“How Participatory Are We?” the Use of Community Participatory Approaches at Primary Health Care Level to Address the Growing Burden of Chronic Diseases in Sri Lanka

“How participatory are we?” The use of community participatory approaches at primary health care level to address the growing burden of chronic diseases in Sri Lanka

Kolitha Wickramage

A thesis in fulfilment of the requirements for the degree of Doctor of Philosophy

School of Public Health and Community Medicine 2015

1 | P a g e

THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet

Surname or Family name: WICKRAMAGE First name: KOLITHA Abbreviation for degree as given in the University calendar: PhD School: School of Public Health and Community Medicine Faculty: Faculty of Medicine Title: The use of community participatory approaches at primary health care level to address the growing burden of chronic diseases in Sri Lanka

Declaration relating to disposition of project thesis/dissertation

I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).

……………………… ………………………...... 11thJune 2015 Signature Witness Date

The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY Date of completion of requirements for Award:

Originality Statement

‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.’

2 | P a g e

Signed......

Date.... 11th June 2015

‘I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only).

I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.'

......

Candidate’s signature

Authenticity Statement

‘I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.’

......

3 | P a g e

ABSTRACT

The growing burden of chronic non-communicable diseases (NCD’s) in Sri Lanka required rethinking current models of primary health care (PHC), and greater involvement of communities in prevention and control programs. Enabling effective community participation (CP) is recognized as essential for prevention and control of NCDs by WHO. However, for CP to be translated from ‘rhetoric into reality’ the perspectives and practices of health workers, that are key catalysts to CP, need to be assessed.

To better understand CP in the context of PHC level NCD prevention and management, this study sought to explore the knowledge, experiences and commitment of PHC workers, health administrators and policy makers; the extent to which community participatory approaches are utilized; factors contributing to meaningful CP and develop a theoretical model in order to define factors that enable, limit, enhance or sustain meaningful CP at PHC level.

The study design utilized multiple-method approaches underpinned by a constructivist epistemology including in-depth interviews, group interviews, and an applied (focused) ethnographic study. NCD prevention interventions utilizing participatory approaches were analysed and developed into two short films. Relevant documents were examined and a descriptive cross-sectional survey conducted with health policy makers.

Despite clear commitment to the value of, and rationale for CP by policy makers, administrators, PHC workers, and within policy frameworks, the study found little evidence of meaningful CP practice addressed at PHC level. Factors contributing to CP were distilled into five broad categories: administrative and resource environment; attitudinal environment; policy and regulatory environment; technical competency and knowledge environment and external environment encompassing factors outside the formal health system that may influence the realization of CP. These were anchored within a complex health systems model.

The study findings clearly indicate that for CP to be meaningfully embraced as a viable public health strategy, health authorities need to themselves promote a culture of iterative and collaborative planning and understand the interplay of factors enabling and inhibiting its emergence. The findings of this thesis may be useful for health planners and analysts within low to middle-income countries interested in adopting a complex systems analysis approach to explore CP at PHC level.

Keywords: Primary Health Care, Non-Communicable Diseases, Community Participation, Health System Reform, Complex Health Systems Modelling, Sri Lanka, Health Policy, Policy Implementation.

4 | P a g e

ACKNOWLEDGEMENTS

There are a number of people who have accompanied me on this journey that I wish to thank. Firstly, without the participation of the health executives, district health administrators, consultants, primary health care workers and managers, community members and patients, this research would not have materialized. To them, I owe a huge thanks for their time, thoughtfulness and honesty. The energy of your dynamic participation is the very essence of this thesis!

My supervisors Professor Anthony Zwi and Dr Husna Razee were unfailingly engaged, providing constructive feedback and were available whenever I needed their guidance. Your feedback was invaluable in helping me to sharpen my writing and arguments. I am deeply grateful to both of you for supporting me through some very stressful situations over the past year. Thank you! Anthony I thank you for being a true friend and mentor over the years and look forward to our future work.

I would like to thank Dr Anne Bunde-Birouste and Prof. Jan Ritchie for their continuous support, friendship and inspiration over the years, and to Dr Palitha Abeyakoon of WHO for the guidance he provided during the formative phases of this national research study.

To Dr Susie Perera for helping to meaningfully integrate this research evidence with the Ministry of Health’s Primary Health Care revitalization process. To my Sri Lankan ‘research crew’ for their support in undertaking the field research across three districts in what were challenging and resource limited rural settings. I thank Charmara and Kavinda for assisting me produce two documentary films for the in-depth case study analysis, and importantly to the ever-dynamic Shar and loving family who sacrificed so much with quiet resolve through the many years of undertaking this research.

I am pleased that this research project, the partnerships it harnessed and the participatory process followed, has already contributed to providing an evidence base for health systems reform for decision-makers at the Ministry of Health. I sincerely hope this work will continue to practically contribute to a positive reform process. The goal of this effort was always about making a meaningful impact, and without the support and participation of those mentioned, this would not have been possible. So thank you and Keep shining!

Kol

5 | P a g e

SCIENTIFIC CONFERENCE & WORKSHOP PROCEEDINGS

1. Paper presentation: “Practical tools to explore community participation and lessons learnt from three district health systems”. Annual Scientific Sessions, College of Community Physicians, 2013 on Conference theme: “Defining Universal Health Access and Coverage in Sri Lanka”, 21-25th September 2013 in Colombo, Sri Lanka.

2. Co-Facilitated (with Director of Policy and Planning, Ministry of Health) Workshop for Ministry of Health executives, and Senior program managers entitled: “Primary Care Model for Sri Lanka”, Palm Garden Village Hotel, Anuradhapura District, on 18-19th March, 2011. Workshop was organised with the objective of harnessing the knowledge and experiences of participants to explore gaps in existing primary health care structure/function and the role and responsibilities of each category of PHC staff in order to formulate suitable policy recommendations for revitalizing and rationalizing Primary Health Care.

3. Two short-films were developed on the Case Studies profiled in this study. These were produced as two short video segments of 15 mins duration and shared with Ministry of Health as an awareness raising tool/training video to profile NCD prevention interventions that utilized community participatory approaches.

“The ways in which health systems engage with communities are best understood by using multi-level approaches. The dynamic of community participation is no doubt complex. However it is not by reducing complexity but rather seeking to uncover it, layer by layer, can we better understand its manifestation in primary health care practice. Uncovering these layers require multiple-approaches…that’s why my work encompasses a range of methods from applied ethnographic and observational studies, document analysis, surveys and interviews to examine worker perceptions, interventional maps to organisational histories…. though challenging and time consuming, this approach led to a more nuanced understanding of participation that may be useful for reformists and scholars gathered at this workshop…”

– Author, at a workshop presentation of initial study findings to Ministry of Health officials.

6 | P a g e

ACRONYMS

CA Content analysis CBI Community-based health intervention CP Community Participation CSO Community Support Officer CCP Consultant Community Physician DA Document analysis DPAD Directorate of Policy Analysis and Development EWARN Early Warning Disease Surveillance System GAP Global Action Plan for the Prevention and Control of NCDs (2013-2020) GT Grounded Theory HAs Health Administrators HCW Health Care Workers HEB Health Education Bureau HEO Health Education Officer HPSR Health Policy and Systems Research HRH Human resources for health HMP Health Sector Master Plan MOH Medical Officer of Health MoH Ministry of Health MO-MCH Medical Officer of Maternal and Child Health MO-MH Medical Officer of Mental Health MO-NCD Medical Officer of Non-Communicable Diseases MO-P Medical Officer (Planning) NCDs Non-Communicable Diseases NHDN National Health Development Network NHC National Health Council NHDC National Health Development Committee NGO Non-Governmental Organisation OECD Organisation for Economic Cooperation and Development OPD Out Patient Department P-COMPASS Participatory Compass (Tool) PPD Participation in Policy Development (Tool) PHC Primary Health Care PHI Public Health Inspector PHM Public Health Midwife RDHS Regional Director of Health Services SDH Social Determinants of Health SOS Salt, Oil, Sugar Reduction Program WHA World Health Assembly WHO World Health Organization

7 | P a g e

Table of Contents (Brief Contents)

CHAPTER 1. INTRODUCTION AND STUDY RATIONALE 1

CHAPTER 2: LITERATURE REVIEW PART 1. COMMUNITY PARTICIPATION: DEFINITIONS AND PRACTICE 18 PART 2. MODELS FOR EXPLORING CP AND EVIDENCE OF EFFECTIVENESS 34 PART 3. CP IN SRI LANKA’S HEALTH SYSTEM 51

CHAPTER 3: OVERALL THEORETICAL FRAMEWORK TO EXPLORE RESEARCH 66 OBJECTIVES

CHAPTER 4: RESEARCH METHODOLOGY, STUDY DESIGN AND METHODS 78 RESEARCH METHODS QUALITATIVE METHODS Key informant interviews 89 Group interviews 91 ‘Community Participatory Compass’ instrument 93 Observational study and use of focused ethnographic method 100 Document review method 109 QUANTITATIVE METHODS Survey rationale and method 112 Analytical process adopted 118 Strategies to ensure research rigour 122

CHAPTER 5: STUDY FINDINGS Findings of key informant and group interviews 131 Observational and focused ethnographic study results: Part A: District Health System profiles 159 Part B: Observational follow-up study 163 Part C: Mapping NCD prevention and control interventions 167 Part D: In-depth-case-study analysis 169 Results of document analysis 189 Results of the survey of policy makers 206

CHAPTER 6: ANALYTICAL FRAMEWORK SYNTHESIS AND DISCUSSION 213

REFERENCES 254 ANNEXURES 278

8 | P a g e

Table of Contents (Full Contents)

CHAPTER 1. INTRODUCTION AND STUDY RATIONALE 1.1 The ‘promise’ of community participation in health………………………………... 2 1.2 Community participation: a key strategy for NCD prevention and control………. 3 1.3 Study rationale and justification………………………………….……………… 4 1.4 Overview of research aims and objectives……………………………………… 7 1.5 Structure of the dissertation………………………………….…………………… 9 1.6 Researcher reflexivity: ‘Peeling the onion’ as research investigator……………… 11

CHAPTER 2: LITERATURE REVIEW PART 1. COMMUNITY PARTICIPATION: DEFINITIONS AND PRACTICE 2.1.1 Structure of literature review………………………………….………………….. 18 2.1.2 ‘Unpacking’ community participation: definitional challenges…………………… 19 2.1.3 Defining community participation…………………………….…………………… 20 2.1.4 Defining community: consumer or community participation? ……………………… 21 2.1.5 Defining ‘community based’ interventions…………………………….…………… 23 2.1.6 Community participation in relation to good governance and democracy……….. 25 Translating rhetoric to reality: emerging lessons in use of CP for decision-making 2.1.7 Creating an enabling policy environment…………………………….…………... 28 2.1.8 Political environment…………………………….……………………………….. 29 2.1.9 Politics of participation…………………………….…………………………….. 30 Why community participation? 2.1.10 Health systems reform through decentralization and the promise of CP…………. 31 2.1.11 Evidence of decentralization in health systems and link to CP……………………. 32

PART 2. MODELS FOR EXPLORING CP AND EVIDENCE OF EFFECTIVENESS 2.2.1 Theoretical frameworks for exploring CP: Ladders, spiders, continuums, rubrics, 34 networks and lists! …………………………….………………………………….. 2.2.2 Challenges in assessing CP and evidence of its effectiveness……………………. 39 2.2.3 Evidence from developing nations…………………………….………………….. 41 2.2.4 CP in primary health care…………………………….………………………….. 42 2.2.5 CP in NCD prevention and control…………………………….…………………. 43 2.2.6 Importance of health worker attitudes in enabling CP…………………………… 46 2.2.7 Placing health workers at the center of health services research………………… 50

PART 3. CP IN SRI LANKA’S HEALTH SYSTEM Sri Lanka country context 2.3.1 Sri Lanka’s early commitment to primary health care…………………………….. 51 2.3.2 Decentralisation and devolution of health decision-making ……………………… 51 2.3.3 Brief overview of Sri Lanka’s health system…………………………….……….. 53

9 | P a g e

2.3.4 Review of interventions utilizing community participatory approaches in Sri 56 Lanka’s health system…………………………….……………………………….. 2.3.5 Current public health challenges in Sri Lanka ……………………………………. 58 2.3.6 Chronic disease care at PHC level …………………………….…………………. 60 2.3.7 Community access to PHC services and phenomena of ‘bypassing’………………. 61 2.3.8 Current efforts to reform Sri Lanka’s PHC model and address NCDs…………….. 63 2.3.9 Section Summary…………………………….……………………………………. 64

CHAPTER 3: OVERALL THEORETICAL FRAMEWORK TO EXPLORE RESEARCH OBJECTIVES 3.1 Developing a complex systems model to explore research objectives …………... 66 3.2 Health Systems Frameworks…………………………….………………………… 66 3.3 The need for a health system model that captures complexity of CP…………….. 67 3.4 Ecological perspective in health…………………………….…………………….. 67 3.5 Adopting a complex heath systems approach to exploring CP…………………... 68 3.6 A ‘progenitor’ model to explore research objectives……………………………... 70 3.7 Research objectives …………………………….………………………………... 72 3.8 Section Summary…………………………….……………………………………. 76

CHAPTER 4: RESEARCH METHODOLOGY, STUDY DESIGN AND METHODS 4.1.1 Chapter overview…………………………….…………………………………... 78 4.1.2 Rationalizing the methodological paradigm …………………………….……….. 79 4.1.3 Study design …………………………….……………………………………….. 79 4.1.4 Research methods, procedures and study instruments……………………………. 85 4.1.5 Participants…………………………….…………………………………………. 85 4.1.6 Sampling strategy and recruitment …………………………….………………… 86 4.1.7 Ethical considerations…………………………….……………………………….. 88

RESEARCH METHODS

QUALITATIVE METHODS 4.2 Key informant interviews 4.2.1 Selection of participants for key informant interviews…………………………… 89 4.2.2 Procedures in undertaking KI’s and ensuring methodological rigour……………… 90 4.3 Group interviews 4.3.1 Procedures in undertaking GI’s …………………………….…………………….. 91 4.3.2 Group composition …………………………….…………………………………. 92 4.3.3 Development of interview guide …………………………….…………………... 92 ‘Community Participatory Compass’ instrument 4.3.3 Developing the instrument…………………………….…………………………... 93 4.3.4 Components and measures of the P-Compass instrument ………………………… 94 4.3.5 Method of applying the P-Compass instrument …………………………….……. 96

10 | P a g e

4.4 Observational study and use of focused ethnographic method 4.4.1 Focused ethnographic method …………………………….……………………... 100 4.4.2 Part A: District health system profile …………………………….……………… 104 4.4.3 Part B: Structured observational study of PHC workers …………………………. 104 4.4.4 Part C: Mapping of NCD prevention and control interventions at PHC level …….. 106 4.4.5 Part D: In-depth case study analysis of community participatory NCD prevention 107 interventions …………………………….………………………………………... 4.5 Document review method 4.5.1 Analysis of health policy and legal documents …………………………….…….. 109 4.5.2 Analytical framework …………………………….……………………………… 110

QUANTITATIVE METHODS 4.6.1 Survey rationale and method …………………………….……………………… 112 4.6.2 Rationale, survey participants and sampling frame……………………………... 113 4.6.3 Overview of survey instrument…………………………….……………………… 114 4.6.4 Pre-testing survey instrument…………………………….……………………….. 115 4.6.5 Administering the survey …………………………….…………………………... 115 4.6.6 Participation in Policy Development (PPD-Tool) used in e-survey………………… 116

4.7 Analytical process adopted 4.7.1 Quantitative data analysis………………………….……………………………. 118 4.7.2 Qualitative data analysis………………………….……………………………… 118 4.7.3 Main steps in data analytic process…………………………….………………... 119 Step 1: Data Immersion…………………………….…………………………….. 119 Step 2: Coding Process…………………………….…………………………….. 119 Stage 3: Hermeneutical process…………………………….……………………. 121

4.8 Strategies to ensure research rigour 4.8.1 Credibility …………………………….………………………………………….. 122 4.8.2 Transferability…………………………….………………………………………. 126 4.8.3 Confirmability……………………………………………………………………... 126 4.8.4 Dependability…………………………………………………………………….. 127 4.8.5 Reflexivity………………………………………………………………………… 129

CHAPTER 5: STUDY FINDINGS

5.0 Chapter overview 131 5.1 Findings of Key informant and group interviews 5.1.1 Participant characteristics………………………………………………………… 132 5.1.2 Definition of community…………………………………………………………… 133 5.1.3 Definition of community participation…………………………………………….. 134 5.1.4 ‘Rights based’ approach to CP…………………………………………………… 135 5.1.5 The modes and means of CP……………………………………………………… 136 11 | P a g e

5.1.6 Role of communities in NCD prevention…………………………………………… 137 5.1.7 Role of PHC workers in NCD prevention…………………………………………. 138

5.2 Factors inhibiting community participatory approaches at PHC Level

Health system factors A. Operational and technical capacities 5.2.1 Existing duty lists that do not emphasise CP approaches ……………………..….. 140 5.2.2 Enhancing skills and knowledge in NCD prevention and management …………... 140 5.2.3 Ad-hoc efforts in capacity building and adopting innovations ……...…………… 141 5.2.4 Monitoring and evaluation mechanisms that capture CP …………………………. 143 5.2.5 Role of curative care institutions in enabling CP for NCD prevention ……………. 143 B. Professional ethos and cultural barriers 5.2.6 Culture of participatory decision-making within the health system……………….. 145 5.2.7 Fear of losing expert authority…………………………………………………… 146 5.2.8 Historical prioritization of MCH interventions……………………………………... 146 5.2.9 Role of General Practitioners in CP ……………………………………………… 147 C. Structural barriers 5.2.10 Scaling down PHC service areas…………………………………………………. 148 5.2.11 Inadequate resource allocation and logistical constraints ……………………….. 149 5.2.12 Addressing human resources for health gaps ……………………………………. 149

Community Factors A. Community attitudes 5.2.13 Lack of perceived benefit in CP by communities …………………………………. 150 5.2.14 Demand for clinically oriented interventions …………………………………...… 151 5.2.15 Time constraints …………………………………………………………………... 151 B. Socio-cultural and political barriers 5.2.16 Social cultural hierarchy …………………………………………………………. 152 5.2.17 Gate-keepers and political opportunism…………………………………………. 152 5.2.18 Corrosion of volunteerism…………………………………………………………. 153 5.3.1 Application of the P-Compass tool 154 5.3.2 Reflections on using the P-Compass……………………………………………….. 154 Section Summary………………………………………………………………….. 155

Findings from the focused ethnographic and observational study Overview of the observational and focused ethnographic study results………….. 159 5.4 Part A: District health system profiles 5.4.1 Demographic and chronic disease profile ……………………………….……… 159 5.4.2 Human resource for health profile……………………………………………….. 160 5.4.3 PHC demarcation and effect on community engagement ………………………... 161

12 | P a g e

5.5 Part B: Observational follow-up study 5.5.1 Characteristics of PHC workers…………………………………………………… 163 5.5.2 Mapping ‘community based’ health interventions………………………………… 163 5.5.3 Resources and expenditures incurred…………………………………………….. 165

5.6 Part C: Mapping NCD prevention and control interventions 167

5.7 Part D: In-depth-case-study analysis 5.7.0 Case-Study 1: Salt, Oil, Sugar (SOS) Reduction Program 5.7.1 Setting ……………………………………………………………………………. 169 5.7.2 Background………………………………………………………………………. 169 5.7.3 Short-film…………………………………………………………………………. 171 5.7.4 Intervention impact and challenges………………………………………………. 171 5.7.5 Resources used for the intervention………………………………………………. 173 5.7.6 Monitoring and evaluation strategies…………………………………………….. 174 Tools to explore the CP dynamics in the described intervention: 5.7.7 Applying the Project Cycle model………………………………………………... 175 5.7.8 Applying Rifkin’s Spidergram model……………………………………………... 175 5.7.9 Applying the P-Compass tool……………………………………………………... 177

5.8 Case-Study 2: Family Nutrition Program 5.8.1 Setting ……………………………………………………………………………. 178 5.8.2 Background……………………………………………………………………….. 178 5.8.3 Short-film…………………………………………………………………………. 179 5.8.4 Intervention impact and challenges……………………………………………….. 179 5.8.5 Resources used for the intervention……………………………………………….. 180 Tools to explore the CP dynamics in the described intervention: 5.8.6 Applying the Project Cycle model………………………………………………... 181 5.8.7 Applying Rifkin’s Spidergram model……………………………………………... 182 5.8.8 Applying the P-Compass tool……………………………………………………... 182 Section summary…………………………………………………………………... 184 5.8.9 Overview of focused ethnographic study results and reflections on methods 185

5.9 Results of document analysis 5.9.1 Domestic policy and legal framework analysis …………………………………... 189 5.9.2 Results of content review of PHC duty lists and training curriculum……………….. 195 5.9.3 Health Sector Master Plan ………………………………………………………. 197 5.9.4 Guidelines for Strengthening Health Care at Primary Level……………………… 197 5.9.5 Domestic legal frameworks pertaining to community participation in Health……. 198 5.9.6 An erosion of participatory space in health care decision making in Sri Lanka? 199 Section Summary…………………………………………………….……………. 203

13 | P a g e

5.10 Results of the Survey of Policy Makers 5.10.1 Demographic characteristics……………………………………………………… 206 5.10.2 CP in the policy making process………………………………………………….. 207 5.10.3 Results of P-Compass and PPD tool analysis……………………………………… 207 Section Summary…………………………………………………….……………. 211

CHAPTER 6: ANALYTICAL FRAMEWORK SYNTHESIS AND DISCUSSION 6.1 Chapter outline 213 6.1.1 Answering the research questions………………………………………………… 214 6.2 ‘Bringing it all together’: Synthesizing an analytical framework to explore CP 222 using a complex health systems approach……………………………………………… 6.2.1 Building the model………………………………………………………………... 222 6.3 Model elements and dynamics…………………………………………………… 225

6.3.1 Administrative and resource environment……………………………………… 224

6.3.2 Attitudinal environment…………………………………………………………. 231

6.3.3 Technical competency and knowledge environment…………………………... 235

6.3.4 Policy and regulatory environment……………………………………………... 239 6.3.5 Use and limitations of the model ………………………………..………………... 243

6.4 Conclusion……………………………………………………………………….. 245 6.5 Relevance for Policy and Practice …………………………………………….. 248 6.6 Recommendations for future research…………………………………………….. 251

REFERENCES 254

ANNEXURES 1 Systematic review of literature: perceptions of health workers toward CP in 274 health care………………………………………………………………………... 2 Review of literature examining CP in Sri Lanka’s health system………………….. 292 3 Ethical Approval ………………………………………………………………….. 313 4 Official endorsement letter from Ministry of Health, Sri Lanka………………….... 315 5 Semi-structured interview guide…………………………………………………... 317 6 Checklist for exploring CP within health policy documents………………………... 319 7 Policies, strategic plans and domestic legal frameworks pertaining to health care 320 in Sri Lanka. 8 Hypothesizing the potential linkages between domains 326 9 Some reflections on study challenges/limitations 329

