://doi.org/10.1016/S2214-109X(18)30348-6

Yakob, B., & Ncama, B. P. (2017). Measuring health system responsiveness at

facility level in Ethiopia : performance , correlates and implications. BMC

Health Services Research, 17(263), 1–12. https://doi.org/10.1186/s12913-017-

2224-1

217

PUBLICATIONS

Publication Paper 1: Understanding the implementation of direct health facility

financing and its effect on health system performance in Tanzania

218

Publication Paper 2: Assessing health system responsiveness in primary health

care facilities in Tanzania

219

APPENDICES

Appendix I: A descriptive results of Health System Responsiveness as perceived

by patients in PPHF before and after implementation of DHFF

Domain Time Responsiveness

% Score, Mean

[95% CI]

Min–Max

% score

Negative

N (%)

Positive

N (%)

Prompt

attention

Base line 79.40[77.48,

81.32]

0.00- 100.00 135(31.99) 287(68.01)

End line 81.01[79.47,

82.55]

28.57-100.00 142(33.65) 280 (66.35)

All 80.21[78.98,

81.44]

0.00-100.00 277

(32.82)

567(67.18)

Respect for