14 | P a g e

LIST OF FIGURES

Figure 1 ‘Peeling the onion’: navigating through social, organisational and 16 setting specific layers to harness perspectives of health professionals Figure 2 Research domains explored in literature review 19 Figure 3 A graphical comparison of centralized vs. decentralized system 31 Figure 4 Link between PHC and preventive health action 45 Figure 5 Literature search strategy 46 Figure 6 Health administrative system of Sri Lanka 55 Figure 7 Literature search strategy 56 Figure 8 Points of participatory engagement within intervention cycle 58 Figure 9 Pathways of community ‘choice’ in accessing health services 62 Figure 10 The proposed model for PHC reform aimed at expanding health 64 service coverage to communities Figure 11 The WHO Health Systems Framework 66 Figure 12 A conceptual framework explore research objectives (‘Progenitor’ 71 model) Figure 13 Graphical overview of Chapter 4 78 Figure 14 Overview of study design 81 Figure 15 Overview of research methods, types of data generated and 84 modes of data analysis Figure 16 Geographical map of districts selected for study 87 Figure 17 The Community Participatory Compass (P-Compass) instrument 98 Figure 18 The four parts of the focused ethnographic and observational 103 study Figure 19 A stepwise approach to mapping PHC interventions aimed at 106 addressing NCDs Figure 20 Key elements in undertaking policy document analysis 110 Figure 21 Tool for assessing extent of participatory engagement within Sri 117 Lanka’s health system (‘PPD-Tool’) Figure 22 The collaborative process adopted for data transcription 128 Figure 23 Graphical overview of Chapter 5 131 Figure 24 Overview of factors inhibiting CP at PHC Level 139 Figure 25 The ‘filter and funnel effect’ 143 Figure 26 Overview of the focused ethnographic study results 158 Figure 27 Identifying the manifestation of CP based on the Project Cycle 175 Model Figure 28 Spidergram for Case-study 1 177 Figure 29 Identifying the manifestation of CP based on the Project Cycle 182 Model Figure 30 Spidergram for Case-study 2 183

1 | P a g e

Figure 31 Sri Lanka Health Policy Timeline (1978-2013) 192 Figure 32 Disbanding the National Health Development Network: An erosion 201 o participatory health care decision making space Figure 33 Participatory dynamics within and between the health system 202 Figure 34 P-Compass analysis: Plotting the level of CP in health policy 209 making Figure 35 Participation in Policy Development Tool (PPD) analysis: CP within 210 health system making Figure 36 Conceptual framework presenting the broad domains influencing 223 CP at PHC level in Sri Lanka. Figure 37 Key factors within the administrative & resource environment 230 Figure 38 Key factors within the attitudinal environment 232 Figure 39 Key factors within the technical competency & knowledge 238 environment Figure 40 Key factors within the policy & regulatory environment 242 Figure 41 Report card on the current status of CP within PHC services 248

LIST OF TABLES

Table 1. Research question and objectives 9 Table 2. Advantages and disadvantages of CP in planning and policy- 27 making Table 3. Barriers and enablers for successful public participation in policy 29 making Table 4. Summary of theoretical frameworks for exploring CP: Ladders, 37 spiders, continuums, rubrics, networks and lists Table 5. Summary of systematic reviews exploring CP and health 40 Table 6. Factors influencing CP 48 Table 7. Overview of the six domains used in health policy survey 114 Table 8. P-Compass Findings 155 Table 9. Human resources for health district mapping (2013): Active posts 161 and vacancies by district, MoH Area and specific cadre Table 10. Community outreach activities by PHMs 164 Table 11. Community outreach activities by PHIs 165 Table 12. Resource expenditures incurred by PHC workers in undertaking 166 community health interventions Table 13. Mapping NCD Prevention and Control Interventions 168 Table 14. Policies articulating CP as way of meeting policy goals 193 Table 15. Duties and work functions of PHC workers relevant to CP and NCD 196 interventions Table 16. Demographic characteristics 206

2 | P a g e

Table 17. CP in the policy making process 208

LIST OF BOXES

Box 1 Whitehead’s typologies of Community Based Interventions 24 Box 2 Decentralization and citizen involvement in health care decision 33 making in Finland Box 3 Characteristics of focused ethnographies 101 Box 4 Nuwara Eliya district mini-case study 162 Box 5 Disbanding the National Health Development Network 199

LIST OF IMAGES

Image 1. Mothers attending a community health clinic conducted by a Public 1 Health Midwife (PHM) in Northern Sri Lanka. Image 2. Community members at a community health nutritional program 17 held at a school in a rural village in North-central province, Sri Lanka. Image 3. A health promotion program targeting children and young 65 adolescents on diarrheal diseases at coastal community in Eastern Province of Sri Lanka. Image 4. PHM discussing with children and mothers attending a community 77 health nutritional program the plan of activities for the session. Image 5. A young mother discussing her involvement in a community based 130 NCD prevention intervention Image 6. Children participating in a community based nutritional program at 212 a pre-school setting Image 7. A community based psychosocial intervention targeting young 274 children living in a displaced camp in Northern Sri Lanka

3 | P a g e

Chapter 1:

Introduction and study rationale

Image 1. Mothers attending a community health clinic conducted by a PHM in Northern Sri Lanka. (Photograph: Kolitha Wickramage)

1 | P a g e

1.1 The ‘promise’ of community participation in health

“With an emphasis on local ownership, primary health care honoured the resilience and ingenuity of the human spirit and made space for solutions created by communities, owned by them, and sustained by them” - Director- General, WHO (1)

Community participation (CP) in health is a complex entity. As will be outlined in the literature review of the next chapter, its meaning is highly contested. The concept of CP in health first appeared over three decades ago as part of a movement for social justice (2). It was envisaged that basic health needs could be met more appropriately and efficiently by the greater involvement of people themselves (3). Indeed, the push for CP within health care systems has been linked to a larger movement of health sector decentralization (4,5). CP has also been viewed as a way of ‘legitimizing’ the actions of government health programs to ensure ‘community buy-in’ (6).

In 1991, the World Health Organization (WHO) (7) summarized the benefits of CP as:  Greater program coverage – involves more people than non-participatory approaches;  Efficiency – promotes better co-ordination and the use of resources;  Effectiveness – goals and strategies are more relevant as a result of participation;  Equity – promotes the notion of providing for those in greatest need and reduces unfair maldistribution of resources and services;  Self-reliance – increases people’s control over their own lives.

Enabling CP in health has therefore been advocated as a key component of a health system. In WHO's Alma Ata Declaration in 1978 it formed a fundamental pillar of the Primary Health Care (PHC) approach (8). National health systems have also enshrined CP as a crucial component of health care delivery. The United Kingdom’s National Institute for Health and Clinical Excellence, for instance, suggest that involving patients and communities should be compulsory for the providers of health services such as NHS trusts (9).

At the WHO conference dedicated to the 30th anniversary of the Alma-Ata Declaration on PHC held in Kazakhstan in 2008, member states of WHO re-affirmed their commitment to working in partnership with communities at the PHC level in addressing health challenges (10). The remarks by the WHO Director General at the conclusion of this conference, an 2 | P a g e excerpt from which was presented at the start of this section, reflects the centrality of CP in the ‘new’ PHC model. CP as a public health approach continues to be actively promoted within global health policy discourses, and by institutions as the WHO Commission on Social Determinants of Health (SDH). The lessons learned by the Commission from low-income countries emphasizes the importance of “CP in decision-making” as a key political factor in reducing health inequalities (11).

CP in health covers a vast range of activities: from vertically oriented information provision and consultation, to horizontally driven community development actions such as citizen- cooperatives within health care institutions. There are also varying degrees of participatory action (12), from passive compliance and tokenistic involvement, through to meaningful collaboration and shared decision making. These are elaborated in Section 2.2.1 of the review of literature.

Scholars have also indicated the importance of ensuring ‘meaningful’ forms of community engagement rather than “simply tick the policy requirement box” (13). ‘Meaningful CP’ refers to a process of community engagement that emphasises collaborative partnership between health care providers and community groups in a sustained and iterative manner. To make CP effective, it may be necessary to move beyond one-off or isolated efforts and consider how participation promotes equity, reshapes power and becomes embedded as an integral part of the relationship between health service provider and community. This view of CP is reflected by Rifkin and colleagues (1988) in presenting their primer on measuring CP in the context of PHC: “…this (14) process is one which focuses on the ability of community groups to improve their health and health care and by exercising effective decisions to force the shift in resources with a view to achieving equity” (15).

1.2 Community participation: a key strategy for NCD prevention and control

Collectively, chronic Non-Communicable Disease (NCDs) – conditions such as cardiovascular disease and diabetes – were responsible for 38 million deaths worldwide in 2012 (16). The countries most affected by this ‘silent epidemic of chronic diseases’ are low-and middle- income countries like Sri Lanka, where more than 28 million deaths occurred (16).

3 | P a g e

Statutory health authorities and international organisations now recognise the urgent need to respond to the growing burden of chronic diseases through structural reform of health systems and by forging partnerships within the health sector: between government agencies, the private and voluntary sectors, and most importantly with communities themselves (17,18). Because of the long-term and frequently complex nature of NCD management, positive health outcomes are best achieved according to WHO when those with the disease, their families, community supporters and healthcare teams ‘work together’ and are ‘informed, motivated, prepared’ (19). Within this partnership, Primary Health Care (PHC) workers have a central role to play in equipping communities with health information for primary and primordial prevention and skills for home based care, as well as organising and implementing health promotion and prevention programs (20).

The importance of enabling meaningful partnerships with communities at the PHC level in addressing NCD risk factors have been iterated in numerous international global health frameworks and World Health Assembly (WHA) resolutions. For instance, the Resolution made during the Sixty-first WHA in 2008 on the Prevention and Control of NCDs urged member states to support “integrated community-based, people centered primary health care services for health promotion, prevention, early detection and treatment of NCDs” (21). The High-Level Meeting on NCDs held at the UN General Assembly in 2011 saw world leaders gather in New York to address the growing burden of chronic diseases afflicting their population (22). Community participatory approaches were explicitly referred to (pp. 11), and were endorsed as a key element within the country action plans and enshrined in the Political Declaration at the heads of state meeting. The Global Action Plan for the Prevention and Control of NCDs for years 2013 to 2020 (‘GAP’) was formally endorsed by member states at the WHA meeting at Geneva in May 2013. It provides a road map and menu of policy options for governments, WHO, international organisations and civil society to combat NCDs (23). The GAP emphasised “the enhancement of community participation” as a core aspect of national response strategies for the prevention and control of NCDs (24).

The rationale for the sustained advocacy of CP models for NCD prevention at the PHC level was shaped through a growing body of research evidence showing the effectiveness of such participatory approaches, especially for primary prevention (25-27). Since PHC workers are a key ‘conduit’ between the health system and communities, professional attitudes and

4 | P a g e work place cultures play a key role in promoting meaningful partnerships with communities for the realization of NCD prevention goals (18) (28). Studies have shown that “professional uncertainty”, due to perceptions by PHC workers that CP may lead to erosion of ‘expert health authority’ and power, may actually hinder effectiveness of participatory processes (29,30). Enabling and supportive attitudes towards community participation from health care workers and health administrators are needed if communities are to be viewed as partners in health.

If CP is to be translated from ‘rhetoric into reality’, then the perspectives, commitment and attitudes of health professionals and decision makers that are key catalysts to community participation need to be assessed. However, as explored in the literature review, there has been limited examination of such perspectives (which I termed ‘attitudinal landscape’) and exploration of interventions using CP approaches (the ‘practice landscape’). This thesis aims at addressing this research gap (‘to map’ CP), within the context of Sri Lanka's health system.

1.3 Study rationale and justification

Well recognised for its advances and early commitment to ensuring primary health care ‘at low cost’ (31,32), Sri Lanka is now faced with a growing burden of NCDs, which now account for nearly 90% of the nation’s disease burden (33). Of the 11 South-East Asia Region countries, Sri Lanka ranks third with NCDs, accounting for 66% of total deaths (34). This has prompted calls for revitalization and reform of healthcare services in Sri Lanka to meet the chronic disease challenge (35).

Studies have revealed a number of factors driving the disease burden. These include: an increase in sedentary lifestyles across all socio-economic strata (36); changing dietary patterns and associated nutritional deficiencies (37); poor community based management of those already afflicted by chronic diseases (33); and weakened capacity of primary care services to deliver effective NCD prevention and control programs (38). Associated push factors for health system reform also include rising health care financing costs (32); professional and consumer expectations around ensuring continuity of care for those with chronic diseases; and a rapidly aging population due to Sri Lanka’s demographic transition (currently the third ‘oldest’ population in Asia) (39). The increasing prevalence of NCDs has therefore emerged as one of the most important public-health challenges in Sri Lanka (40).

5 | P a g e

The Sri Lankan Government, through the efforts of the Ministry of Health, has committed itself to a process of revitalizing PHC in order to meet such challenges (Section 2.3.8). Understanding the intervention space for NCD prevention and control at community level and the experience and expectations of PHC workers is therefore crucial to inform current government efforts.

Since the historic Alma Ata declaration in 1978, Sri Lanka has been an active supporter and signatory to major global health frameworks for ensuring health sector reforms, and has been credited with following ‘best practice’ for achievements in community focused health interventions (41,42). In recent years, Sri Lanka formally ‘signed on’ the Political Declaration on NCDs at the UN General Assembly in 2011, which articulates the importance of enabling ‘effective partnerships between the health system and communities’ to tackle the growing NCD burden. As a signatory to the 30th Anniversary of the Alma Ata Declaration in Jakarta in 2008, Sri Lanka reaffirmed its commitment to a return towards a ‘community centered health service platform’ and health system reforms where CP is an integral force (43).

Despite the historical commitment and current policy rhetoric of CP in health care in Sri Lanka, very little is known or has been researched on the values, knowledge, attitudes and perceptions of PHC workers, health system administrators and policy makers concerning CP. I argued that in order for such reforms to bear fruit within routine PHC practices, there is a need for a better understanding of how participation emerges, how it is enabled, sustained, and inhibited, and how innovation is translated into routine health services. Such understanding and analysis can contribute to promoting much-needed evidence-informed policy and practice (44).

There has not yet been a systematic exploration of how participatory approaches manifest themselves or are pursued within routine practices and formal work functions of the health system in Sri Lanka. This finding emerged through reviews of published and grey literature, and was emphasized through interactions with health care professionals and academics during the formative phase of this research.

Understanding how CP is enshrined within existing national health policies and legal frameworks, its manifestation in routine programs, and the perceptions by both policy makers and primary health care workers of its utility becomes crucial. Further, identifying

6 | P a g e factors that may sustain, limit or enhance a culture of community involvement in PHC and within NCD interventions may be useful for health policy makers and planners. Finally, given the resource implications of working intensively with communities over a long period of time to enable community and health service development (45), there is also a need for critical debate on the purpose of CP as well as evidence of the benefits claimed for it from the perspectives of primary care workers and health administrators. This research project, entitled “How participatory are we?”, aims to contribute to building such an evidence base, to support health care decision making and program planning.

The research is therefore framed within the perspective and tradition of Health Policy and Systems Research (HPSR) as iterated by the WHO that seeks to promote an evidence- informed decision making logic for revitalizing health systems (46). HPSR research enables the identification of gaps in capacity, barriers to efficient functioning and effective performance of the health system, analysis of partnerships with communities and methods by which the existing resources can be optimally utilized (46,47).

The research questions and findings will be useful not only for policy makers, health administrators and scholars currently committed to undertaking reform of Sri Lanka’s primary health care (PHC) system in alignment with the goals of the Alma Ata Declaration, but also to those interested in harnessing evidence on how CP manifests within routine PHC practice.

“Participation used to be the rallying cry of radicals. Its presence is now effectively obligatory in all policy documents and project proposals from the international donors and implementing agencies. Community Participation may have won the war of words, but beyond the rhetoric, its success is less evident. Part of the problem is clearly political. True participation is a threat to powerful and vested interests…”

- Dudley, E. (1993). The critical villager: Beyond community participation. Psychology Press, UK.

1.4 Overview of research aims and objectives

The primary research objective is encapsulated as a simple rhetorical question: ‘How participatory are we?’, where ‘we’ refers to Sri Lanka’s health system, and ‘participatory’ refers to CP in health care decision making and service delivery at the PHC level. The study sought to explore: the knowledge, experiences and commitment of PHC workers, health

7 | P a g e administrators and policy makers toward the use of CP in addressing NCDs at the PHC level; the extent to which community participatory approaches are utilized (if at all) at the PHC level to tackle the growing NCD burden; and the factors that enable, limit, enhance or sustain meaningful CP. This research aim was ‘deconstructed’ to form four research objectives (Table 1).

This research does not seek to empirically evaluate the effectiveness of CP in health care interventions, nor evaluate the extent of participation of Sri Lanka’s public health system. Rather, the research intends to explore if and how CP is manifest in PHC practice, the determinants of such manifestations, and, most importantly, the perspectives of the health professionals who work within the health care system - from ‘front line’ PHC workers to executive decision-makers.

As indicated in previous section, this research is framed from the perspective of the statutory health authorities (comprising policy makers, planners and health administrators) and of ‘front line’ PHC health care professionals, and not of communities. Documenting the experiences and perspectives of communities, patients and/or health care consumers in relation to CP in PHC is indeed essential to ‘complete the circle of understanding’ on complexity of the participatory dynamics. However, as described in Chapter 3, this warrants a dedicated research agenda, with sufficient time and resources devoted to such an in- depth exploration of the wider community voice. Whilst community perspectives on CP and PHC interventions targeting NCD have indeed been captured and analysed (Chapter 5), these should not be viewed as comprehensive analysis of the community perspective.

I was specifically interested in exploring the extent to which Sri Lanka’s health system engages with communities and utilizes community participatory approaches for the delivery of primary health care services, and in particular for those interventions targeted at addressing NCDs. The rhetorical question to which I repeatedly return: ‘How participatory are we?' aims at encapsulating this research perspective.

8 | P a g e

Table 1. Research question and objectives

Research Question: To what extent are community participatory approaches utilized in Sri Lanka’s primary health care system to address the growing burden of chronic diseases?

Research Objective 1: Analyse the extent to which CP approaches are enshrined within national health policies, domestic legal frameworks, PHC worker duty lists and routine job functions of Sri Lanka’s health system

Research Objective 2: Explore the knowledge, experiences, perceptions and attitudes PHC workers, health administrators and policy makers have toward the use of CP approaches for NCD prevention and control at the PHC level.

Research Objective 3: To explore the extent to which community participatory approaches are utilized (if at all) in NCD prevention and promotion interventions, and investigate the nature, form and rationale for their use at the PHC level.

Research Objective 4. Identify the major determinants and factors that enable, limit, enhance and sustain meaningful community participatory practices in Sri Lanka’s health system, through drawing on the experiences of PHC workers, health administrators and policy makers; assessing the health policy and operational environment; and synthesizing the evidence on NCD interventions within PHC settings.

1.5 Structure of the dissertation

The following presents an overview of the thesis:

Chapter 1 outlines the background and motivations for the study, and my perspective as the research investigator studying CP within a health systems approach. The research problem and research objectives are briefly introduced.

Chapter 2 presents a comprehensive review of literature encompassing the historical foundations of CP, its emergence as a movement in public health, definitional challenges and the rationale for its advocacy and use by statutory health organisations. This is complemented by a number of appendixes reflecting further work undertaken around the literature.

9 | P a g e

Chapter 3 presents an overview of the conceptual frameworks which emerged from a review of literature intended to explore the dynamics and factors of CP. Based on these models, the core theoretical framework and approach of my study is highlighted.

Chapter 4 explores the overall methodological approach and rationale for the ‘interpretive stance’ based on research objectives. The research methods, settings, participants, their selection and ethics relating to research are presented. The process of data collection and analytical methods are described. Issues pertaining to study rigor, assumptions and limitations of the research are also highlighted.

Chapter 5 presents the results of my study, structured according to methods of data collection: qualitative findings distilled from key informant and group interviews; results of the focused ethnographic study and follow-up observational study of PHC workers (including in-depth case-study analysis of two NCD prevention interventions); document analysis of domestic policy and legal frameworks; and finally, results of a survey of policy makers.

Chapter 6 integrates the key findings in relation to the research questions, and shows how they converge and contribute to the formation of a composite health systems framework to better understand CP within routine PHC practice. The comprehensive analytical framework that is presented utilizes a complex health systems approach to explore CP. There is discussion of the relevance of these findings to current PHC reform processes in Sri Lanka, and to efforts to combat NCDs in low to middle-income countries, along with recommendations for further research.

10 | P a g e

1.6 Researcher reflexivity: ‘Peeling the onion’ as research investigator

“Researchers see the world through different cultural, philosophical, or professional lenses” - Cohen (2000)(48).

In this section I reflect on my own values and beliefs concerning CP and motivations for undertaking the study. There were multiple events, inclinations, irregular and logical pathways that led to the formulation and construction of my research questions and final project. It is difficult to synthesise these into a single commentary. This section is therefore written as ‘reflective fragments’ that fit together to form the fabric of my research inquiry.

My early career and training began in virology and HIV research, where I worked at the Prince of Wales hospital in Sydney. I shifted to a public health focus after completing a post-graduate Masters in Public Health, and largely because I felt drawn to the rapidly evolving protracted civil conflict of my country of birth, Sri Lanka. Since 2004, I have worked in Sri Lanka as a public health project coordinator with United Nations agencies such as the WHO, to support the Ministry of Health in rehabilitating conflict-affected health systems and address the health challenges faced by populations forcibly displaced by war and the devastating Boxing Day Tsunami disaster. I managed a range of public health programs in the war torn North-Eastern provinces of Sri Lanka, such as the rapid establishment of basic primary care treatment centres, emergency medical referrals, camp- based infectious disease surveillance programs and health promotion interventions targeting diarrheal and vector-borne diseases. With the cessation of ethnic conflict and transition to post-conflict recovery, my role changed to supporting the MoH in health systems recovery and capacity building. For instance, between 2008 and 2009 I assisted in the implementation (and later evaluation) of a WHO-funded pilot program across 4 districts aiming to boost community mental health capacity at the primary care level. As Sri Lanka rapidly climbed into a post-war economic development boom, I witnessed a new battle emerging on the frontlines of the primary care interface – the growing burden of chronic disease.

I valued CP approaches in health service delivery not only for the enrichment they bring to both health professionals and participating community members – in terms of understanding the ‘other’, knowledge and skills empowerment – but also for its practical use in enhancing intervention coverage and diffusion in community settings such as internally displaced persons camps. Although challenging to implement in resource-limited, time-bound and

11 | P a g e donor-driven programs, I utilized CP approaches within specific public health interventions I managed. Such interventions ranged from community engagement in communicable disease surveillance and control programs within mass population displacement contexts (49), to the establishment of district level health promotion programs (50). Despite experiencing the challenge of harnessing such participatory approaches, I only began to critically reflect on the nature of participation and the scientific literature around participation much later in my career. In completing a Masters degree in Human Rights, my thesis had explored the ‘right to health’ among war-displaced communities in Northern Sri Lanka. An integral aspect of the right to health is the ‘right to participate’. The training and exposure to rights based discourses stimulated my desire to explore participation in health as a right of citizens, and the role of the state health system in ensuring this right.

After undertaking a comprehensive review of literature on CP, two key findings emerged. First, there remains a persistent rhetoric emphasising CP in enhancing health services, as echoed in Alma Ata in 1978. Second, even taking into account the difficulty in measuring CP, there is limited evidence for its effectiveness (51,52). Where positive links were found on CP and improved health status, they were context-specific and not generalizable (53).

There also appears to be a scarcity of studies that adopt more complex health systems approaches to explore CP. As Rifkin (2014) indicates, despite the relatively large body of published literature on CP over the past 30 years, “new frameworks of analysis are needed” because CP is often “seen only as a solution to one particular health problem without considering other systemic factors” inherent within the health system (51). Other authors have also highlighted the lack of a comprehensive and generalizable approach to understanding CP that embraces the complexity of modern health systems (52).

Genesis of the primary research question

During the formative research phase, I consulted senior policy makers and directors from the Ministry of Health, including a number of country representatives of development partner agencies, who spoke of the need to ‘find real solutions’ in enhancing the way Sri Lanka’s health system can work better with communities at the PHC level. In the early part of my research I also had the opportunity, at the request of the MoH, to co-organise a two-day technical workshop for senior health system analysts from the MoH and agencies like WHO, UNICEF and IOM on the topic of revitalizing the conflict-affected PHC system in the Northern 12 | P a g e

Province. This workshop was part of a series of three national workshops aimed at ushering in new PHC system reforms and redefining the role of PHC workers in the context of a growing NCD burden1. The workshop highlighted the limited information and evidence base for CP approaches available to health authorities (and their discontent!) in planning strategies for PHC reform. The scarcity of published research on exploring CP in Sri Lanka was also noted. At the conclusion of the formative research phase, I identified the following knowledge gaps in exploring CP:

 How is CP defined and how is it enshrined within existing national health policies and legal and operational frameworks of the health system?  How do professional cultures and PHC practice dynamics and organisational ethos enable, limit or enhance a culture of CP in health?  What is the effectiveness of CP in addressing health outcomes and social determinants of health?

The primary question which emerged from this formative phase, and which rhetorically echoes throughout this research is: ‘How participatory are we?’ How much emphasis does the current health system place on the use of CP? What is the value and meaning we place on community engagement within routine primary care practices? Do we enable and facilitate CP in practice alongside the ongoing rhetorical commitment as an idea?

From a philosophical perspective, I was keen to explore how the pervasive rhetoric of CP enshrined in various government policies has been translated into the reality of health service delivery at the primary care interface. I was also curious as to why research on CP was scarce despite the fact that Sri Lanka’s public health system has been universally heralded as a model of PHC service delivery, especially since CP is a core pillar of PHC.

From a methodological perspective, I was drawn to the challenge of exploring the complex, multi-layered social phenomena of community participation through an evidence- based approach. Rather than ‘measuring’ CP to gauge its effectiveness in meeting specific health intervention or program goals, as has been done across many settings previously, I sought to describe how it is defined, manifested, diffused or is rejected within routine PHC

1 I have analysed the content of these workshops and presented these in the form of a paper to be published. 13 | P a g e practice. Rather than ‘measure’, the challenge was to capture the broader underlying determinants of its use and emergence within district health systems.

Having worked with health systems in mainly resource limited developing country settings over the past decade, it is my firm belief that health system issues are best understood by using multilevel approaches. It is not by reducing complexity but rather uncovering it, layer- by-layer, and utilizing a multiple method approach that I feel one may be better able to understand the dynamic of CP in a health system. This encompasses policy framework analysis, direct observations of practice, capturing worker perceptions and by looking back at organisational histories. International debates on improving health system performance and quality of care are strongly influenced by systems thinking that embraces multiple perspectives (47,54).

From an empowerment perspective, I sought to harness the perceptions of both ‘front-line’ workers and ‘board room’ level policy makers, in understanding their constructions of CP, how it emerges and is sustained in routine practice.

The efforts of some key leaders within the Ministry of Health in taking a pro-active approach to push for a PHC reform agenda against a backdrop of the growing burden of chronic diseases is to be commended. Despite engagement with such processes at central level the question remained: where are the perspectives of front-line primary care workers in all of this? Rather than viewing frontline PHC workers merely as ‘conduits’ for health service delivery at community level, health workers across all levels need be engaged as protagonists for new ideas and innovations. Their views and perspectives are vital for any exploration of CP in PHC. Indeed, health systems research that puts the ‘health worker at the centre’ of the research agenda is increasingly been acknowledged as a ‘missing branch’ of systems research (55,56).

From a pragmatic perspective, I intended this research to ultimately make a real difference to the way health systems engage with communities by offering insights to the push and pull factors that catalyse or inhibit participatory practice. I intended to contribute to the development of a theoretical compass to explore CP within a PHC system, thereby enabling frontline workers and policy makers to make better-informed decisions. A compass is not a

14 | P a g e prescriptive map of instruction or didactic information, rather a tool to help health authorities and researchers navigate and appreciate the complexities of CP in health.

‘Peeling the onion’ as research investigator

I migrated to Australia from Sri Lanka in the late 1980’s to escape the protracted civil conflict in Sri Lanka. Despite my Sri Lankan identity, there were occasions where some health workers would perceive me as a ‘foreign other’. At its worst, this ‘othering’ attitude was expressed through subtle actions of resistance by a medical officer in charge of a PHC unit to withhold cooperation in undertaking the field research component for that particular PHC setting. Although this was resolved relatively quickly through official channels of correspondence, it highlighted the many sensitivities of undertaking research in Sri Lanka’s post-conflict society where notions and fears of the foreign ‘other’ are still prevalent.

I constructed a simple model of how I had to navigate the multiple layers of social-cultural sensitivities and the organisational layers of the Sri Lankan health system in order to undertake the research - a process I entitled ‘Peeling the onion’ (Figure 1).

15 | P a g e

Figure 1. ‘Peeling the onion’: navigating through social, organisational and setting specific layers to harness perspectives of health professionals

The diagram is also a reflection of my perspective as a researcher: the person I think I am (in my beliefs on CP for instance), the person health authorities think I am (perception of insider/outsider), the person communities think I am (a health professional interested in understanding ‘their voice’). How such ‘positionality’ is negotiated and how it may influence data collection and interpretation is a process I termed ‘peeling the onion’.

Extensive efforts were made in providing information and education to all participants on study goals and consent procedures. Beyond the usual ethical approval processes with central level MoH authorities, endorsement of the study was obtained at district, provincial and divisional (village) level health administrators.

16 | P a g e

Chapter 2:

Literature review Part 1: Community participation: definitions and practice

Image 2. Community members at a community health nutritional program held at a school in a rural village in North-central province, Sri Lanka.

17 | P a g e

2.1.1. Structure of literature review

A wide range of research perspectives may be adopted not only for defining CP but more importantly for examining the phenomenon of CP in health (51). The narrative reviews embedded in this chapter aim to unpack this complexity and outline the rationale for exploring the research questions identified in the previous chapter. Anchored to a complex health systems framework, my research aimed at exploring CP from the perspective of health professionals in Sri Lanka’s health system. I was motivated to explore the meaning, value and commitment placed on CP by ‘front-line’ PHC workers to ‘board-room’ policy makers.

The first section of this review aim at unpacking the definitions of CP. The second section explores the praxis of CP within PHC service delivery, and examines its utility in NCD prevention and control. The third and final component of the review focuses on the country context, highlighting current efforts to reform Sri Lanka’s PHC system and the challenge of addressing the growing burden of NCDs. Figure 2 presents an overview of the four research domains and the relationship of these literatures.

The first domain - ‘CP in health’ – examines the meaning of the concepts of ‘community’ and ‘community participation’ in relation to health and health systems. The section describes the historical foundations of CP, its emergence as a movement in public health, its evolution within global health policy discourse and its use by statutory health organisations. The literature dealing with the challenges in evaluating the effectiveness of CP is also presented.

The second and third domains explore the global frameworks promoting the use of CP as a strategy in PHC, and as a strategy in addressing the burden of NCDs at the community level in light of empirical research. The factors that may limit or enhance its diffusion in PHC programs are also described. The review then deals with the importance of health care worker attitudes in building CP, and there is a focused systematic review of literature aimed at mapping health worker attitudes to CP.

The final domain of the review provides an overview of Sri Lanka’s historical commitment to PHC, the evolution of health sector decentralization, the organisation of community health services, and the phenomenon of ‘bypassing’ primary health care and its relevance for CP.

18 | P a g e

This final component focuses on the challenges of NCD prevention and control in Sri Lanka given current efforts to reform PHC services.

The domains of literature reviewed follows a logical approach in an effort to contextualise the debate, ranging from conceptual definitions to exploring the intervention space in Sri Lanka itself.

Figure 2. Research domains explored in literature review

2.1.2 ‘Unpacking’ community participation: definitional challenges

The Alma Ata Declaration heralded by WHO in 1978 advocated for community participatory approaches as a core pillar in re-aligning health systems for PHC (57). Community engagement in health care continues to be embraced as central to the primary care strategies of many countries (58-61). Indeed, the idea of CP continues to have currency with health service development and system planners in both developed and developing contexts. The concepts of CP, along with its variants such as ‘consumer involvement’, have become a central tenet of the public health movement (62,63). CP in health activities is assumed to facilitate both health and social benefits for participants and has been promoted as an essential component of effective, efficient, and locally owned health care. The rationale for CP in health care may be viewed in part as contributing to a movement towards decentralization in health systems (4,5), to ensuring and promoting greater accountability and democracy (64,65), and to enabling a ‘rights based’ approach to health 19 | P a g e

(66,67). The outcome of participation may be seen as the improvement in quality of care, appropriateness of health service delivery for users, patient satisfaction and enhanced utilization (68).

2.1.3 Defining community participation

“Participation has unfortunately become a cliché word which makes many people squirm, long before it has developed any content. Everyone plays lip service to it, without having much idea as yet of what it means and how you achieve it, and often also without genuinely wanting it except to the extent they can use it to further their own ends” - Bryson, (1981) (69). As the above quote suggests, the term ‘participation’ is considered so self-evident and ubiquitous that it is rarely defined in the health literature (70). Analysts are confronted with a multiplicity of definitions for CP in which its meaning is highly contested (71,72).

Two broad conceptual components are evident, however: participation as a ‘means’ and participation as a ‘process’ (15,73-76). CP as a ‘means’ emerges from a somewhat technocratic approach in which a health system or program utilizes CP approaches as a means to ensuring 'successful' intervention goals, or as a strategy for achieving particular outcomes, whether these are for the system or its beneficiaries. CP is framed as a means of legitimizing health goals and facilitating the implementation of health programs.

CP can also be viewed as a desirable process and a goal in itself (73, 74). Oakley (1999) articulated this by defining CP as the “empowerment of people in terms of their acquisition of the skills, knowledge, and experience to take greater responsibility for their development” (74). The WHO also defines CP within an empowerment framework in which participation enables people to become actively involved in defining the issues of concern to them and in reshaping and ultimately influencing changes in services at a local level (7). It provides a method for assessing local needs, identifying priorities and empowering local people as active agents of change. Within this definition, CP contributes to decreasing marginalization and increasing inclusivity, access and equity (77).

Rather than pursuing CP to ensure a better ‘technical fit’ or to derive efficiencies for (often) vertically driven public health interventions, this typology of CP is dominated by horizontal notions of shared ownership and power for decision making in health. The Centre for Disease Control (USA) defines CP as a ‘powerful vehicle for bringing about environmental and behavioural changes’ that will improve the health of the community and its members (78). CP

20 | P a g e may thus be defined as a process that ‘gives voice’ to the community and, in turn, builds sustainable integrated communities which themselves are crucial to generating social capital and addressing inequalities in health (79).

To summarize, CP may be ‘constructed’ in two ways: first, within a technocratic framework that seeks to deliver effective interventions for beneficiaries/patients; second, within an empowerment framework which seeks to promote the acquisition of skills, knowledge, and agency to influence their own health and wellbeing. Within the empowerment approach, health systems and health care providers work ‘with’ communities rather than ‘on’ them.

2.1.4 Defining community: consumer or community participation?

As with the term ‘participation’, the definition of ‘community’ has been contentious, and the term used rather loosely (80). Even within the CP literature I reviewed, the 'community' that participates was not clearly defined and in most publications was ignored altogether. Moreover, there was evidence to suggest that definitions of CP vary by professional category of worker and geographic setting (81).

A study in the UK showed GPs were more likely to define community as the ‘practice population’, while practice nurses viewed community in terms of either locality or shared interest groups (82). In a study conducted in New Zealand, GPs and heads of primary care clinics viewed community members enrolled within their practice registers as the ‘community’ (13). GPs in the study also defined participation as a means of ensuring a ‘quality improvement process’ that encompassed consumer feedback, through use of patient satisfaction questionnaires.

It appears that when people are defined as ‘consumers’, the aim is to increase participation of such individuals in a health program or service. In the consumerist model, it is the individual, rather than a group, that ‘participates’ within a given health care program or setting. This concept reinforces the notion of health care as a ‘product’ to be consumed, rather than a process of iterative engagement with the service provider. It places emphasis on individual responsibility for health rather than a process of collective action (80). These concepts of consumer-driven rather than community participatory approaches to health,

21 | P a g e according to Baum (1996), fit within a “managerialist approach to health service provision” (83). From this perspective, CP usually involves a degree of ‘representation’ from those interacting with or using a health institution within a defined geographic catchment area (84).

Current economic reforms and public sector restructuring reflect the dominance of market economics in health and other sectors. This, along with the push for health sector privatization and service arrangements between providers and health care financing, has contributed to the development of a consumerist definition of community (80,85,86). Ryan (2001) argued that constructing communities as consumers has transformed the interaction between a service user and health provider to a “passive commercial transaction rather than an interactive political engagement” (87). Gideon (2005) analysed the changing landscape of CP in Chile, where health consumerism has been actively supported by civil society and community advocates that had traditionally rallied against such models (88). The co-option of NGOs and civil society groups by government health authorities with minimal challenge has left health policy-making largely a ‘top-down’ process.

Johnson (1996), in her research on PHC management, argued that while a reliance on comprehensive forms of community engagement greatly contributes to the process of defining a shared vision between communities and statutory health authorities, it may also lead to the impression “that the service is unfocused” (89). Johnson argued that a health service based on the principles of PHC “might from time to time look as if it lacks direction. This can happen when the service is actively seeking community direction rather than planning and making the decisions itself” (89).

In reviewing the literature, it becomes clear that there are differences between community and consumer participation at both pragmatic and conceptual levels. It is also apparent that how CP is articulated and defined by decision makers and health professionals becomes important in the pursuit of ‘community based’ health solutions.

If health system planners and PHC workers are to work effectively in a participatory manner then understanding how they conceptualize CP becomes crucial. As Taylor suggests, “understanding of community narratives about community participation is essential for planners

22 | P a g e and policy makers” (90). Indeed, one of the objectives of my research will be to explore what ‘participation’ and ‘community’ mean to PHC workers, health policy makers and planners, and to consider the values and duties implied by participation. Whether it is consumer participation, community participation, or both that is required, a better conceptual understanding of participation by health authorities may be needed.

2.1.5 Defining ‘community based’ interventions

A ‘perpetual allure’ of community-based health interventions is the promise they bring as a cost-effective means of quickly increasing program coverage and showing generalizability of health program benefits (91,92). Yet, just as ‘community’ and ‘participation’ have a wide range of meanings, understanding what exactly constitutes a ‘community-based’ health intervention (CBI) is also challenging.

CBI studies have been typically implemented according to geographic units or boundaries such as cities, counties, or villages (93) (91). Empirical reviews of community-based intervention studies also include small social units such as workplaces and schools, alongside geographically defined groups (94,95), as constituting the intervention space. McLeroy et al. (2003) define four categories of CBIs based on implicit constructions of community: community as ‘setting’, community as ‘target’, community as ‘agent’, and community as ‘resource’ (96):

(1) Community as setting - the community is defined geographically and is the locus in which interventions are implemented. The focus is primarily on changing individuals’ behaviors as a method for reducing the population’s risk of disease. Promoted by WHO as the ‘settings based approach’, such interventions may be citywide, using mass media or other approaches, or may take place within community institutions, such as neighborhoods, schools and work sites. Communities may be engaged through advisory committees or community coalitions that can assist in tailoring interventions. (2) Community as target - refers to the goal of addressing specific health status indicators of a community. (3) Community as resource - is commonly applied in health promotion initiatives that inculcate a high degree of community ownership. As McLeroy indicates, such

23 | P a g e

programs are “aimed at marshaling a community’s internal resources or assets, often across community sectors, to strategically focus their attention on a selected set of priority health-related strategies”. (4) Community as agent – emphasizes communities as agents to support their own development, although he warns that “communities are defined as much by whom they exclude as whom they include”.

Unlike McLeroy, Whitehead (2002) (97) defined CBIs based upon who initiates, drives, and carries out the intervention. He proposed seven types of CBIs, ranging from self-sufficient programs, driven and funded exclusively by the community (‘Type 1’: see Box 1 below), to those planned and implemented as equitable partnerships by the community in collaboration with an external change agent (‘Type 7’). The continuum involves the recipient community to different degrees, from passive program recipients to active partners in program implementation, with the ‘ideal’ CBI being a true partnership between technical experts and the communities they serve. The former contribute conceptual strength, comprehensive design, and rigorous implementation, while community endorsement and support increase the likelihood of the program becoming incorporated into its sociocultural context, strengthening sustainability, ownership and diffusion.

Box 1. Whitehead’s typologies of Community Based Interventions (97)

 Type 1: Programs initiated by individuals or groups indigenous to the community to be served by a program (target community), without any external (to that community) support  Type 2: Programs initiated by individuals or community groups indigenous to the community, who also recruit external, technical (expertise) support  Type 3: Programs in which individuals or community based organisations (CBOs) pursue external fiscal support or funding  Type 4: Programs in which individuals or CBOs indigenous to the target community initiate and recruit external technical and fiscal support  Type 5: Programs which are initiated by external change agencies (e.g. a local NGO) within a target community, but without any input from individual residents or organisations of that community, except as program recipients;  Type 6: Programs which are planned and initiated by external change agencies, and in which community members are eventually invited to participate on community advisory committees, or as lower level project staff such as "community outreach workers", or as volunteers  Type 7: Programs which are planned and implemented as an equitable partnership by CBOs and an external change agent or technical organisation.

24 | P a g e

The differing typologies illustrate the difficulties in summarizing results across the array of CBIs, although both authors recognise that some interventions may have characteristics of multiple forms. Common to both is the use of social ecology as a framework for exploring CBIs. A socio-ecological approach emphasizes the interdependence of social systems that encompasses the interactions within social, institutional, and cultural contexts that contribute to health and wellbeing.

2.1.6 Community participation in relation to good governance and democracy

“Every couple of decades governments decide that they need to involve citizens more in public decision- making processes…a growing loss of faith in the traditional institutions of government have once again prompted political decision makers to explore options for enhanced citizen participation” – Chung (2012)(98).

This section of the review explores the value and rationale for its use within health systems and, more importantly, for health care reform. A key objective of my research (Objectives 1.1 to 1.3) is to explore the extent to which communities participate in health policy and decision-making. This component draws upon the broader literature on governance and politics, going beyond the confines of the public health and health management literature.

The statement by Chung (2012) at start of this section suggests the ‘vogue’ of engaging consumers in decision-making. Public policy theorists and political scientists have long promoted the value of CP in policy making (99) (100). Two broad philosophical perspectives are cited as driving the need for CP. Within the ‘participatory democracy’ approach, CP offers opportunities for communities to participate in actively ‘shaping their world’ (101). This approach is also labelled by some scholars as ‘deliberative democracy’ (100), and aims at developing policies and designing services that respond to the local, spatial and context-specific needs of communities. A fundamental tenet in support of this approach is for CP to create notions of ‘legitimacy’ in priority-setting and decision-making for the government authority (102,103). Abels (2007) indicates that engaging communities for policy making and planning processes is done “in order to legitimize governance decisions” (104). However, in the pursuit of seeking legitimacy and representativeness, CP can deliberately be engineered towards creating symbolic rather than meaningful or substantive forms of collaborative community engagement (103). 25 | P a g e

The second perspective driving the need for CP seeks to harness the collective wisdom of communities to enrich policy formulation and organisational practice (105). Such citizen engagement is ‘not only the right thing to do but will provide a rich new source of ideas to government’ (106). A multiplicity of ‘entry points’ for CP in decision-making is also apparent. These include CP through civil society organisations, health consumer groups, community websites, incorporated entities, NGOs and peak bodies, and more formally organised forums such as citizen committees at hospitals and health trusts (107).

By engaging with citizens, governments can benefit from expert knowledge beyond their immediate realm of information, expertise and advice, while simultaneously creating opportunities to educate people about policy alternatives. This perspective assumes that by enabling CP in health governments may develop a better appreciation of public opinion, and may seize the opportunity to challenge, inform and shape people’s preferences. Officials can also test the public’s likely reaction to a policy proposal.

Irvin and Stansbury (2004) reviewed political science and policy discourses and suggested a list of conditions under which CP may be advantageous or disadvantageous for policy making (108). These are summarised into a table format in Table 2. Despite the many advantages of investing in CP processes such as building trust and mitigating public anxiety towards policy decisions, there were also real disadvantages, such as cost, time and participatory trajectories that may actually backfire and create more hostility toward government.

26 | P a g e

Table 2. Advantages and disadvantages of CP in planning and policy-making

Advantages to: Disadvantages to: Citizen participants Government Citizen participants Government  Education (learn  Education (learn  Time consuming  Time consuming from and inform from and inform  Pointless if input is  Costly government citizens) ignored  May backfire, representatives)  Better policy and  Limited creating more  Persuade and implementation participation by hostility toward enlighten decisions marginalized or government government  Persuade citizens; linguistically  Loss of decision  Gain some control build trust and diverse making control over policy process allay anxiety or communities  Scope for poor  Better policy and hostility  Worse policy decisions that are implementation  Build strategic decision if heavily politically decisions alliances influenced by impossible to  Gain activism skills  Gain legitimacy opposing interest ignore for decisions groups  Reduced budget  Avoid litigation for implementation costs

27 | P a g e

Translating rhetoric to reality: emerging lessons in use of CP for decision making

“In a world where large organisations control wide swaths of both public and private sector activities, the possibility of establishing more locally operated, locally responsible institutions holds out great attraction” – Saltman (2007) (109).

2.1.7 Creating an enabling policy environment

A comprehensive global assessment of CP in public policy making in human services was undertaken by the Organisation for Economic Cooperation and Development (OECD) in 2001 (110). This assessment was influential in guiding CP in governance within democratic systems, including the health sector (101,103). The report concluded that while many nations articulated the need for CP in delivery of human services, efforts to engage citizens in policymaking are still often undertaken on a ‘pilot basis’. Whilst there were regulatory provisions for participation enshrined within most domestic and regional coordination frameworks (for instance at European Union level), these were ‘mostly experimental’ (110). A follow-up report in 2009 showed considerable progress had been made across a greater number of countries to ‘genuinely strengthen’ and mainstream CP in policy development concerning the design and delivery of services (101). An important finding of the 2009 report was that CP is often shaped and conditioned by a local culture of ‘civic traditions’, and thrived in countries with vibrant civil societies. Although smaller in scope to the OECD assessment, scholars at the European Institute for Public Participation undertook an in-depth review of EU nations to identify requirements for successful CP in public policy making (111,112). In assessing the three reports, I summarised the key barriers and enablers for successful public participation in policy making (see Table 3).

A major finding of these assessments was that for CP to ‘thrive’ there needed to be explicit formulation of CP within domestic policy and legal instruments. A key component of my own research drew on this and focused on how CP is enshrined and applied within the guiding domestic legal and policy frameworks of Sri Lanka’s health system.

28 | P a g e

Table 3. Barriers and enablers for successful public participation in policy making

Enablers for successful public participation Barriers to participation  Clearly defined constitutional legal framework for  Costs (time, money, political public participation cost)  Systematic approach to public participation via  Complexity enabling public participation processes  Undertaking research on CP and rigorous evaluation  Representativeness

Investing in dedicated programs for CP was also highlighted as a ‘priority action’ for those authorities with an interest in sustaining CP mechanisms. Without such stimulation, innovative participatory approaches may not emerge within routine practices. Research Objective 3 of my study sought to assess the nature, form and extent to which CP practice emerged within routine health system settings.

The review also outlined the need to create a culture of research and evaluation of CP at country level in order to ensure continued financial support and evidence-informed development. The importance of research was emphasized in order to stimulate shared learning and development of CP methods that work.

2.1.8 Political environment

Political commitment to CP is another major determinant of successful community-engaged policy and programming. President Barack Obama, as his first executive action upon assuming office in February 2009, issued to all government agencies a Memorandum on Transparency and Open Government aimed at ‘ensuring public trust and to establish a system of transparency, public participation, and collaboration’ (113). The British Prime Minister David Cameron declared in July 2010 that ‘community participation is at the heart of public sector reform’ and called for a reform in governance that ‘unleashes community engagement’ and devolved decision-making (114).

As already indicated, if participation is poorly handled, it may serve to disenfranchise civic engagement and future funding potential (115). However, exposing political decision- making to the ‘collective wisdom’ of citizens can benefit even complex, technical policies in the arena of health, energy and transport (116). The premise here is that appropriate forms

29 | P a g e of CP can indeed resolve problems and enable a deliberative and collaborative search for solutions (117).

2.1.9 Politics of participation

A common way health authorities seek to engage communities within decision-making processors is by employing ‘community representatives’ through establishment of consumer panels or policy advisory committees. Such civil formations may minimize cost factors by averting the need for statutory health institutions to undertake large-scale community consultations. They may also serve to ‘legitimize’ health programs undertaken by the health system with the endorsement of such panels (118). A key challenge that emerges concerns the ‘representativeness’ of those ‘representing’ the community. Translated as a simple rhetorical statement in Latin: “Quis custodiet ipsos custodes?” - who watches those community representatives?

Achieving community representation within the modern world that encompasses a diverse and heterogeneous society is indeed a challenge (119). Scholars argue that, in encouraging diversity in community representative processes, authorities should have the skill to identify the inherent weaknesses of such representatives and the power-dynamics between groups when deliberating CP processes (120).

30 | P a g e

Why community participation?

2.1.10 Health systems reform through decentralization and the promise of CP "Decentralization is central to participation. It involves the increased involvement of local jurisdictions and civil society in the management of their affairs, with new forms of participation, consultation, and partnerships." – The Earth Institute (2013)(121).

Figure 3. A graphical comparison of centralized vs. decentralized system (122).

Decentralization, which refers to the transfer of power, authority and functions from central to local authorities (123), has been recognised as an important means of improving delivery of public services (124). In centralized systems individual units (represented by nodes in Figure 3), like local governments, are directly controlled by a central power. In decentralized systems power is distributed in a hierarchical fashion such that there are middle tier powers between the central and local nodes.

Health sector reforms during the 1990s were explicit in aiming to enhance service responsiveness and improve accountability (125). Decentralization in health care was viewed as an essential ‘administrative reform’ needed to improve the efficiency and quality of services and to promote democracy and accountability to the local population (126). Indeed, WHO recommended the importance of decentralization as crucial for health care reform for its member states (127).

31 | P a g e

CP and decentralization are intrinsically linked through the idea that partnerships between local communities and health systems are essential in order to ensure culturally appropriate, conflict sensitive, needs specific and locally responsive levels of health care (128,129). A decentralized system is seen as complementary to enhancing CP since it (theoretically) emphasizes the values of shared decision-making, consolidating partnerships with local health authorities and fostering local or indigenous capacities to transform health.

2.1.11 Evidence of decentralization in health systems and link to CP

Decentralization is core to the realization and effective implementation of the PHC approach (130,131), with evidence suggesting that CP can add value in the delivery of more ‘selective’ packages of PHC interventions (20,132). Decentralization has also been linked to greater opportunities for CP and an enhanced democratic voice (133,134). A World Bank study showed that of 75 developing and transitional nations having populations greater than 5 million, 63 have initiated reforms aimed at transferring political power to local units of government (135). The main goal of such reforms was to enhance equity, increase efficiency and ensure more participation and responsiveness of government to citizens. The UN Development Programme (UNDP) recommends decentralization needs to adopt approaches in which communities should be seen as an ‘entry point’, since many innovative solutions are “most likely to emerge from the people themselves” (136).

Given the wide variation in the organisation of health systems, as well as differences in democratic processes, there is no ‘standardized recipe’ of decentralization (137). Pursuing meaningful CP in health may also necessitate overcoming entrenched cultures of inhibitory professional attitudes, in which participatory approaches are not valued, and local politics and centralized bureaucracies may be antagonistic to reforms for the sharing of power.

Administrative reforms alone may also be inadequate to promote CP. A review of low and middle income country contexts revealed that observed patterns of effective CP were more closely linked to a high degree of political commitment to decentralization, than to existing national policy commitments for CP (126). In the absence of such high level commitment, meaningful community involvement may generally be found only in small-scale programmes, in specific locations, guided by highly dedicated and charismatic individuals (138-140). The literature demonstrates that effective decentralization cannot be achieved only through 32 | P a g e devolution of ‘administrative’ power, but must be accompanied by a range of measures to stimulate enabling processes that empower communities and make participation more feasible.

Scandinavian countries have some of the best healthcare systems in the world, with high levels of public satisfaction with health and social services (67). This importance may in part be due to a key underlying value within these societies, focusing on equity and prioritizing participatory engagement (141-143). Another major push factor has been local and regional governance structures that actively promote innovative practices from peripheral health units and village-level health institutions (144). Case Box 2 (see below) provides a case study from Finland concerning efforts made by health authorities to enhance citizens’ right to participate in health care decision making. Similar approaches to ensure community involvement in health care policymaking and program formulation have been described in the British (145), Australian (146), and Irish health systems (147,148).

Box 2. Decentralization and citizen involvement in health care decision making in Finland

Finland is ranked 31st in WHO’s ranking of Health Systems, despite healthcare spending per capita being amongst the lowest of OECD countries (149). Finland is divided into self-governing municipalities, which are autonomous and responsible for providing – among other things – health and social care services. The most important channel for the public to participate in health decision- making is through municipal council health committees which have community boards to represent the local population (150). Health services are therefore tailored according to the needs of local residents. There are also various patients’ associations and community clubs which lobby decision- makers on health care issues (150).

33 | P a g e

Literature review (Part 2) Theoretical frameworks for exploring CP and evidence of effectiveness

2.2.1 Theoretical frameworks for exploring CP: Ladders, spiders, continuums, rubrics, networks and lists! The challenges in defining CP also extend to the process of describing its form and evaluating its practice in real world settings. With the increasing purchase and promotion of CP in health systems globally, there is an increasing need for conceptual models to document and describe participatory processors in health. Conceptual frameworks are intended as a resource for reflection in order to explore complex constructs. I undertook a comprehensive review of literature to explore the theoretical frameworks that aimed at exploring the various dimensions of CP. I did not limit my search to those models that were only geared at measuring participation from a health system perspective, but included approaches from the broader social science, political science and community development literature.

The purpose of reviewing such theoretical literature was to shape my own research frame and conceptual analysis (presented in the next section). As noted earlier, I was concerned less with evaluating CP, but with in-depth qualitative analysis of the dynamics of participatory practice from the perspective of health workers, policy makers and planners in Sri Lanka’s health system. More specifically, this framework analysis aimed at:  Aiding the construction of an overall theoretical framework to structure my research objectives (presented in Section 3.1)  Moulding a reflective tool (entitled the ‘Participatory compass’) that was applied during interviews and within surveys to capture the participatory dynamic within specific PHC settings (Section 4.3.3)  Assessing CP in the NCD prevention intervention case-studies (Section 5.7)

A number of distinctive patterns emerged in my synthesis of the literature. All theoretical approaches grappled with two definitional perspectives of CP: those that were constructed to assess participation as a process and those that looked at evaluating CP focusing on specific effects and outcomes. Numerous models have been formulated that were based on

34 | P a g e one or more of these processors (see Table 4 below), with some aiming to combine both processors within a multi-dimensional rubric model. Butterfoss (2006), for instance, articulated CP as a process along a continuum which enabled communities to maximize their potential and progress from individual action to collective social and political change, whilst Rifkin’s ‘spidergram’ (1988) (151) utilized a hierarchical scale to numerically rate the level of CP from restrictive engagement to iterative collaboration and partnership.

Rifkin’s spidergram remains the most widely used model, in both developing and developed country settings. This may stem from its appeal as a tool for providing a scalable ‘measure of CP’ (from 0 to 5) that could be visualised across factors considered to influence the breadth or depth of CP. Within the health literature, there is a greater emphasis on formulating tools for evaluating CP within given health interventions to enable outcome driven health care and a desire for ensuring program efficiency (152).

The question of how power is shaped between health professionals and communities is central to most approaches. Some models explored CP as part of an organisational change process (107), whilst others sought to map factors which contribute to a lack of effective strategic partnerships between communities and health system stakeholders (153). Only three studies – Zakus & Lysack (3), Kelly and Vlanenderen (154) and Pickin et al. (153) – specifically sought to unpack the perspective of PHC workers in exploring CP. A detailed analysis of each model has been prepared as a separate publication but excluded here due to word limitations.

In order to be helpful, conceptual frameworks generally have to be adaptable to local structures, environments and needs (155). The real challenge and skill was in the application of each framework within operational realities and constraints. It is best to tailor and adapt a model to suit specific research questions, definitions of CP, and research perspectives (health professional vs. community participant).

I utilized elements of Health Canada's Public Involvement Continuum Model (107) in constructing the Participatory compass (section 4.3.3) since it provided a thick descriptive categorization of the types and modes of CP within a health system. The visual aid used within the model was highly appealing in deconstructing the participatory typologies.

35 | P a g e

Pickin and colleagues (2002) undertook empirical research, consisting of in-depth interviews with a purposive sample of people who had recently been involved in a community– statutory sector partnership, and then formulated a model to describe the factors which may contribute to a lack of effective strategic partnerships between communities and health system stakeholders (153). Whilst findings were useful in outlining the participatory dynamics of the particular intervention examined, authors recognised the limitations in generalizing such findings across the health system. Adapting the same logical process as Pickin (2002) (153), I utilized a more rigorous in-depth research approach in my study. This was done by engaging multiple types of health actors; exploring CP across multiple settings of the health system over a 3-year period; and using multiple research methods. The paper also provided many insights into constructing a factor analysis model of the various themes distilled through the qualitative research in relation to a health system. I adapted this factor analytic approach in the development of the final theoretical framework of my thesis, although there were distinct variations in the hermeneutical approach that will be described in Section 6.2.

More generally, I found the approach described by Zakus and Lysack (3) highly appealing due to its pragmatic health system planning perspective. Authors recognised that the implementation of CP is the ultimate responsibility of local health programme initiators, and it is at the local level where the day-to-day realities of incorporating CP into health service delivery are confronted. An aspiration of my study was to assess the participatory dynamic of Sri Lanka’s PHC units. Links with local political structures and legitimacy (or illegitimacy) of the community ‘representatives’ from the population which the health agencies serve were highlighted as a factor influencing attitudes towards CP. Finally, I utilized Rifkin’s widely used Spidergram model (387) to assess CP in the NCD prevention intervention case studies (Section 5.7).

36 | P a g e

Table 4. Summary of theoretical frameworks for exploring CP: Ladders, spiders, continuums, rubrics, networks and lists Type Author/Reference Brief description Ladder Arnstein's ladder of The analogy of a ‘ladder’ highlights the divergence of power between those who have power and those who do ‘citizen not. The ladder has eight rungs, each one representing a level of citizen involvement. At the top of the ladder, participation’ citizens had the most control over decision-making. The middle rungs represented degrees of tokenism (156) (placation, consultation and informing), and the bottom rungs describe levels of non-participation.

Spidergram Rifkin's Spidergram In Rifkin's approach, five factors are considered to influence the breadth or depth of CP in a community health model (151)(387) programme. The five factors are: needs assessment, leadership, organisation, resource mobilization, and management. Dimensions may be substituted with other dynamics of participation such as gender aspects of participation.

Rubric Cohen-Uphoff Cohen and Uphoff suggest that CP can best be analysed by asking four critical questions of: ‘Who participates? rubric model (157) Why do they participate? When do they participate? How and where do they participate?’. The answers to these questions are then constructed with what authors term the ‘Dimensions of participation’. These are: the kind of participation that is taking place; the sets of individuals in the participatory process; the various features of how engagement occurs; and the purpose of participation. The context to which participation takes place (defined as historical, environmental and socioeconomic) are also examined. Gaventa’s Power Gaventa’s framework places special emphasis on exploration of the ‘representativeness’ in CP as it relates to the Cube rubric model organisational structure in the community. This may actually exclude some persons from decision-making. (158) Gaventa’s approach comprises three power dimensions: the observance of conflicts in decision-making, exploring who participates, who gains and who loses in CP; the organisational structure of society; and the ‘social myths’ and norms that may shape, legitimize and determine patterns of participation in communities. Continuum Health Canada’s The framework was specifically aimed at capturing the range of pathways by which the public can be informed, ‘Public Involvement involved and engaged with healthcare organisations in Canada. The continuum framework defines five levels of Continuum’ (107) public involvement together with criteria appropriate for the selected level of involvement. The hallmark of the model lies in its articulation of the methods health system planners can utilize in determining a strategy for CP engagement. The community This model captures the dynamics of partnership between communities and an agency (or health system) over engagement time. The distinguishing feature between this model and that of Arnstein is that it accounts for the changing

37 | P a g e

continuum (159) nature of CP, which may evolve over time. The model is firmly enshrined through a technocratic lens where communities are engaged to ultimately effect health system goals and programmatic changes. Networks Pickin and This model describes the factors which may contribute to a lack of effective strategic partnerships between colleagues communities and health systems. The five components are: the community’s capacity to engage; the skills and ‘network’ model competencies of organisational staff; the dominant professional service culture; the organisational culture; and (153) the dynamics of the local and national political systems.

Lists Molyneux’s model Molyneux and colleagues constructed a model of ‘community accountability’ in health by listing factors that (68) influence the functioning and impact of community participatory processors within a given health system. The model encompasses local resource capacities, community perceptions, ‘who’ participates, the modality of participation, and how such determinants work synergistically or antagonistically to ultimately affect CP in a health service. Zakus and Lysack’s Zakus and Lysack (1998) constructed a list of factors and ‘predisposing conditions’ that need to be considered for list (3) enabling CP in health. Since the implementation of CP is the ultimate responsibility of local health programme initiators, it is at the local level where the day-to-day realities of incorporating CP into health service delivery are confronted. The model places emphasis on political context, health system factors and community factors such as the legitimacy of community representatives in influencing CP. Butterfoss’s Butterfoss defined a set of program indicators to guide planners in the process evaluating CP within a given indicator list (161) intervention. This included for instance the mapping of participant and organisational characteristics, their capacities, shared interests and histories. Kelly and Kelly and Vlanenderen (1995) noted that the evaluation of participatory processes in a program should consider: Vlanenderen’s list the modes of participation and the degree to which participation influenced power dynamics between the parties (154) involved. The authors highlight the importance of communities’ own perception in influencing how they behave in group-based projects.

38 | P a g e

2.2.2 Challenges in assessing CP and evidence of its effectiveness

Health reform processes in many developed nations within the OCED have promoted CP in routine health services as a means of increasing accountability and relevance through partnerships with communities (162). Building trust, accountability and legitimacy through CP in health is seen as a prerequisite for an inclusive society and for policy formulation (163) (164). A number of guides that promote public involvement in the planning and development of health services have also emerged over the last two decades (165).

Despite the rhetoric, a critique of the CP approach has been the lack of objective measures to verify whether or not consumer participation has improved the effectiveness, efficiency or other quality aspects of health care. Linked to this is the lack of consensus on which CP methods and strategies are most effective. In the tradition of evidence-based medicine, there have been efforts by a number of researchers to empirically assess the impact of CP in health. Six systematic reviews exploring the nexus of CP and health care are presented in Table 5.

Most reviews concluded that involving communities had contributed to some improvements in service provision, although the effects of CP interventions on quality of care and on health and social outcomes were not definitive. These results are not surprising given the complex nature of CP and difficulty in linking inputs and outcomes through causal linkages. There are likely to be numerous confounders which are difficult to control for without comparative groups, which are difficult to establish in real-time policy change. Causality is also difficult to determine in complex community-based interventions (73,166).

An important lesson in analysing the systematic reviews was to understand their theoretical basis, the proposed causal links or theory of change in the use of CP approaches, and the whether the outcomes considered either health status or health system responsiveness. Only one systematic review explicitly presented its theoretical underpinnings and offered a critique of the proposed model utilising the evidence derived from the review (68).

39 | P a g e

Table 5. Summary of systematic reviews exploring CP and health

Study Objective Findings Simpson Systematic review  Mental health services that involved current and previous (2002) involving users in the service users in providing training and feedback to (167) delivery or evaluation of professional staff and employees had more positive mental health services. attitudes toward users than those that did not harness user participation.

McCoy A systematic review for  The review found some evidence that community (2012) evidence on health facility involvement in health facility committees (HFCs) can be (168) committees in low and effective in terms of improving the quality and coverage of middle-income countries. health care, as well as impacting on health outcomes. However, the external validity of these studies is limited given the varying roles and functions of HFCs and the complex set of factors influencing their functioning.

Repper A systematic review to  There is tentative evidence that consumer involvement in (2007) describe methods of training enhances workers' skills in the manner prioritized (169) involving consumers in by consumers. However, if consumer involvement in training healthcare education. and education is to facilitate services that reflect the priorities of the people using them, it must be developed in partnership with service providers.

Evans A systematic review of  The review demonstrated that there is very little evidence (2010) health and social outcomes of participatory approaches by UK public health units (166) of participatory having any noteworthy impact on health and social approaches by UK public outcomes. health units.

Molyneux A systematic review of  The review explored the impact of community involvement (2012) (68) methods and modalities in decision making processors at peripheral health facilities used to ensure community in developing countries. Research focused specifically on accountability in PHC community representation on ‘health facility committees’. settings.  Relatively few studies presented good quality quantitative data using observable measures of impact. More often, authors drew on views and perceptions of the committees and community members, which typically differed by stakeholder group (i.e. between health workers and committee members, or committee members and community members). While this is expected given the different potential losses and gains by different parties, such differences may also relate to interview bias.

Crawford Systematic review to  A systematic review of involving patients in the planning (2002) examine the effects of and development of health care found few empirical (170) involving patients in the studies which have examined the effects of involving planning and development patients on quality of care, consumer satisfaction and other of health care. health service outcomes.

40 | P a g e

2.2.3 Evidence from developing nations

In a systematic review of literature, Molyneux and colleagues (2012) examined the main forms of community engagement at peripheral health facilities in low or middle-income countries (68). Of the 21 studies included in the final analysis, half were from sub-Saharan Africa, and other countries included India, Colombia, Mexico, Cuba, Peru and Nepal. The main form of community engagement was found to be through participation through committees and advisory groups at the facility. This corresponds to McLeroy’s ‘Community as setting’ (96), and Whitehead’s Type 6 CBI typology (97).

Key influences that impacted CP related to how committee members were selected, their motivation for involvement, their relationship with health workers and managers at the institution, and the provision of adequate resources. Authors also noted specific challenges to sustained engagement such as the community members’ lack of clarity concerning roles and responsibilities on the committee, avoiding politicization and dilemmas related to voluntary participation and remuneration.

In undertaking an independent examination of each paper included in Molyneux’s review, I noted that many of the efforts and interventions to inculcate CP were established by external agencies and not initiated by the government health authority. There was also limited information on sustainability and ‘uptake’ by health systems and the perspective of health workers. Ensuring program sustainability and integration across the health system beyond the successes of pilot projects requires sustained commitment, leadership and financing from the nation’s health authority (171-173).

A central theme of my research is the exploration of CP within ‘routine’ health system programs, devoid of any external agency support, pilot programs or other such vertically driven time-limited interventions. Molyneux’s (2012) concluding statement was that “very few studies presented good quality quantitative data”: this statement is equally made in relation to empirical qualitative data, an issue that drove my research and the contribution I seek to make.

41 | P a g e

2.2.4 CP in primary health care

“Primary health care is based on practical, scientifically sound and socially acceptable methods and technology made universally accessible through people’s full participation and at a cost that the community and country can afford” -WHO, Declaration of Alma Ata, 1978.

As highlighted in above quote, CP is considered a central ‘tenet’ of Primary Health Care (PHC), and one of the four pillars of the Alma Ata Declaration on PHC in 1978. In 2008, on the 30th anniversary of the Alma Ata Declaration, the Jakarta Declaration for Primary Health Care reaffirmed the need for CP. WHO cautioned member states not to ‘lose focus’ on prioritizing CP in health care reforms (57), and to ‘return’ to the original precepts of the 1978 declaration (174). In its call for action, the Jakarta Declaration emphasized the necessity of participation in terms of “actions carried out by and with people, not on or to people” (175).

Figure 4 depicts a theoretical construction of the PHC ‘intervention space’ based on WHO’s ‘community and home-based’ health care model (176). The diagram is a representation of the key components of preventive health action at the PHC level. Though CP is theoretically enshrined (as articulated in Alma Ata Declaration) within the PHC layer, its approach (and reach) is diffused through primary, secondary and tertiary preventive actions. Primary prevention comprises activities for health promotion and prevention of illness at the individual, family and community level. Secondary prevention focuses on screening, early detection, provision of treatment and care for common illnesses and ailments, and appropriate referral. Tertiary prevention comprises the provision of rehabilitative and palliative care for patients with chronic illness and disability.

In addition to the Alma Ata Declaration, CP has emerged as a core policy priority in a number of international health frameworks. The Ottawa Charter for Health Promotion in 1986 proposed ‘strengthening community action’ as one of its five action areas, where “concrete and effective community participation” was needed in setting priorities and implementing them to achieve better health (177). More recently, the Global action framework for the prevention and control of NCDs launched by WHO in 2012 (178) emphasized the importance of CP, and the need for governments to create ‘supportive policy environments, appropriate legislation and inter-sectoral action’ for health to stimulate such participation (24).

42 | P a g e

2.2.5 CP in NCD prevention and control

The importance of enabling partnerships with communities at the PHC level to address NCD risk factors have been iterated in numerous global health frameworks (24,179,180), WHA resolutions (181-187), and by scholars (173,188). A number of themes were distilled from the review of literature on addressing NCDs using participatory approaches.

Strategies to tackle major risk factors for NCDs such as tobacco use, obesity, and excessive salt consumption demand collaborative engagement with other sectors (189). The need for a multi-sectoral policy approach to the prevention and control of NCDs is also required since many social determinants of NCDs fall ‘outside’ the health sector (190).

Most global frameworks have been shaped by research evidence of the impact PHC workers have on the reduction and control of NCDs (25) (26) (27). Meaningful partnerships with communities and the health system may be enabled through the ‘conduit’ of PHC workers. A major obstacle in the fight against NCDs has therefore been the human resources for health (HRH) gaps at the PHC level in the health systems of many developing nations, and the skills, knowledge and capacities PHC workers need to address NCDs (18). With appropriate training and equipment, employing non-physician clinicians in PHC settings to address the HRH gaps and manage NCDs, undertake community based screening and promote adherence to treatment within evidence-based guidelines has been shown to function effectively, and in some cases even better than interventions by physicians (25) (27).

Since communities access several health care providers in both public and private sectors, having joint strategies to prevent, refer and manage NCDs at community level is deemed essential (189). Such networking may also improve continuity of care and clinical management of those with chronic illnesses, and support public-private partnerships in preventive programs to combat NCDs. PHC worker attitudes towards CP (18,28), and work place cultures (191) are key factors that may limit the realization of PHC goals (explored in detail in next Section 2.2.6) .

43 | P a g e

To summarize: many nations have recognised the need to respond to the growing burden of chronic diseases through the structural reform of health systems and by forging collaborations within the health and private sectors, and – importantly – with communities themselves (17,18).

44 | P a g e

Figure 4. Link between PHC and preventive health action

2.2.6 Importance of health worker attitudes in enabling CP

Since the ‘heart’ of CP involves the reshaping of power dynamics between authorities and the communities they serve, it is vital to understand the perspectives and attitudes of health professionals toward participatory approaches in exploring how CP emerges and is sustained in practice. In examining the literature, the perspectives of health care workers on the value and utilization of CP had not been systematically analysed. I synthesized available evidence by undertaking a dedicated review, presented in full in Annexure 1.

After applying a specific inclusion and exclusion criteria (outlined in Annexure 1) and removing duplicates, a total of 131 papers were identified from 1,798 articles retrieved from electronic databases and hand searches2. A total of 39 papers met screening criteria and were included in the final analysis.

Figure 5. Literature search strategy

2 Peer-reviewed publications between years 1980–2013 Page | 46

The main finding of the review was that while the majority (46%) of health workers actively supported the idea of CP in health care, there was mixed evidence for the commitment needed for its meaningful implementation within health programs and in routine practice. A significant number of studies (21%) identified dissatisfaction and non-commitment to CP on the part of health professionals.

Case-studies across a diverse range of health care settings examined in the review highlighted worker attitudes and workplace cultures as key factors that may limit or enable the realization of PHC goals (18,28) (170,192,193) (194-196). The findings suggest that health sector stewardship and staff attitudes are vital ingredients to enable health systems to effectively engage with communities in PHC settings (191,197). Longstanding and dominant collegial structures within PHC settings which do not value community engagement in health care decision-making may disenfranchise even the most motivated PHC workers (198,199). Limitations to CP occurred where there was reluctance or unwillingness to listen to community views due to workplace cultures or time and resource constraints. In some settings, health workers identified the need to enhancing their own technical capacities to explore more effective ways to facilitate and sustain CP.

In analysing the content of each selected paper, I distilled a number of factors that were important in enabling or inhibiting CP. These were then categorized in four broad thematic areas: a) the administrative dynamic of the health setting or institution; the attitudes of health workers; b) the technical competency of health workers in relation to skills pertaining to community mobilization and community development; c) the resource environment such as time and other monetary resource; and, d) finally a policy and regulatory environment. The results factors are summarised in Table 6. Understanding such determinants may be useful if meaningful partnerships between statutory organisations and communities are to be implemented and sustained. I also drew upon these thematic areas identified through this review to help formulate an overall theoretical framework for my study (see Section 3.1).

Page | 47

Table 6. Factors influencing CP

Author/ Findings Factors (inhibitory and enabling) Reference Bowl (1996)  Confusion about the meaning and purpose of CP in health  Administrative (CP approaches not (200)  Lack of consensus on how health consumers/service users can best be represented on stipulated by management, part of committees job duty; confusion over role in CP)  Unwillingness to listen to community views  Attitudinal (low value placed on CP,  Limited resources to enable participation by management not wanting engagement from non-  CP does not directly relate to their spheres of responsibility and thus forms a push factor health stakeholders) against its implementation  Technical competency (methods for effective CP practice unknown)  Resources (time and resource constraints not allowing for CP practices) Freyens (1993)  Health workers tending to underestimate people's potential for action and reaction, and to  Attitudinal (little value placed on (201) insist upon the need for a hierarchical structure CP)  Confusion about the notion of community, multiple definitions provided for understanding of  Technical competency (methods for community participation effective CP practice unknown)  Resource limitations where even basic services are unable to be provided  Resources (time and resource  Lack of financial resources constraints not allowing for CP  Non-conducive management practices practices)  Ignorance about the value of participation  Administrative (CP approaches not  Taboos, customs, and traditions producing resistance to change stipulated by management, part of job duty; confusion over role in CP)  Confusion by community about their roles in CP  Lack of education by community on their roles in CP Hogg (2008)  An increasingly medicalised view of health promotion pushes out CP in routine practice  Attitudinal (low value placed on CP (202)  The perceived lack of leadership and interest from some communities prevents CP from being in medicalised approach to service pursued delivery)  Attitudinal (perception that community does not want CP) McCann  Policies that promote participation have little impact to promote culture of CP  Policy implementation (policy (2008) (203)  Younger staff are more willing to engage in new methods than older ones rhetoric not translating to practice) Bryant (2008)  Belief in the inadequacy of consumers' skills,  Attitudinal (perception that (204)  Perception consumers' lack of interest community does not possess capacity  Little value attached to consumer involvement in health decision-making. to engage)  Attitudinal (little value placed on CP in medicalised approach to service delivery) Page | 48

Kidd (2007)  A systemic change to the way consumer participation is funded and incorporated into mental  Administrative (205) health services is required  Resources  Need for the establishment of clear purposes, policies and roles, and commensurate funding to enable consumers to be stakeholders with similar resources to other stakeholder groups Happell (2002)  Ambivalence regarding the value and necessity of CP  Attitudinal (value placed on CP (206) limited)

Brown (2001)  Dominance of medical practitioners in establishing organisational values during public  Attitudinal (little value placed on (207) involvement activities CP in medicalised approach to  Local political tensions in primary care influences CP service delivery) Brandstetter  The importance of having community members’ involvement is important but should only be  Attitudinal (value placed on CP (2012) (208) included in the evaluation process limited to only evaluation of programs) Nathan (2013)  The need for health services to employ a facilitator who can support, engage, navigate and  Resources (209) advocate for the community representative’s participation and influence in health service policy  Technical competency and practice. In the absence of a CP facilitator, it becomes difficult to build skills and confidence of health care workers (HCWs) for CP and engage them in agendas for action Haigh (2008)  Lack of insight into appropriate methods of engaging communities  Technical competency (methods for (210)  Health professionals prefer consultative rather than decision-making role for consumes. effective CP practice unknown)  Variable patient interest and health professional attitudes  Attitudinal

Fridinger F.  Limited time available due to busy schedule for communication between health care  Resources (time constraints not (1992) (211) professionals to engage in CP allowing for CP practices)

Rifkin (1983)  HCWs in Asia have a ‘medicalised’ view of health and want responsibility to remain in the  Attitudinal (little value placed on (212) hands of the medical professional. CP poor in medicalised approach to service delivery) Simpson  The primary obstacle is funding CP strategies and lack of political commitment to  Administrative and policy (lack of (1981) (213) decentralization decentralized approach of health governance non-conducive for CP)

Page | 49

2.2.7 Placing health workers at the center of health services research

International debates on improving health system performance and quality of care are increasingly shaped by systems thinking and evidence-based approaches. Health policy and systems research (HPSR), promoted by WHO, seeks to understand how health systems can be revitalized and better organised (214).

However, a recent review indicated that when health systems research is considered only through bio-medical research processes, it is systematically under-valued by health care decision makers and administrators (215). As Koon and colleges (2013) highlight, the more successful approaches in HPSR are typically interpretive and engage stakeholders at the interface of community and health services (54).

Most interventions at primary care level are undertaken or mediated through PHC health workers, often considered the ‘foundation stones’ of a health system (55). Their views and perspectives are vital for any exploration of CP in PHC. Indeed, health systems research that puts ‘health workers at the center of the research agenda’ is increasingly acknowledged by WHO as an important yet ‘missing branch’ of systems research (55).

Rather than viewing PHC workers merely as the ‘conduits’ for health service delivery, the WHO Alliance for HPSR suggest that health workers be engaged as protagonists for new ideas and innovations (216). Gross and colleges (2012) highlighted the importance of analytical approaches aimed at understanding health worker capacities to access resources (for instance human, financial and symbolic capital), and transfer them to the community (217).

In the present research study, the phenomenon of CP was explored through the perspectives of PHC workers, health administrators and policy makers and by undertaking detailed observations of their routine practice environments. By exploring the perspectives of health workers, Gross and colleagues (2012) argued that deeper understanding of public health system resources, capacities, practices and performance could be identified, providing an evidence base for health system improvement (217).

Page | 50

Literature review (Part 3) Community participation in Sri Lanka’s health system

Sri Lanka country context

This section presents an overview of Sri Lanka’s health system, focusing on the challenge of tackling NCD’s, the evolution of PHC services and more recent push for reforms.

2.3.1 Sri Lanka’s early commitment to primary health care

Sri Lanka was hailed as a ‘success story’ in PHC, with its early adoption of the Alma Ata principles and achievement of health outcomes comparable to those of high-income nations (31). A number of factors have been cited as contributing to Sri Lanka’s success in improving the health status of its citizens: public financing of health care since gaining independence in 1948; early adoption of PHC in 1978; investments in a preventive health model via a network of trained PHC workers at village level; and high levels of women’s autonomy and relative gender equality (32). Whilst it is difficult to qualify the causal links between these factors on health outcomes, these trends have clearly contributed to impressive health outcomes associated with maternal and child health, low levels of communicable diseases and longer life-expectancy (32).

2.3.2 Decentralisation and devolution of health decision-making

As discussed in Section 2.1.10, decentralization is central to CP and to health system reform(135). It entails the increased involvement of local jurisdictions and civil society in the management of their own affairs, with new forms of participation, consultation, and partnerships.

Decentralization of Sri Lanka’s health system may be traced to 1954, which saw the appointment of 15 Superintendents of Health Services (218). Although the central health authority at the Ministry of Health still had final jurisdiction over human and financial resource allocation, this model of health governance allowed for a degree of autonomy in health program decision making within each Superintendent area (219). The demarcation of

Page | 51

each Superintendent of Health Service area was however largely based on geo-political determinants.

Another significant event in the history of the health services in Sri Lanka was the signing of the Charter for Health Development in 1980. In doing so, the Government committed itself to strengthening PHC services, with the importance of community health strategies clearly articulated. A strategy for inter-sectoral action in health was also launched in the same year with the establishment of the National Health Development Network3. The Network comprised the National Health Council (NHC), as the peak health policymaking body, and the National Health Development Committee (NHDC), to implement the decisions of the Council. The NHC took the responsibility in setting out health policies and coordinating multi- sectoral action in health. The NHC also enabled a dedicated mechanism to coordinate inter- sectoral health action at sub-national (district) levels.

In 1987, major constitutional reform in Sri Lanka led to the 13th Amendment of the Constitution, which saw the creation of provincial levels of governance. The system of Superintendents of Health Services were abolished, and the health administration was devolved to a provincial health system. Provincial councils in turn encompassed multiple district demarcations. The introduction of a provincial council system restructured the entire provincial, district and divisional levels of administration. These demarcations were based on geopolitical boundaries, which remain to the present day. With the devolution of powers, the provincial Ministries of Health were given the responsibility for the management of all hospitals up to district level, and the responsibility for implementing national level strategies. The central level MoH were given responsibility for policy formulation, providing strategic guidance, technical training institutions, bulk purchases of medical supplies, management of teaching hospitals and specialized campaigns.

A key milestone for Sri Lanka's health system post-independence was the launch of the National Health Policy in late 1992. The policy focused on health inequalities of low socioeconomic groups in society, and emphasized the importance of “community involvement

3Members of the NHC included Ministers of; Health, Agricultural Development and Research, Higher Education, Education, Finance and Planning, Local Government, Housing and Construction, Home Affairs, Labour, & Rural Development. Page | 52

in the planning, implementation and evaluation for all health and health related activities” (220).

Guided by the NHC, a Presidential Task Force was appointed in 1997 to re-formulate the National Health Policy of Sri Lanka (221). This was done with the financial and technical support of the World Bank in partnership with health system experts (222). The major recommendation was for the introduction of a comprehensive Health sector Master Plan to deal with the range of human resource and equity issues present in the system. The development of the master plan was supported by the Japanese Government, and was launched in 2002. It aimed at delivering comprehensive health services to Sri Lankan citizens and highlighted the “need to empower communities towards more active participation in maintaining their health” (223).

In 1999, the MoH was restructured under the direction of the Health Minister. At the central level, the responsibility for policy development, planning, and monitoring of the public health services rested with the Director General of Health Services. The executive arm of the MoH, comprising the Health Minister and Health Secretary, approved the initiation of all health policies and facilitated the process of policy endorsement through government (224).

Note on Post-conflict and Estate health system

Sri Lanka has recently emerged from nearly three decades of protracted conflict, which came to an end in 2009. A recent review suggested there are clear deficiencies and unmet needs in the health systems of post-conflict regions such as increased mortality and morbidity, lack of health workers, low access to services, low levels of knowledge in health issues, low levels of health promotion and awareness programmes4. Greater investments in health for the North including establishment of health worker training institutes translated to enhancement in service coverage and early indicators for positive reduction on health care burden.

The creation of a formal estate sector commenced during the colonial period by the influx of South India migrant workers bought into Sri Lanka by the British to work in tea plantations.

4 Siriwardhana, C., & Wickramage, K. (2014). Conflict, forced displacement and health in Sri Lanka: a review of the research landscape. Conflict and health, 8(1), 1-9. Page | 53

The tea sector constitutes an important export market for Sri Lanka. The sector remains state owned but leased to management companies. The health provision is shared to varying degrees between services run by the MOH and the plantation management. Some areas of the estate sector consistently reported the highest maternal and infant mortality rates in the country, well exceeding those in conflict affected areas. Difficult terrain and the long distances that mothers had to travel to government institutions that provide emergency obstetric care contributed to some of these deaths. In recent years intensive efforts have been made to improve converge of the basic package of MCH services to mothers and children across the estate sector. Trained midwives, family welfare supervisors, Assistant Medical Practitioners and Estate Medical Assistants provided the services. Women were provided with transport facilities and paid leave to attend antenatal clinics. A dedicated directorate of estate health has also been established by the MOH with greater investments made for providing outreach services.

2.3.4 Brief overview of Sri Lanka’s health system

Sri Lanka’s PHC system encompasses both preventative and promoting health services that are coordinated and delivered at the village level through a network of Medical Officer of Health (MOH) units. The basic organisational structure of the PHC system in Sri Lanka is depicted in Figure 6.

The Medical Office of Health (MOH) unit forms the most fundamental PHC unit of Sri Lanka’s health system (225). This primary interface between the community and the health system is therefore critical for the exploration of CP. There are 286 MOH offices in Sri Lanka, with the average catchment population ranging from 60,000 to 80,000 people (224). MOH’s are usually organised according to government administrative divisions within a district, and increase in number depending on the size and population of each district. The curative care services at primary care level are offered through a network of district and rural hospitals. A medical officer heads each MOH unit and is a physician by training. The other health professional staff attached to the MOH include Public Health Inspectors (PHI), Public Health Midwifes (PHM), Public Health nursing sisters (PHNS), Senior Public Health Inspectors (SPHI), Senior Public Health Midwifes (SPHM) and health volunteers. PHIs are responsible for environmental health action (vector-borne diseases, water and sanitation and food inspection control), and on average, each covers a population of 9,000. PHMs are primarily Page | 54

responsible for maternal, reproductive and child health services and are allotted a population of 3,000 persons. These PHC workers are also responsible for implementing vertically driven health interventions developed at the central level, such as environmental health campaigns and disease-specific prevention initiatives.

In summary, this section has described the process of decentralization and the degree of devolution in health decision-making in Sri Lanka’s health system. Indeed, the organisational apparatus appears conducive to enabling CP approaches. The next section explores the evidence of community participatory practice. Figure 6. Health administrative system of Sri Lanka

Page | 55

2.3.4 Review of interventions utilizing community participatory approaches in Sri Lanka’s health system

Stimulated in part by the historical declaration of ‘Health for All’ in 1978, the health authorities in Sri Lanka have consistently demonstrated policy commitment to community participatory approaches in health services. Indeed, Sri Lanka’s community health worker model was showcased as an example of good practice at the inaugural conference in Alma Ata (226,227). Despite these commitments, there has been no review of the research evidence exploring the actual manifestation of CP within the health system in Sri Lanka. I undertook a comprehensive review of both published and grey literature to identify any interventions or initiatives, their characteristics and contribution to the realization of health goals.

Figure 7. Literature search strategy

Results from 9 electronic databases, including hand searches from Ministry of Health, Post-graduate Institute of Medicine and University databases (n= 785)

Application of eligibility criteria: 73 titles/abstracts met search criteria after removing duplicates

Application of inclusion/exclusion criteria: 20 papers met criteria

Application of Critical appraisal criteria: 10 papers met criteria and included for final analysis

The review process outlined in Figure 7 found only 11 studies that described the involvement of communities in the planning, implementation or development of health care programs in Sri Lanka. The inclusion and exclusion criteria are presented separately in Annexure 2.

Page | 56

No study was found to specifically explore the dynamics of CP or how it was defined, nor to describe the perspectives of health workers or community members on its utility. Descriptions of CP were secondary or incidental to the core purpose of the research being conducted, or were case-study presentations. All articles described community health interventions that were undertaken as part of a pilot project and were in all cases led by an external agency or were donor-funded intervention research projects. None described interventions that were catalysed by workers within the health system.

Of all studies reviewed, the meaning of ‘community’ was explicitly defined in only two studies. Pearson and colleagues (2010) articulated the idea of an ‘epistemic’ community – “a network of professionals with recognised expertise and competence in a particular domain and an authoritative claim to policy-relevant knowledge in that domain or issue area” (228). The development and consideration of policy options within this epistemic community enabled a deeper understanding of evidence, knowledge and linkages for health policymaking in suicide prevention due to ingestion of pesticide. Wickramage (2008) defined community as “those displaced persons within the confines of a IDP camp” (49).

By reading descriptions of the actual points of community engagement in the intervention against the steps of the classic project cycle, it became evident that CP was manifested mainly during program implementation and design phases (Figure 8). This result is not surprising given the highly technocratic approach adopted by health planners in utilizing CP as a means of boosting the ‘reach’ of health interventions in communities.

Four studies described effects of enhanced service delivery, better access and acceptability of services as a result of partnerships between communities and health workers. For instance, an early-warning disease surveillance and outbreak response system established in partnership with communities not only improved early detection and health promotion capacity, but also built social cohesion amongst conflicting groups within the displaced community (49). Other studies suggested that involving communities may have led to enhanced service coverage (229), enhancing health seeking behaviours and improvements in health knowledge of communities (230,231). Further research is needed to understand the processes through which participation facilitates positive health outcomes and empowers communities. Page | 57

In summary, despite Sri Lanka’s longstanding health policy commitment to CP, the published literature revealed a scarcity of interventions and research on CP in health.

Figure 8. Points of participatory engagement within intervention cycle. CP ‘manifests’ mainly during program implementation (Step 4) and program design (Step 3) phases.

Project cycle phase Paper Step Step Step Step Step reference 1 2 3 4 5 1 Jayawardene x , 2011(232) 2 Ranasinghe, x 2011(231) 3 Konradsen, x x x x x 2000(233) 4 Yasuoka, x 2006 (230) 5 Wickramage x x x x ,2008 (234) 6 Devendra, x x x 1984(235) 7 Holmes, x x 2011(236) 8 Gammanpila x x x ,1997(237) 9 Jayasekera, x x 2000(238) 10 Perera, x x 1985(239) 11 Pearson, x x x 2010(228)

2.3.5 Current public health challenges in Sri Lanka

Sri Lanka has an impressive track record of health care provision, with substantial accomplishments in health outcomes compared to countries with similar development indices. These include low maternal and infant mortality (240), communicable disease burden (241), near elimination of malaria (242) and improved life-expectancy at 78 years for females and 71 for males (31). Sri Lanka's tax-financed public health system has played a significant role in achieving high levels of health status for its people, despite government expenditure on health being one of the lowest in the world (at 1.2 to 3% of GDP) (229).

In the aftermath of the 28-year protracted civil conflict, the Sri Lankan government has embarked on a rapid economic development pathway (243). Studies have shown that

Page | 58

improving economic status with increased rate of urbanisation brings both positive and negative effects on population health (244,245). Sedentary lifestyles, changing dietary habits, consumption of refined and processed foods, living in polluted urban settings and road traffic accidents through expanding road networks may increase risk exposure to NCDs (244).

Sri Lanka is also experiencing an advanced stage of a demographic transition and is currently the fastest ageing South Asian population (39,246). In 2010 it became the ‘third oldest country’ in Asia after Japan and Singapore (247). Accordingly, the chronic disease burden and heightened fiscal cost for long-term care of this ageing population are emerging challenges (243).

Sri Lanka’s epidemiologic transition has seen a shift in the disease pattern from infectious diseases towards chronic NCDs, which now account for nearly 90% of the disease burden (248). Whilst mortality due to infectious diseases have been declining, from 42% of deaths in 1945 to 20% in 2003, mortality attributable to NCDs are rising. During the past half- century the proportion of deaths due to circulatory disease across all ages (such as heart disease and stroke) has increased from 3 to 24% (33). The largest increase in the rate of hospitalization is due to hypertensive disease (249). Obesity levels show a clear upward trend (250), with the prevalence of Sri Lankan adults being overweight, obese and having central obesity at 25.2%, 9.2% and 26.2% respectively in 2005–2006 (251).

The NCD burden also has a complex pattern of distribution in Sri Lanka and varies for each specific condition. For instance, deaths from heart disease are higher among the rich while those from asthma are highest among the poor (33). The age-adjusted prevalence of diabetes showed marked variation, with the highest prevalence (18.6%) in the urbanised communities of Western Province, and the lowest prevalence in the highly rural Uva province (6.8%) (252).

An increase in sedentary lifestyles (36), changing dietary patterns with nutritional deficiencies (37), alcohol and tobacco consumption in males (458), poor community based management of those already affected by chronic diseases (33), and gaps in the capacity of district health systems to deliver effective NCD prevention and control programs (40),

Page | 59

have led to the increasing prevalence of NCDs. This has therefore emerged as one of the most important public-health challenges in Sri Lanka (38).

2.3.6 Chronic disease care at PHC level

Within Sri Lanka’s socialist health care policy framework, all government health services are available free of charge to all citizens, including all inpatient, outpatient, and community health services. While most Sri Lankans live within 3 kilometres of a government health facility (253), disparities exist in the functional capacity of health care facilities in the post- conflict, rural and estate sectors (254).

In Sri Lanka’s health system, health promotion and prevention efforts are driven by the District’s PHC system, via the MOH unit (see Figure 6). Treatment for chronic diseases is largely delivered and managed at specialist clinics in higher-level (secondary or tertiary level) facilities. Outpatient clinics have no adequate or standardised systems to ensure long- term clinical management and follow-up for NCD patients (33,255). Most privately run outpatient services at village level are provided by government medical officers working in their off-duty hours near their government institution posting (253).

Studies in Sri Lanka have shown that the cost of treating NCDs is high (33) and that a large proportion of households affected by NCDs (up to 31%) have to borrow money when seeking treatment to cover out-of-pocket expenditures (256). The WHO considers treatment and secondary prevention of NCDs at primary care to be a vital and cost-effective strategy that “should be promoted over the provision of such services from larger hospitals” (257). The most basic treatment facilities available at village level are the ‘divisional level’ (or village level) clinics and rural hospitals that have the potential to provide treatment for those with chronic diseases such as diabetes and cardiovascular disease. However these facilities lack even the most essential drugs such as lipid lowering statins, basic laboratory facilities, and NCD assessment and management guidelines (33). As a result, these institutions remain underutilized in addressing NCDs.

Page | 60

2.3.7 Community access to PHC services and phenomena of ‘bypassing’

Access to all health services in Sri Lanka is made flexible by pluralist policies of permitting patients to visit any hospital in the country without restriction. Community members have a large degree of ‘choice’ in Sri Lanka’s health system with regards to selection of their health care provider. The Ministry of Health has not established a universal patient referral system in Sri Lanka. People can choose their own pathways to any primary, secondary or tertiary facilities, in both public and private health sectors. For instance, a person may access the outpatient department at a major tertiary care institution in his/her district by ‘bypassing’ the PHC clinic facilities in his/her village (Figure 9).

In the absence of policy guidelines on health system referrals, health consumers readily bypass primary care institutions to access larger OPDs at secondary and tertiary institutions, even for the most minor ailments (253). This has led to an over-utilisation of such units at secondary and tertiary care services. There is no established system for integrated care plans for patients with chronic disease. Primary care providers have little guidance or training in family medicine practice and the private practitioner network is largely unregulated.

There are no formal partnership programs between the MoH and private sector providers for joint NCD prevention and health promotion programs, nor are there any guidelines for reporting requirements vital for epidemiological surveillance and resource planning. If communities view secondary and tertiary care institutions as their primary foci for chronic disease management, then partnerships between preventive programs and these institutions are vital for NCD prevention and control.

Page | 61

Figure 9. Pathways of community ‘choice’ in accessing health services. A person with chronic disease seeking health care have multiple choice in pathways to seek health care. As indicated, the person may choose to bypass primary level care to enter an out-patient clinic at a large tertiary care hospital for their consultation, or alternatively enter an Ayurvedic clinic at primary level. The MOH office is also indicated as the dedicated conduit within the public health system that provides preventive health programs. Page | 62

2.3.8 Current efforts to reform Sri Lanka’s PHC model and address NCDs

Successful prevention of NCDs not only calls for inter-sectoral action to reduce the negative social determinants and enforce positive ones that affect chronic diseases, but for reorienting PHC services to enable them to better address NCD prevention and control functions (51).

Sri Lanka’s MoH has recently intensified efforts to address the growing burden of NCDs through the formulation of a National Policy for the Prevention and Control of NCDs in 2009. This policy articulates CP at PHC level as a priority action area for the fight against chronic diseases. A key policy directive has been the appointment of a new cadre of ‘Medical Officers for NCD’ (MO-NCD) to be based at each district health system to help catalyse and direct NCD programs at the district health system level. However, as of 2014, the work functions and duty lists for MO-NCDs posts remain undefined, with no direct budgetary allocations to support interventions at district level. A 2012 World Bank review of the NCD burden and health sector response revealed that district health systems in Sri Lanka showed limited capacity to promote effective population-level outreach and prevention activities for NCDs, with their current focus remaining firmly on maternal and child health and infectious diseases (33).

A major health system reform effort undertaken by the MoH that has been spurred on by the urgency to address the growing burden of NCDs has been to revitalize Sri Lanka’s PHC service delivery model (258,259). Driven by the MoH‘s Directorate of Policy Analysis and Development (DPAD), the model emphasizes revitalizing PHC services through better integration of preventive services (at MOH level) with curative care services such as those offered at district level hospitals. The reform process seeks to enhance community-based health care by improving health promotion, continuity of care, better penetration of screening and prevention programs, and ensuring effective clinical management and follow- up practices. A schematic diagram depicting the enhanced scope of PHC service coverage and enhanced CP within the new model is shown in Figure 10.

Page | 63

Figure 10. The proposed model for PHC reform aimed at expanding health service coverage to communities

Proposed reform strategies emphasize increased engagement with communities at PHC level and a ‘referral and back referral system’ to evolve through meaningful partnerships with communities and PHC workers. The draft guidelines developed by the DPAD for revitalizing the role of PHC workers emphasise a shift towards a “community and family centred approach” (258). If such CP reforms are to diffuse through the health system and be translated into reality, then it is essential to understand the perspectives and attitudes of health professionals, who are the key catalysts to engagement. This thesis addresses, in part, capturing such evidence.

2.3.9 Section Summary

The growing burden of chronic diseases in Sri Lanka, with concomitant increases in the costs of treatment and the challenges of a rapidly aging population, has required health care authorities and policy makers to rethink PHC service delivery and prioritize strategies that work towards better community based prevention programs (259). Despite this, the rhetoric promoted through policy statements and related materials are not matched with effective implementation, training, resources and other mechanisms to facilitate the uptake and adoption of these emerging approaches.

Page | 64

Chapter 3: Theoretical framework for exploring research objectives

Image 3. A health promotion program targeting children and young adolescents on diarrheal diseases at coastal community in Eastern Province of Sri Lanka.

Page | 65

3.1 Developing a complex systems model to explore research objectives This section presents the development of a conceptual framework to ‘situate’ the research questions of my thesis within a health systems approach.

Exploring CP from the perspective of health professionals within Sri Lanka’s PHC system forms the basis of my thesis. A conceptual model adopting a socio-ecological perspective within a complex health systems approach was constructed as the guiding ‘compass’ through which to explore these research questions and objectives.

3.2 Health systems frameworks

WHO defines a health system as consisting of organisations, people and actions whose primary goal is to promote, restore or maintain health (260). Although the structure of health systems varies substantially from country to country, health systems frameworks are useful tools for policy makers and researches in analysing the various components and determinants which affect the provision of population health and services within their system (261). Van Olmen and colleagues provide an in-depth overview of different frameworks for health systems analysis (262,263).

Figure 11. The WHO Health Systems Framework

The WHO Health Systems Framework model depicted in (Figure 11) comprises of six building blocks, which can also be considered as ‘policy levers’ to affect system change and revitalization (264). Whist I uncovered variations of this framework in reviewing the published literature (265-267), its core elements remain much the same. The first lever is

Page | 66

fundamental to health system management and involves ‘health sector leadership and governance’. This includes stewardship, administrative function, policy development, regulatory action and ensuring accountability. The second lever involves human resources for health management. The third covers health care financing, while the fourth involves the use of technology, medical products and diffusion of such technological innovations. The fifth relates to the evidence-based nature of practice and its value in such areas as health worker training and curricula design. Modification of one or all of these levers enables policy makers to reshape health system deliverables and goals.

3.3 The need for a health system model that captures complexity of CP

Whilst the WHO Health Systems Framework model is useful for macro-level analysis and creating a ‘common language for shared understanding’ (268), it is limited in capturing the complexity that operates within any given health system (265). Scholars suggest that researchers “need to expand the framework analysis” in order to explore specific phenomena (265).

Exploring complex phenomena like CP at the PHC level requires a holistic health systems viewpoint, as argued by Pickin et al. (2002), who reiterate the system-wide approaches needed (153). The authors also drew similar conclusions to those articulated in my literature review, calling for greater understanding and better approaches to supporting health authorities and statutory health institutions in working more effectively with communities.

3.4 Ecological perspective in health

The ecological perspective in health involves capturing the dynamic interplay and influence of biological determinants, material, social, psychological processes, and cultural and behavioural processes (269-271). The strength of the ecological approach is that it captures complexity - a key feature of modern conceptions of health and health systems (272,273). To address the growing burden of obesity, for instance, ecological approaches have been adapted to harness deeper understandings of the interactions between communities and NCD prevention and control programs (274,275). The dynamic interplay between disease burden, communities and health systems underpins the ecological perspective, and highlights the challenge policy makers face in effecting meaningful health system reform.

Page | 67

The value of adopting an ecological perspective in assessing the delivery of community health programs by statutory health institutions has also been acknowledged and promoted by WHO and health system scholars (269,273). Despite the rhetoric, however, the practical application of such an approach by health decision makers has been somewhat limited (190,272).

In undertaking health systems analysis, it is also not enough to merely list the various components or describe the environments within the ecology of a health system, but rather to explore how each of these components relate to, and are influenced by, one another.

A key component of my research (Research Objective 2) involves capturing the perceptions of decision makers and PHC workers involved in health care delivery and exploring how these relate or translate to community participatory practices and policy formulation. Authors exploring diffusion theory (276), organisational theory (277), and individual behaviour change theory (278), also suggest that implementation of health promotion interventions at community level may be influenced by a constellation of factors such as worker motivation, characteristics of the service organisation, and environmental context.

Health system issues are thus best understood by using multilevel complex approaches (279). By assessing existing policy frameworks and organisational histories and structures, complemented by direct observations of physical characteristics in space and practice, I attempted to capture such complexity in exploring the phenomenon of CP within Sri Lanka’s health system.

3.5 Adopting a complex health systems approach to exploring CP

The first view of complexity, in effect, means ‘complicated’! Complexity is a property of a system, not an intervention (280). Health systems have been viewed through this lens of complexity since there is great diversity of organisational forms, delivery modes and interactions between communities and among organisational units (281,282). A complex system is one that is adaptive to changes in its local environment, and is “built up from a number of components, which may act both independently and inter-dependently” (283). A key challenge of this thesis (as I articulate in Research Objective 4) is to identify what the ‘active ingredients’ are in Sri Lanka’s health system, notably those affecting community participation Page | 68

at the primary health care interface. The research findings (drawn from Research Objectives 1 to 3) have been used in the formulation of a ‘theoretical compass’ to assist health authorities to navigate the PHC intervention space and understand some of the key enablers and barriers to meaningful engagement with communities within the health system.

In relation to the ‘classic’ WHO system framework where the component parts of a system are characterised, the complexity perspective (also labelled as ‘complex health systems’ approach (280) or ‘multi-scale’ health systems analysis (284)) aims at unpacking the influences and relationships among the elements within a system. As Begun and colleagues (2003) suggest, complex systems operate through relationships among health system stakeholders, involving an analysis of the quality, emergence, and outcomes of relationships among individuals, groups of individuals, and organisations (281).

Scholars have argued that, too often, health planners and some researchers adopt a reductive perspective that focuses on structural components of a health system such as health policies, human resources for health and technical capacities of workers (214,285). The equally important subjective components influencing practices such as the perceptions and attitudes of professionals working within the system is ignored.

Although challenging, adopting a complex systems approach may yield practical insights for health care revitalization, even within individual health care settings. Campbell and colleagues (286) applied a complex systems approach to evaluate the routine operations of a stroke unit. After mapping the core health system elements as outlined in the WHO health systems model, they used multiple research methods to investigate staff attitudes, power dynamics, management protocols and resource availability at the unit. They were able to link variations within system components to service outcomes of the setting. Atun and colleagues (2006) also explored the structural, organizational, financial, clinical and relational changes of PHC reforms in Estonia using a multi-scale health systems approaches (287). Marchal and colleagues (288) took user fee exemption policies for maternal care in West Africa and Morocco to explore the methodological consequences of complexity for health policy research and evaluation. Authors suggested research approaches need to be dynamic so as to provide a “holistic and systemic view on the problem and/or solution”, and where they can be adapted in response to the insights that emerge (288).

Page | 69

3.6 A ‘progenitor’ model to explore research objectives

In order to anchor the research questions within a broader health systems framework and do so within an ecological approach, a basic model was constructed (Figure 12). Encompassing the core dimensions of WHO’s health system model, this conceptualisation recognises the importance of a range of factors – context, stewardship and PHC work practices – in ultimately influencing the emergence of community participatory practice at PHC level.

The area within the black circle (Figure 12) represents the health system. Within this are contained the six health system building blocks identified in the WHO model (numbered). There are two broad ‘environments’ that influence and affect community participatory practice at primary health care level. The first is defined as the ‘policy and structural environment’ encompassing health sector stewardship functions, statutory health policy and legal frameworks, and the general organization of health services. Second is the ‘practice environment’ that includes the actual delivery of services and encompasses determinants that influence service delivery such as worker attitudes and technical capacities (285). The PHC ‘intervention space’ is depicted by the solid sphere at the centre of the diagram. The Individual factors within these environments may work independently or synergistically to ultimately influence the PHC intervention space – the inner sphere at the heart of the health system. These potential inter-relationships are symbolized by the flow of arrows depicted within the model.

The ‘external environment’ encompasses factors outside the health system such as political influences, socio-cultural transitions and economic dynamics that may interact with elements of the health system, both directly and indirectly, to ultimately influence health interventions. For instance, a push for greater decentralization and devolution of power to district level governance (currently debated in Sri Lanka) may provide more autonomy and redirect power to district and village level health units.

Page | 70

Figure 12. A conceptual framework to explore the research objectives (the ‘Progenitor’ model)

This initial or ‘progenitor’ model is presented not as a dedicated analytical framework, but as a means of framing the research. I used this conceptual framework to symbolically link each research objective into the broad health system domains it sought to explore. I also used graphical icons based on the model to frame each research objective (see next section), and use these icons throughout my thesis to help refer the reader to the relevant health system environment.

Importantly, the model provided the basic theoretical ‘scaffolding’ on which a more robust model was constructed to explore CP at the PHC level and to ground the research data collection and analysis. By applying principles of grounded theory (through hermeneutic analysis as elaborated in the Methods chapter), I aimed to build upon this progenitor form, and ‘unpack’ the health system factors and determinants influencing the meaningful realization of CP practice at the PHC level.

Page | 71

3.7 Research Objectives

As described in Chapter 1, the primary research objective was encapsulated as a simple rhetorical question: How participatory are we?, in which the ‘we’ refers to Sri Lanka’s health system, and ‘participatory’ refers to CP in health care decision-making and service delivery at the PHC level. The study sought to: explore the experiences, values and commitment of PHC workers, health administrators and policy makers relating to the use of CP in health and for addressing NCDs; examine the extent to which CP approaches are utilized (if at all) within routine PHC practice; and explore factors that enable, limit, enhance or sustain meaningful CP in Sri Lanka’s health system.

A logical approach underlies the construction and presentation of the research objectives. The research aim was ‘deconstructed’ to form four research objectives, and further disaggregated into multiple sub-objectives to allow for a deeper analysis (elaborated in text-boxes below). The graphical icons adjacent to each research objective correspond to the conceptual domain of the health system framework to analyse CP as outlined in Figure 12, Section 3.7.

As explored in the review of literature (Sections 2.1.6 to 2.1.8), building an understanding of the enabling structural environment for CP by exploring existing policy and procedural frameworks is vital to exploring how CP is manifested within the PHC intervention space.

The first objective aimed at analysing the extent to which community participatory approaches are enshrined within existing national health policies, domestic legal frameworks, health sector regulatory frameworks, routine PHC worker duty lists and reporting and evaluation mechanisms. The construction of meaning of ‘community’ and ‘community participation’ within policies, and the rationale and implementation of CP in planning processes are also explored. This research objective also includes examining the involvement communities have (if at all) in health care decision and policy-making processes in Sri Lanka, especially in relation to PHC service delivery. The historical and current discourse on CP in health system reform and decentralization is also examined.

Page | 72

Sub-Objectives:

1.1 To examine how CP is defined, its value ascribed and its rationale articulated within domestic health sector policies, legal frameworks, political manifestos and health sector master plans. [Analyse existing policies and action plans relating to chronic disease prevention and control in order to assess the extent to which CP approaches have been incorporated or promoted as a strategy to reach policy or program goals]

1.2 To examine the rationale for pursuing community participatory approaches in achieving health policy goals or health program outcomes. [Is the rationale for pursuing CP a ‘technocratic’ one, which views participation as a ‘means of ensuring successful interventional outcomes for communities as health consumers’, or rationalized as an ‘end’ goal, manifesting as the ‘empowerment’ of communities in their acquisition of skills, knowledge and experience?]

1.3 To examine the mode and extent to which communities have been engaged in health policy making and program planning processes in Sri Lanka. [Assess the mechanics of how a health system engages communities in policy and planning processes, assess the specific phase(s) within the heuristic policy and program cycle where participation was enabled, where it was sustained, and where it dissipated.]

1.4 To examine the current state of decentralization within Sri Lanka’s health care system, and describe the devolution of decision-making power within the health system. [Analyse the level of participatory engagement between central, provincial, district and primary health care units within the health system in planning health interventions.]

1.5 To analyse the extent to which community participatory approaches are enshrined within routine PHC worker duty lists and reporting and evaluation mechanisms. [Mapping existing human, financial and material resource allocation at primary care level to enable community participatory practices.]

The second research objective aimed at exploring the meaning and value assigned to CP from the perspectives of PHC workers, health executives, policy makers and health administrators. Their knowledge, experience, perceptions, attitudes and commitment toward

Page | 73

the use of CP approaches in the delivery of PHC services, and more specifically NCD prevention and control, were examined.

Research Objective 2: Explore the knowledge, experiences, perceptions and attitudes that PHC workers, health administrators and policy makers have toward the use of CP approaches for NCD prevention and control at PHC level.

Sub-Objectives:

2.1 What is the value and importance that health professionals assign to community participatory approaches? [How is participation defined? What rights and duties does it imply? What do they view as the benefits of CP?]

2.2 To describe the knowledge and awareness PHC workers have of the growing burden of NCDs, their perceived role in addressing NCDs, and the use of community participatory methodologies for NCD prevention and promotion. [How important are the issues of NCDs to PHC workers? What value do PHC workers give to NCDs in relation to their other health priorities, and in relation to their routine work practices?]

2.3 What types of strategies and interventions do health administrators and PHC workers perceive as being effective in addressing the growing burden of NCDs at the community level? What specific strategies do they identify to tackle NCD risk factors?

2.4 What are the technical skills, resource capacities, training, supervision requirements and evaluation processes that PHC workers, health administrators and policy makers perceive to be important in enabling CP at PHC level?

The third objective sought to examine the extent to which community participatory approaches are utilized (if at all) within current PHC health programs and interventions aimed at addressing NCDs at community level, and to investigate the nature, form, and rationale for their use.

As described earlier, the intention was not to ‘evaluate’ CP, measure health outcomes or investigate how effective such approaches are within PHC service delivery. What was crucial in addressing the primary research question was to document how participatory processes emerged within the routine PHC ‘intervention space’ and capture the interactions and dynamics between PHC workers and communities. The scope of this research objective Page | 74

also aimed at investigating the daily practice and work environment of PHC workers to understand how their time and resources are prioritized and allocated in conducting community health interventions. The opportunities and limitations of participatory approaches within PHC units were also explored.

Sub-Objectives:

3.1 To analyse interventions aimed at addressing NCDs at PHC level; identify stakeholders involved in formulating such interventions; outline characteristics of the communities involved and factors such as time and resource commitments by PHC workers; and describe the evolution and development of such interventions and the role played by PHC workers.

3.2 To document the participatory processes in such interventions, notably the interactions between health workers and communities and their manifestation through ‘deliberative’ action or ‘opportunistic’ means. [Did the intervention achieve/contribute to any intended program or health outcomes?]

3.3 To describe the nature of community engagement within the intervention? Was their participation meaningful? [Describe the level of participation from a gradient ranging from little or no participation, ‘consultation’ to intensive community engagement].

3.4 To describe the involvement and contribution of health authorities (if at all) at peripheral and central levels, and to explore how administrative structures within PHC units were mobilized and decisions on resources and time allocations made [Explore the scope for opportunities using participatory approaches within routine PHC work functions].

The fourth and final research objective aimed at synthesizing the findings from previous objectives to identify the determinants and factors that enabled, inhibited, enhanced or sustained meaningful community participatory practices within Sri Lanka’s PHC system. In essence, it formed the basis for the conclusions of the overall study. A secondary objective was to develop a reflective tool to enable health workers, planners and researchers to explore the nature, form and dynamic of CP within a PHC setting in Sri Lanka. The aim was

Page | 75

not to formulate a ‘standardized instrument’ for measuring CP, but rather to construct a simple framework that allows health workers and planners to reflect on the most dominant form or manifestation of CP.

Research Objective 4. In drawing on the experiences of PHC workers, health administrators and policy makers in assessing the health policy and operational environment within PHC settings, to identify the key determinants and factors that enable, inhibit, enhance and sustain meaningful community participatory practices at PHC level.

Sub-Objectives:

4.1 To determine the main factors and determinants that enable and motivate PHC units to engage in community participatory approaches in addressing NCDs. [Which factors and determinants constrain, disable or disempower the adoption participatory approaches? Which factors lead to disenfranchisement and poor sustainability of CP practice? What are the dynamics/relationships between such factors and whether they work independently or synergistically in enabling or disabling CP in PHC work practices and culture?]

4.2 To develop a robust health system model that encapsulates CP within any given health setting or intervention.

3.8 Section Summary

In order to explore the complex and multi-dimensional phenomenon of CP, I applied an ecological model of health system analysis within a complex health systems approach. The research aim and objectives were constructed in order to identify the broader determinants and factors that enable, limit, enhance and sustain CP practices within PHC settings in the Sri Lankan health system. The core elements of this early model evolved to form the basis of the conceptual framework used to analyse CP in PHC and explored in the last two chapters of this thesis; modifications emerged during the data collection and iterative analysis of data.

Page | 76

Chapter 4: Research Methodology, Study Design and Methods

Image 4. PHM discussing with children and mothers attending a community health nutritional program the plan of activities for the session.

Page | 77

4.1.1 Chapter overview

My study aimed to assess the legal, policy and administrative environment for CP in Sri Lanka, identify the meaning and value assigned by health workers and administrators to CP, and investigate the experience and challenges faced in its practical implementation within routine PHC practice. In doing so I aimed at building a conceptual model for exploring community participatory approaches to address NCDs through a health systems framework.

This chapter describes the methods I used to collect data in order to answer the research questions outlined. A multiple-method [QUAL+quant] research design was utilised to obtain data. As depicted in Figure 13, a step-wise progression towards the formulation of a conceptual model for exploring CP in PHC is followed. Each corresponding step (indicated as a unique colour icon) is ‘unpacked’ and elaborated further in this Chapter.

Figure 13. Graphical Overview of Chapter 4

Page | 78

4.1.2 Rationalizing the methodological paradigm

Research can be broadly classified as positivist, interpretive or critical based on the philosophical assumptions (the epistemological stance) adopted by the researcher. The broad epistemological stance of my research is from an interpretive-constructivist perspective (289). This is because I was interested in exploring the phenomenon of CP through the constructions of how field level health workers, policy makers and health administrators themselves defined CP, described their experiences and values related to CP, and how CP is actually manifested in routine primary health care practice. The interpretive- constructivist researcher explores participants' views of the situation being studied, whilst recognising the impact on the research of their own background and experiences (290). While I was interested in the ‘lived experience’ of the participants, I found myself in agreement with Miles and Huberman’s (1994) pragmatic view that these meanings and intentions are worked out within the frameworks of these social structures: “structures that are invisible, but nonetheless real... and exist objectively in the world and exert strong influences over human activities because people construe them in common ways” (291).

It was therefore important to explore the realities of the context in which health workers and administrators were situated, their interactions with communities, the health system and the overarching administrative and operational parameters of their routine practice. Understanding their intervention environment also included exploring policy and legal framework through deeper analysis including of job function. This approach embraces a health systems perspective (elaborated in Section 3.3) for understanding the complexities of the organizational environment (‘ecological approach’).

4.1.3 Study design

A multiple-method qualitative research study design was used, underpinned by a constructivist epistemology. Such study designs have been used within PHC settings, and as Creswell (2004) indicates, “have the potential for rigorous, methodologically sound investigations in primary care” (292). The research method most appropriate for

Page | 79

answering each research question was selected. Figure 14 presents an overview of the study design, research participants and how each research objective links to an appropriate method.

Although an interpretive-constructivist paradigm yields to an overall qualitative methodological approach (293), the observational and focused ethnographic method used – for instance, in exploring research Objective 3 – also generated layers of quantitative data on routine work practices of PHC workers and their work settings. The qualitative and quantitative data generated using this multiple-method approach helped capture the ‘intervention space’ available to PHC workers for engagement with communities, and described resource realities and program dynamics within each primary care setting. When combined with the interview data, the use of multiple methods may build a more comprehensive understanding of the determinants that inhibit or enable the adoption of CP approaches. Health system scholars have suggested that adopting a multiple-method approach results in “expanding the breadth and range of research inquiry” (294), and can play a valuable role in health services research (295).

Research Objective 3 adopted a descriptive case-study methodology to explore NCD interventions catalyzed by PHC workers that utilized community participatory approaches. The descriptive case study approach is used to describe an intervention in the real-life context in which it occurs (296). Qualitative data were collected through in-depth interviews with health workers, community members and from observational visits captured through video-graphic methods. A hallmark of a descriptive case-study is the use of such multiple methods (297).

In reviewing the literature, I appreciated that no single methodology is superior to another in every circumstance. What is critical, however, is the selection of the appropriate research methodology for the inquiry at hand (292). Indeed, to pursue my research goals, each method was specifically matched to answer a given research objective. Documenting the CP landscape in Sri Lanka’s health system requires unpacking complex dynamics between health authorities, PHC and communities across the selected health care settings. A multiple- method approach enabled building an understanding of the underpinning layers of evidence. Page | 80

Figure 14. Overview of study design. Linking conceptual model and research objectives with overarching methodology, analytical framework, research methods and participants.

Conceptual Research Objective Research Methods Participants and Setting framework [Overarching Methodology]

(Analytical frame)

Objective 1: Analyse the extent 1) Document review. Review of following As part of data gathering policy to which CP approaches are documents: Sri Lanka’s national health makers at Central level MoH were enshrined within national health policies and domestic legal frameworks, interviewed to identify relevant policy policies, domestic legal Policy manifestos, draft heath policy and legal content. Consultative frameworks, PHC worker duty lists documents, MoH Circulars, PHC worker meetings with experts in Sri Lankan and routine job functions of Sri duty lists, training curricula, routine health law (at Colombo university) Lanka’s health system reporting formats of PHC workers and was also undertaken. relevant peer-reviewed publications) [Document analysis using a 2) Systematic literature review of peer- deductive approach in examining reviewed and grey literature pertaining CP in legal and policy documents] to CP in Sri Lanka

(Content Analysis)

Page | 81

Objective 2. Explore the a) Current Policy makers & planners knowledge, experiences, 1. In-depth Interviews with Key informants at Central level perceptions and attitudes that such as health administrators (HA’s), policy b) Health administrators at District PHC workers, health makers, health managers at district level. level administrators and policy 2. Group Interviews (with PHC workers) c) Technical experts/advisors at makers have toward the use of Central level CP approaches for NCD d) Health Service focal points at prevention and control at PHC District level level. e) Technical experts/advisors from Sri Lanka’s health sector development [Qualitative Research partners (e.g. WHO, UNICEF) Methodology] f) PHC workers at District level: Medical officer of Health (MOH), (Thematic Analysis using principles Public Health Inspector (PHI), Public of Grounded theory approach) Health Midwife (PHM), Senior Public Health Inspector (SPHI), Senior Public Health Midwife (SPHM), Health Volunteers (HVs). 3. Survey of policy makers and planners g) Persons involved in health policy [Quantitative Research making in Sri Lanka (includes Methodology] academics, experts, peak bodies, retired policy makers and planners (Quantitative Analysis) of MOH)

Page | 82

Objective 3. To explore the a) PHC workers extent to which community 1. Key informant interviews with District b) District Level Health administrators participatory approaches are health administrators and PHC workers to c) Community members/beneficiaries utilized (if at all) in NCD ‘map’/identify existing interventions aimed engaged in the intervention and their prevention and promotion at addressing NCDs using CP approaches family members interventions, and investigate the 2. Focused Ethnographic study and nature, form and rationale for Observational follow up study of PHC their use at PHC level. work practices and worker intervention 3. Descriptive case-study of NCD [Qualitative Research; and interventions utilizing CP approaches Descriptive Case-study developed through interviews with PHC methodology to map CP workers and community members. Data interventions within routine collected from observational visits and also practice] captured through video-graphic methods. (Thematic Analysis using principles of Grounded theory approach)

Objective 4. In drawing on the experiences of PHC workers, health administrators and policy makers in assessing the health policy and operational environment within PHC settings, to identify the key determinants and factors that enable, inhibit, enhance and sustain meaningful community participatory practices at PHC level.

[Hermeneutical analytic process - analytic method that integrates study findings exploring CP in PHC into a theoretical canvass.]

(Formulates an overall health systems theoretical framework)

Page | 83

Figure 15. Overview of research methods, types of data generated and modes of data analysis

Page | 84

4.1.4 Research methods, procedures and study instruments

This section describes the methods and strategic approach used for collecting data. Care has been taken to present each method as a discrete heading, recognizing however that there are overlaps that exist. Figure 15 (previous page) provides an overview of research methods used and analytic methods applied.

4.1.5 Participants

As outlined in the overview of the study design (Figure 14), the study population consisted of MoH professionals who were PHC workers, Medical Officers of Health (MOH) responsible for each PHC unit, health administrators at district or central levels and policy makers responsible for national level service planning and policy formulation. Those involved in health policy formulation participated in the policy maker survey.

The focused ethnographic study component captured perspectives of community members and other stakeholders during field level observations and when documenting in-depth case studies on NCD prevention interventions.

4.1.6 Sampling strategy and recruitment

Sampling strategies in qualitative research are largely purposive and determined by the purpose of the study and aims of the researcher (298,299). The challenge for me was to capture perspectives of PHC workers from the large range of PHC settings in Sri Lanka’s health system. District health systems are non-homogenous complex systems exhibiting a wide degree of diversity (300). In undertaking health systems research it was especially critical for me to ensure that I was capturing a range of perceptions of PHC workers from a multiplicity of settings.

In qualitative research, researchers need not seek representativeness through statistical sampling since the strengths of qualitative inquiry emerge from understanding issues related to how and why (301-303). However, a common misconception is that numbers and generalizability are ‘unimportant’ when ensuring adequacy of a qualitative sampling strategy (299,302). A 2008 National Science Foundation Report suggested samples Page | 85

purposively selected may be too small to support claims of having achieved either informational redundancy or theoretical saturation (303). The purposive sampling method employed in selecting the study settings and seeking to allow for maximum variation are described below.

In 2008 the MoH, in partnership with the Sri Lanka Finance Commission and the World Bank, devised an indicator matrix to identify the ‘best, worst and average' performing district health systems in Sri Lanka (304). Undertaken over a period of three years, the district performance classification was based on tracking a wide range of health system indicators such as population health outcomes, morbidity and mortality registers, human resource for health capacities and health care financing. The performance classification was intended to be used as a guide for policy makers and district administrators to strengthen district and provincial level health system planning.

I used this district performance review as a means of rationalising the selection of districts in my study. A district from each performance classification was selected. The rationale for such district selection was not undertaken to perform cross-district comparisons or apply statistical probability to final results but rather as a means of selecting a wide range of PHC units across the country for the study, thereby allowing for maximum variation.

The selected districts (shown in Figure 16) were Pollonaruwa, Hambantota and Nuwara- Eliya districts. There was a high degree of diversity in the selected districts based on urban to rural population densities, ethnic profiles, population distribution, conflict experience and socio-economic gradients. These are explored further in the results section.

Page | 86

Figure 16. Geographical map of districts selected for study

As described in Chapter 2, MOH unit forms the most fundamental PHC unit of Sri Lanka’s health system, critical for the exploration of CP. Rather than undertake interviews with PHC workers within a single MOH setting, PHC workers from six different PHC settings, across three districts were selected. Discussions were held with the Regional Director of Health Services (RDHS) in each district to select the MOH units for the study. The choice of PHC units was made arbitrarily by each RDHS. Again, this step was undertaken not to infer any statistical strength or probability. The six selected MOH areas were Thamankaduwa and Lankapura (Pollonaruwa District); Beliaththa and Tangalle (Hambantota District) and Kothmale and Nawathispane (Nuwara Eliya District).

Page | 87

In qualitative research, sufficient data needs to be collected to achieve a ‘saturation threshold’ and ensure methodological rigor (305). Saturation thresholds may be reached after no new concepts and themes emerge from undertaking further data collection activities. Six PHC settings across three distinct district settings were needed to ensure maximum variation sampling as indicated earlier. However, since I intended to identify and ‘map’ community-based NCD prevention interventions carried out within each MOH, and administer to participants the Participatory compass and other such tools elaborated later in this section to ascertain the level of participatory action, then including a large number of PHC units was deemed useful.

4.1.7 Ethical considerations

Ethics Committee approval was obtained from the University of New South Wales Ethics Committee, Sydney, Australia (Ethics approval code: HREC 11240) and the University of Sri Jayawardanapura Ethics Committee in Sri Lanka (Annexure 3 and 4). The study also received formal endorsement from the MoH, Sri Lanka.

Written informed consent was obtained from all interviewees and participants in all research components, and provisions were made to allow participants to withdraw at any time without any negative consequences. Permission to digitally audio record interviews and video record persons involved in community health interventions was requested. The completed short-films on NCD interventions were shared with the PHC workers including national and district health administrators via workshops. Multiple copies were provided for their use and distribution as part of the dissemination of research findings to those that were directly involved in study.

Page | 88

Research Methods

A multiple-method approach was used adopting both qualitative (QUAL) and quantitative (quant) research methods, although qualitative methods dominated the study design (QUAL+quant).

QUALITATIVE METHODS

The qualitative researcher’s goal is to attain an ‘insider view of the group under study’ (306). Qualitative methods usually entail rich descriptions of social phenomena by encouraging participants to speak freely to obtain insight into the phenomenon perceived or experienced by the participant (307). Building a partnership with study participants leads to deeper insights into the context under study, adding richness and depth to the data (307). My research strategy involved developing partnerships with district health administrators and PHC workers at field level over a period of up to two and a half years. I was able to observe, identify and document interventions utilizing CP approaches at primary care level.

4.2 Key informant interviews

4.2.1 Selection of participants for key informant interviews. Key informant interviews (KI) formed the main qualitative method for exploring the perceptions and attitudes of health administrators towards the use of CP approaches in addressing NCDs. KIs were conducted with health administrators from central, district and divisional (PHC) levels. In the exploration of PHC interventions using CP approaches, community members were also interviewed.

The MoH health administrators interviewed were of three types. Central level health administrators (referred to as ‘Directors’) responsible for country-wide portfolios at the nexus of primary care, NCDs and policy were interviewed. Technical experts or advisors to these Directors called ‘Consultant Community Physicians’ at National level were also interviewed. Second, health administrators responsible for leading specific health programs

Page | 89

and in health stewardship functions at district, village and PHC level health were interviewed. These included for instance, the health program managers working under the guidance and purview of the RDHS (Figure 6, Chapter 1). The third tier of administrator is the Medical Officer of Health (MOH) responsible for managing each PHC unit.

Selected non-MoH stakeholders were also included and were based in UN agencies operating in Sri Lanka such as the WHO, scholars, donor agencies and the World Bank. These are responsible for providing technical advice and assistance to the MoH in addressing NCDs and health systems reform and therefore had valuable insights into systems development and operation.

4.2.2 Procedures in undertaking KIs and ensuring methodological rigour

A semi-structured interview format was used as a guide to facilitate the interview (see Annexure 5). In undertaking the interviews, I was not restricted within the confines of these questions and prompts. Rather, I encouraged the free exploration of ideas as they emerged during the interview based on participant responses. Ultimately this enabled a richer (308) and more grounded approach to exploring the phenomenon of CP in health.

In opening the interview, participants were encouraged to express their views and values on the definitions, role and importance CP plays (if at all) in PHC service delivery. Next, their experiences of utilising CP approaches were explored. The dynamics of existing strategies and interventions to address NCDs at the community level were discussed. The final component involved the application of the Community Participatory Compass (see next section).

All KIs were conducted in English and interviews were audio-recorded in digital (mp3) format. Hand-written notes were taken with digital pen using EvernoteTM software. Audio- recording and transcription more accurately reflect participants’ views than contemporaneous researcher notes. One problem affecting this linear translation process is how misinterpretations of language, for instance, can be ‘carried downstream’, affecting content validity. I support MacLean’s (2004) view that the process of transcription is not merely verbatim copying from recording devices and ensuring correct semantics, but rather capturing the intent of the participant and accurately describing the ideas underpinning Page | 90

what was said (309). I adopted a collaborative and iterative approach to translation that is elaborated in the section on research rigour (Section 4.8).

4.3 Group interviews

Group interviews (GIs) were used to facilitate discussion among PHC workers on CP. GIs undertaken with workers within PHC service units enable better insight and understanding of the functions and activities of the unit as a whole. GIs have been used in health systems research to deconstruct experiences and reflections within the PHC intervention space (310,311).

4.3.1 Procedures in undertaking GIs

Official letters of correspondence were facilitated with the support of the MoH to each of the three Regional Directors of Health Services. Letters outlined details of investigators, supervisors, UNSW institutional details, study aims and purpose in undertaking field research and consent forms. Individual district level research plans were formulated with participation of the RDHS in each of the three districts. Each RDHS agreed to ensure an equal number of public health inspectors (PHIs) and public health midwives (PHMs) be present at each MOH unit for the GI, that there be at least six and maximum of 10 persons in a GI, and that the MOH be interviewed separately. Logistical arrangements were made with each MOH and a final ‘confirmed’ list of participants compiled the week before interviews commenced. All participants were provided an information sheet on the objectives of the study and provided with consent forms. Participants were assured that their confidentiality and anonymity would be maintained, and that they were able to withdraw from the study at any point.

I aimed at conducting the interviews in a trusting environment that encouraged a diversity of views and that did not attempt to persuade or coerce (312). Interviews were held in the staff conference or meeting room of each MOH office. At the start of each group interview, I drew a quick sketch of where each PHC worker was positioned (as shown in image below). These positions were allocated codes and were used to reference each response. This process was useful in attributing comments made by each member of the group.

Page | 91

Figure: Sketch of PHC worker positioning during interview

4.3.2 Group composition

The supervising Medical Officer of Health was not included in the group interview with PHC workers. It was assumed that the presence of the MOH might reinforce authority and hierarchy, which may keep some members of the PHC unit from sharing experiences and opinions. As mentioned earlier, from a health systems perspective, it was important to explore the perspective of PHC workers as a single PHC service delivery unit. The inclusion of both PHMs and PHIs were thus made within a single group interview format rather than undertaking these separately.

4.3.3 Development of interview guide

A semi-structured interview guide was formulated to ascertain participant perceptions on CP approaches to PHC (Annexure 5). The development and pre-testing of the interview guide occurred through three phases. Early drafts were first presented to the formative research group and debated, and the feedback synthesized to subsequent versions. Two ‘mock’ group interviews were conducted involving research assistants and MoH staff (consultant community physicians and a Senior Health Director). The interview guide was again modified after feedback. The discussions were useful in order to familiarize terms, concepts and question flow, including instruments such as the Community Participatory Compass tool.

The second pre-test, was a field pilot involving five PHC workers from an MOH area in Pollonaruwa district. This field-pilot was particularly useful in introducing the prompts (i.e. questions aimed at stimulating and eliciting discussion around sub-themes). Considerable flexibility was allowed during this pilot interview to enable participants to discuss issues that were most important to them and clarify questions posed. After the pre-test phase, a final Page | 92

workshop was held with field level researchers to discuss the ‘lessons learned’, revisit flow of the interview guide, explore transition questions and discuss interview closure strategy.

‘Community Participatory Compass’ instrument

The Community Participatory Compass tool (labelled succinctly as ‘P-Compass’), was developed after assessing multiple frameworks on measuring CP that emerged through an in-depth review of literature (Table 4). The tool was introduced to participants at the end of the interview as a means of gauging the level of CP in the PHC setting in which they worked in.

The P-Compass is a reflective instrument, a ‘compass’ that provides general direction. It is not a prescriptive map that provides directive measurement or detailed information. The instrument was not intended to evaluate or measure the level of CP (as outlined in Rifkin’s spidergram model for instance highlighted in Table 4), but aimed at providing a quick yet robust way of capturing the nature and dynamics of CP from the perspective of health professionals within the health system.

This section describes the theoretical underpinnings and characteristics of the components of the P-Compass.

4.3.3 Developing the instrument

Health instruments are often used to reliably measure a phenomenon of interest and ensure repeatability of the result. In contrast, the P-Compass was not constructed with the intention of formulating an empirical tool in providing a quantifiable measure of CP. Rather it aimed at providing a reflective tool to enable and empower PHC workers, managers and policy makers to reflect on their routine interventions and service culture and utilize the P-Compass to capture the main dynamics of community participatory action.

As described in Part 2 of the literature review (Section 2.2.1), no existing instruments or tools specifically designed to be applied by health workers themselves in characterising CP were identified. Most available models also required researchers with knowledge and understanding of the instrument mechanics to apply them in a research setting. Commonly

Page | 93

used models such as Rifkin’s Spidergram (212), for instance, were usually applied by researchers to assess the level of CP within specific programs or interventions.

The P-Compass is intended to be used as a reflective tool in gauging the overall perceptions of PHC workers and administrators of the participatory dynamic and characteristics of CP in a health care setting as a whole. The intuitive value of the P-Compass is vested in the simple construction of concepts and the ease of understanding them.

4.3.4 Components and measures of the P-Compass instrument

As described in Part 2 of the literature review, a major factor influencing the meaningful manifestation of participatory action is how power is reshaped between health professionals and communities (313) (314). These power dimensions are reflected in the P-Compass model (depicted in Figure 17) as the following: the ‘Degree of Participation’, ‘Community Participants action’ (the mode of community management), ‘Mobilization Method’ and ‘Public Involvement Continuum’.

Degree of Participation with communities ranges from ‘collective action’, ‘co-learning and cooperation’ to ‘consultation’, ‘compliance’ and ‘co-option’ (column two, Figure 17). ‘Community action’ indicated in Figure 17 ranges from situations where communities have little control in intervention processes, where they are the passive recipients of health interventions (working ‘on a community’), to a participatory dynamic in which health care units engage communities in jointly planning parts or entire aspects of the intervention (working ‘with a community’). ‘By community’ refers to situations where communities and health workers jointly engage and have an equal say in decision making processors.

The Public Involvement Continuum within the instrument ranges from providing communities with information and communication on a health problem (‘inform’), to workers actively supporting communities to develop skills to identify and implement tailored solutions and advocate for health programs (‘empower’). The power dimensions are thus thematically linked to each other. This allows the user to simply identify a conceptual frame, and then ‘work across the page’ to the far left hand side to define the overall ‘level of participation’. Participants were also guided through and assisted with using the tool (see next section).

Page | 94

The P-Compass is presented in the form of both a ladder and continuum model which enabled the user to visually refer down and across to any form of power dimensions, and then refer to the corresponding ‘Level of participation’ to obtain the participatory score. Alternatively, the user may start with any given level of interaction, and refer across the matrix to identify what they generally regard as the participant form or CP dynamic. Since the P-Compass construction is in the form of a dynamic continuum, it allows the reader multiple entry points to explore CP. The usefulness of a reflective tool lies in its ability to allow users to scan the dimensions and arrive at a tentative assessment which can then be reaffirmed on deeper consideration.

If a health worker felt that routine PHC services undertaken placed a high value on community partnership, community ownership and empowerment in planning and delivery health services, then a P-Compass score card reflective of a ‘high’ to ‘very high’ level of CP was recorded. If the dominant service culture was for a ‘vertically driven’ intervention mode that did not enable CP, this was reflected as a ‘lower level’ of CP with a low participatory score.

The numerical classification (0 to 4) sought to reflect the participant’s perceptions of the overall dynamics of CP within the health unit or program. The lowest modes of participation, characterized by ‘0’, involve little or no community engagement, where information provision is the basis of interaction between community and health team. The health system here works ‘on communities’ rather than ‘with communities’.

At the highest level of participation (‘4’) termed ‘collective action’, the community is meaningfully engaged, with co-learning and co-sharing occurring with health professionals. At these levels, community participants have an equal level of decision-making power and influence as the health system focal points, and interactions occur in an iterative manner between health workers and community members. At the ‘co-learning’ and ‘co-operation’ level of participation, the health worker’s focus is to 'collaborate' and partner with the community in decision-making. At the next ‘consultative’ mode of participation, the community is consulted and invited to ‘give feedback’, rather than to meaningfully become involved. In the ‘co-option’ level of CP, community members are the passive recipients of an intervention.

Page | 95

In the final category the community is kept uninformed of the service, program or policy intervention affecting them.

In utilizing the P-Compass, it is important to note that no negative value is placed nor inferred with regards to ‘low’ levels of community participatory action such as ‘co-option’ and ‘compliance’, nor is a positive value assigned to the higher scores. Distinctions are made to describe the dominant form of engagement by communities in a given PHC setting. Thus CP that involved meaningful and iterative modes of engagement between health workers and communities was not necessarily ‘more positively valued’ than CP in which communities were only ‘consulted’.

To summarize, the P-Compass tool was designed to provide a quick yet robust way of capturing the nature of CP from the perspective of those within the health system. The P- Compass needs to be viewed a tool that offers a reflective lens on the form of CP rather than an instrument which provides a detailed map of participation and explicitly measures it. The P-Compass was used by health workers and administrators during the research interview process as a means of gauging the level of participatory action in the health settings and programs in which they worked.

The P-Compass instrument was translated into both Tamil and Sinhala languages, pilot tested and underwent many iterations before final field use. The process of field-testing and formulation are presented in Section 4.8.

4.3.5 Method of applying the P-Compass instrument

Key informants and participants in group interviews were provided with a printed copy of the P-Compass card. The card was translated and printed in English, Sinhalese and Tamil. A multi-media projector was used in the group interviews to project the P-Compass. The CP typologies were presented, and the interviewer would then briefly introduce the conceptual framework and scoring pattern using the graphical interface guide described in Figure 17. In developing the instrument, care was placed on designing each graphical node to best represent the different forms of CP.

Page | 96

PHC workers were asked to reflect on their own practices and interventions in their PHC setting and circled the level which they felt was appropriate. The cards were submitted at the end of the session to the interviewer or field assistant. Participants were given as much time to complete the P-Compass as they needed, and were free to withdraw, discontinue or provide feedback later by contacting the research team. The instrument was completed on an individual basis as a personal reflection rather than as a result of a group discussion in which consensus was sought.

The process of administering the P-Compass within interview contexts was rehearsed during the formative research phase where pilot GIs were conducted in Pollonaruwa and Nuwa- Eliya districts. The P-Compass was administered at the conclusion of each interview. The average presentation time taken was 5-7 mins with participants averaging 1.5 to12 mins in completing the form. Most surprising was how quickly the PHC workers understood the conceptual framework presented in the P-Compass. My field notes (recorded on EvernoteTM) from the conclusion of the field-pilot stated: “It was almost as if they [the five PHC workers] intuitively deciphered from the graphics the meaning of the [participatory score card] scale…to think that this was added at the last minute as a way of breaking up the monotony of text with images!… This totally squashed fears I had earlier from when we [the research group] debated this...they didn’t get it as easy. It really flips the script of what was expected” (5.6.0 Evernote: 5th March, 04:7.20pm).

Page | 97

Figure 17. The Community Participatory Compass (P-Compass) instrument

Score Level of Participatory Power Dimensions Health worker/setting participation activity type

Degree of Community Mobilization Public Involvement Continuum Participation participant Method action

Very High Collective Has control Community Empower: To actively support Asks community to identify action Participatory stakeholders in developing the problem/issues and assist Development processes and structures necessary them in providing input to to identify issues and to implement make key decisions on goal 4 “By solutions. and strategies. The health community” worker/system enables communities to accomplish these goals. Co-learning Has Community Collaborate: To partner with the Identifies and presents a delegated Collaboration stakeholders in each aspect of problem to the community. authority decision-making (including Defines limits and asks development of alternatives and community to make a series High identification of the preferred of decisions which can be solution). embodied in a strategic plan/action which is led by 3 community. Co-operation Plans jointly Community Presents defines plan and Collaboration invites input from communities. An iterative process of planning is “With followed. community”

Page | 98

Moderate Consultation Advises Community Involve: To work directly with Presents a plan and invites mobilization stakeholders in planning and policy questions. Prepared to processes to ensure their concerns change plan only if and aspirations are consistently absolutely necessary. “For understood and considered. community”

2 Moderate Compliance Is consulted Community Consult: To consult stakeholders on Makes plans, announces it. consultation draft plans or on issues; feedback Community is convened for from stakeholders has an influence informational purposes. on decisions. Compliance is expected.

Low Co-option Receives Community Inform: To provide the stakeholders Communities are passive information compliance with balanced and objective recipients of intervention. information to assist them in understanding the problem, 1 alternatives, opportunities and/or “On solutions community”

0 none none none none. Community told nothing.

Page | 99

4.4 Observational study and use of focused ethnographic method

The socio-ecological and social determinants approach presented in Chapter 2 highlighted the importance of defining the factors that affect health care service delivery (315). The skills and competencies of a PHC worker in catalysing participatory approaches, and how the worker navigates or mitigates the following factors within a health system, may ultimately influence the extent to which communities are engaged:  Health system factors: human resource gaps within PHC units, availability of material and monetary resources for health interventions, socio-political context, geographic environment;  Individual factors: such as attitudes towards CP, commitment to principles of CP;  Skills, competencies and capacities: knowledge of community engagement methods such as catalysing effective NCD prevention initiatives.

A primary objective of my research was to explore the phenomena of CP in Sri Lanka’s health system and study its manifestation within routine PHC practice settings. Seeking to understand context and identify factors that may act individually or synergistically to influence community engagement is therefore important: focused ethnography was utilized to do so.

4.4.1 Focused ethnographic method

“focused ethnography allows the researcher to better understand the complexities surrounding issues from the participants’ perspectives while bringing the outsider’s framework to the study…” - Roper and Shapira, 2000 (316).

Ethnography is a form of naturalistic enquiry that is a principal approach of anthropologists, and is particularly valuable because of the attention it gives to context (317). It is an observation method that involves an investigator viewing participants as they work in their natural settings in the field, and where the investigator records notes on the activity taking place (318). Focused ethnography is an applied research technique (319,320) that is particularly useful for researchers who wish to focus on a distinct issue like CP within a specific context, amongst a specific group of people (such as PHC workers) living or working in a bigger society or system (such as a district health system) (316).

Page | 100

Focused ethnographic methods have been particularly useful in understanding health worker practices (319,321,322) and offer a holistic way of exploring the relationship between the different kinds of evidence that underpin health care practice (323).

Focused ethnography allow the researcher to gain better understanding of a specific phenomena of interest (316). In contrast to traditional ethnography, focused ethnographies tend to have pre-selected topics of enquiry that inform the participant observation process (319). Applied ethnography fieldwork may be premised on specifically formulated research questions focused on a discrete group within a community or organization (as in the case of Research Objective 3 of this study). The main features of focused ethnographies are listed in Box 3.

Box 3. Characteristics of focused ethnographies as described by Muecke (324)  Problem-focused and context-specific  Pre-selected topics of enquiry that inform participant observation  Focus on a discrete community or organization or social phenomenon.  Involvement of a limited number of participants.  Episodic participation observation.  Focused ethnographies can make effective use of video or audio recordings and even subsequent ‘data sessions’ whereby the data is ‘opened socially to other perspectives’ in groups.  Used in academia and in development of healthcare services research. For instance, Spiers and Wood (2010) explored the perceptions and actions of community mental health nurses in building a therapeutic alliance and the factors that helped or impeded its development of such community mental health services (325).

The distinctions between conventional ethnography and focused ethnographies have been discussed in detail by methodological scholars (320,326), although differences between this applied form and traditional method have been described by Hart (2004) as “differences of degree, rather than of kind” (327).

Both qualitative and quantitative data are captured within a focused ethnographic study. As Savage (2006) states “applied ethnography sits like a chameleon within the tradition of qualitative research” (323). The applied ethnographic perspective in my research allowed me to capture the routine health practices of PHC workers and provide insight and interpretations relating to their real world interactions. I used a combination of direct Page | 101

observation, interviews, analysis of notes, informal social interactions and documenting context through audio, video and images.

Utilising observational assessments, I captured data on the nature of community health interventions, resources utilized by PHC workers to engage with communities, average length of time for PHC workers to undertake an intervention, and costs incurred in community outreach activities. Participating as an active observer, I was able to take part in the ‘daily routine’ of PHC workers and witness how they engage in health interventions and meetings and how they interact with communities, patients and other health workers. During the fieldwork I was also able to capture information through the use of digital HD video camera, still-camera and electronic field journal.

Through sequential capturing for one month of data on worker activities, I was able to compile a quantitative dataset. By combining datasets of PHC workers followed-up across other settings, I was able to construct a detailed PHC practice map. The applied ethnographic approach enabled me to structure my observations by focusing on those most important in studying the phenomena of CP within the intervention space of a PHC worker.

To summarize, the four parts of the focused ethnographic and observational study are presented in Figure 18.

Page | 102

Figure 18. The four parts of the focused ethnographic and observational study

Page | 103

4.4.2 Part A: District health system profile

The structural components of a health system (260), such as human resources for health or financial resources, play a role in how health services are delivered at community level. Health system profiling is important for capturing such determinants and providing contextual layers of information to understand the challenges within the PHC intervention space.

Data collection procedures

Socio-demographic data on populations across the three selected districts were captured from MoH and Census data. The burden of chronic diseases was gathered from disease surveillance registries held at the district level. Maps of currently functioning PHC institutions were collected from the office of the RDHS from each selected district in the study.

The profile exercise also aimed at capturing the human resources for health dynamics within each PHC setting. First, the current registry of all active workers and vacancies by type and MOH unit were obtained. The field research assistant verified the actual number of PHC workers working within each setting against the number allocated by district level plans. Budgetary allocations for NCD prevention and promotion programs, health promotion initiatives and resources provided to PHC workers to undertake community outreach activities (such as allowances for community outreach), and information, exchange and communication (IEC) material relevant to programs on chronic disease prevention were also gathered from RDHS and MOH offices. Monitoring and reporting documents pertaining to PHC worker practices, reporting templates, epidemiological returns and outreach clinic files were collected for analysis.

4.4.3 Part B: Structured observational study of PHC workers

Observing how PHC workers engage with communities during routine community health interventions may provide useful insights into the dynamics of participation. Structured observational techniques are an effective research method for studying executed tasks and processes in public health services (318,328,329).

Page | 104

Participant selection and data recording procedures

Two PHMs and two PHIs from each selected MOH (total of 12 PHC workers across 3 districts) were invited to participate in a structured observational study documenting community health interventions (CHI). The observational period was for a minimum of one month.

PHC workers were first requested to submit their monthly work program (known as ‘advanced programs’) which outline the clinic schedules and outreach activities. Contact details were shared by those workers participating in the study and the field researcher. Communication was made on the day prior to interview in order to confirm scheduling so as to accompany the worker to the intervention site or program. As the observational month progressed, most PHC workers would ask us to convene at their home or clinic office and then proceed to the community intervention site.

The observation was ‘structured’ because it was limited to observing interventions undertaken by PHC workers that involved communities, such as community-based outreach programs. These included antenatal clinics and home-visits (for PHMs) and school medical inspections and adolescent health clinics (for PHIs). An observational guide was used to record characteristics of such interventions. An observational checklist was formulated with data nodes that included: health intervention type; profile and composition of community participants; resources used by PHC worker; and time taken to conduct intervention. Descriptions of techniques and strategies workers used in engaging with communities and relevant data, such as the length of time a health worker devoted to delivering a health education talk (as a proportion of total intervention time), were also documented.

‘Unstructured’ observations on the setting, for instance the sense of remoteness or harshness of terrain in the PHC service coverage area, were also recorded. When as an observer I would witness something unusual or interesting, I would ask myself “Well why did he do that?” and explore further. Engagement between a worker and their community during such outreach services and home-visits were also recorded in field notes. A digital notebook software (EvernoteTM using an Android smart phone) was used to recorded field notes.

Page | 105

Although time consuming and labour intensive, such direct observational methods allowed me to view what PHC workers actually do in routine practice.

4.4.4 Part C: Mapping of NCD prevention and control interventions at PHC level

In pursuit of participation in routine practice settings

The goal of the ‘intervention mapping’ exercise was to identify existing NCD prevention and health promotion interventions that utilized CP approaches. A characteristic feature of vertically driven interventions is that they are usually well resourced for the duration of the pilot phase (often as external donor driven projects). After heralding the success of the pilot, often through a formal evaluation, the interventions are labelled ‘best practice’. The intervention protagonists often with support of health authorities ‘recommend’ these as best practice, where the expectation is for local health workers to diffuse such innovations into routine practices [49]. Scientifically observing and examining the phenomena of CP within functions of a ‘routine’ health system, devoid of any one-off pilot projects is therefore critically important to understanding the translation and sustainability of CP practices in existing ‘real world’ health care settings.

Intervention selection and data collection procedures

Community health interventions addressing NCDs that emerged or were organized by the Government’s district health system, and not a part of any ‘vertically driven’ pilot project or an NGO driven program, were identified. The inclusion criteria established for mapping NCD prevention and promotion interventions are summarized in Figure 19. By canvassing health administrators and consulting PHC workers across the three district settings over the field research period, a map of the NCD prevention ‘interventional landscape’ was synthesised.

Page | 106

Figure 19. A stepwise approach to mapping PHC interventions aimed at addressing NCDs

4.4.5 Part D: In-depth case study analysis of community participatory NCD prevention interventions

NCD prevention interventions captured from the mapping exercise (Part C) were assessed for those that were currently active and utilized CP approaches. The PHC workers that were the protagonists responsible for the intervention or who were actively involved in implementation were contacted and face-to-face discussion held (in coordination with RDHS). Informed consent to undertake an in-depth-case study analysis was obtained from each PHC worker after outlining study goals5. Permission was also sought to digitally video-record (in High Definition video) the community engagement processes involved in each intervention for a 6-month follow-up period.

Focused ethnographic methods were used to document the intervention process and map the resources and capacities that were utilized to sustain intervention. This involved documenting the motivations, rationale and ‘interventional footsteps’ of the PHC worker as s/he engaged with communities, and capturing the health system and community resources used. Interviews were conducted with community members and relevant health and non-health sector stakeholders at village level that were involved in the intervention. Such interviews also

5 Note: Since the PHC workers had already participated in the interview process, they were familiar with the overall study objectives and had a familiarity with the research team. Page | 107

elucidated the phases within the project intervention cycle (i.e. planning, implementation, monitoring, and evaluation) where CP was manifested and how such community engagement occurred. A stakeholder map of those actors involved in this intervention and their roles and responsibilities were also captured. Observations were made as to how communities engaged with the PHC worker, applied intervention strategies and what effect (if at all) such processes had on their wellbeing and development. Informed consent was also sought to record these video-graphically.

After documenting the intervention within its natural environment, I applied the P-Compass tool in consultation with the field researcher who assisted with the observational study and interview process to explore the dynamic of CP within each intervention. Rifkin’s Spidergram tool, which aims to assess the level of participatory action, was also applied (387). It is important to note that the aim of applying the P-Compass and Spidergram tools was not to critically evaluate the effectiveness of CP in achieving health outcomes or impact (although wherever possible such data were also recorded), but primarily to document the rationale for CP and the mechanism of the participatory approach.

The reporting template was developed to present the case-study:

In summary, adopting a focused ethnographic approach allowed for an immersive experience of documenting the varied perspectives of participants in the health intervention, and provided a layered contextual understanding of CP.

Page | 108

Document review method

Document review allows interpretation by the researcher to give voice and meaning to an assessment topic or research phenomenon (330). I examined documents in two categories, intending to investigate the extent to which CP and community participatory approaches have been defined, described and contextualized in health care systems in Sri Lanka (Research Objective 1). These were: 1. Domestic health policy and legal documents pertaining to health care in Sri Lanka 2. Documents relating to PHC worker role and functions, duty lists, training curricular and reporting templates.

I limited my document search to official government records, peer-reviewed papers and published grey literature. This included a rich diversity of documents such as MoH reports, official meeting minutes, technical reports, policy manuals, training and guidance handbooks, strategic plans, standard operating procedures, and health worker training curricula and teaching syllabi.

4.5.1 Analysis of health policy and legal documents

Policies offer a strategic guide for decision and actions, reflecting the values of society and of providers of human services. They demarcate who has power and how it is reshaped, and provide a platform for future trends (331). I examined the content of existing health policies and domestic legal frameworks to assess how CP is rationalized, described, and defined (if at all), and examine how participation is articulated within the policy goals and implementation frameworks.

Inclusion criteria: All MoH policies, including those that were published in the final draft stage prior to Cabinet approval up to December 2014, were included in the analysis. The policy documents were selected from searching MoH libraries and performing hand searches of publications from relevant directorates within the MoH. With assistance from staff of the Directorate of Policy and Planning, official documents and plans from within the health bureaucracy were extracted. Current political manifestos on health were also examined as

Page | 109

these serve as influential documents in guiding national health program priorities and resource allocation.

The domestic legal frameworks pertaining to health were identified through two electronic databases. ‘Law Net’ is a legal information portal maintained by the Sri Lankan Ministry of Justice (332). It is intended to be a tool for the legal community and researchers and provides up-to-date legal information about Sri Lanka's constantly evolving legal framework. The second was the Commonwealth and Common Law Database, published by the Commonwealth Legal Information Institute (2013)(333). The analysis of Sri Lanka’s legislative framework pertaining to health care necessitated an understanding of the nomenclature and content of legal documents (laws, acts, legislation and circulars). As part of this effort, I consulted two senior academics involved in researching health and human rights at the Faculty of Law, University of Colombo.

4.5.2 Analytical Framework

Two approaches to assessing CP in health policy and legal frameworks were utilized. The first was entitled ‘Best Practice Principles for Inclusion in a Participation Policy’ (334), and the second a policy analysis guide to strengthening relationships between communities and health systems in Australia (335). Analytic guides have been successfully used to explore policy content pertaining to CP (336,337). The graphical overview of key points for policy exploration are shown in Figure 20.

I summarized the key features of each guide into a simple checklist (presented in Annexure 6). An important aspect of this tool was that it elicited the health organisation's position in relation to CP in terms of strategic planning, service delivery and evaluation, and how the organization seeks to understand community by building and maintaining comprehensive knowledge about the local community.

Page | 110

Figure 20. Key elements in undertaking policy document analysis

Policy coherence is an important but often neglected aspect of policy analysis. Coherence is defined as the assessment of a policy that is internally consistent and attuned to objectives pursued within other health policy frameworks (338,339). For instance, one policy may emphasize CP as population based screening, but another may explicitly articulate CP through a social determinant perspective. After analysing the content of each policy in terms of how CP was defined, I mapped areas of convergence and divergence in concepts of CP and rationales for its use within a policy implementation framework. As part of the assessment of coherence, I chronologically plotted the emergence of each health policy in Sri Lanka since the Alma Ata Declaration in 1978, and cross-referenced their emergence with the major Sri Lankan health system reform phases and adaptations.

Page | 111

QUANTITATIVE METHODS

4.6.1 Survey rationale and method

Health policies are important governance instruments to guide health care programs and practices, and reflect the highest decision making processes that affect populations (340). Policy surveys of health decision makers have provided valuable data to inform research agendas, policy formulation, priority setting and advocacy (341-343). As per Research Objective 2, I intended to explore the extent to which communities have been engaged in health care planning and policy formulation in Sri Lanka, and the extent to which PHC workers have participated in such policy decision making.

4.6.2 Rationale, survey participants and sampling frame

Health policy making in Sri Lanka is a primary function of the MoH (344). Health policy making is usually performed in a collaborative process with consultation from a multiplicity of stakeholders (345). These mainly comprise subject experts on the specific policy issue or problem, and are sourced from within the MoH, academia, industry, civil society and experts from international organizations (345). Sri Lanka’s health policy-making community is therefore diverse, and ‘capturing’ those with previous history of health policy involvement, and who had played a role in drafting or driving a piece of health policy or legislation, constituted a methodological challenge. There is no registry within the MoH that systematically details all participants involved in past health policy processes. Encapsulating the true sampling frame of policy contributors is therefore difficult.

The study aimed to survey a diverse group of policymakers and elicited participation from individuals who have been involved or are currently responsible for health policy making in Sri Lanka. ‘Eligible responders’ were defined as any person that was directly involved in the process of health policy making and/or was involved in developing strategic frameworks for such health policies in Sri Lanka. A screening question reflecting this was added to the start of the survey instrument to clarify the eligibility criteria.

Page | 112

A descriptive cross-sectional self-administered survey was conducted with health policy makers using an e-survey platform. Online surveys provide an efficient way to collect data (346). I used an online survey strategy to disseminate a policy survey questionnaire to a wider pool of professionals to capture those who had been involved in the health policymaking process over the past three decades. Four different strategies were used to recruit survey participants:

First, the most recent staff list of all executive heads and directors of the MoH was obtained (n=20) and letters sent to their respective officers detailing the purpose of the policy survey and the mode of administration. Follow-up phone calls were made to confirm the receipt of letter and ensure survey completion. Face-to-face meetings were held with any directors that requested support with completing the online survey.

Second, a partnership with the Sri Lanka Medical Association (SLMA) was forged In order to maximize capture of those professionals involved in health policy making. The SLMA is the highest academic medical institution in Sri Lanka, bringing together medical practitioners of all grades and all branches of health care. Meetings were held with the President of the SLMA and relevant staff to seek permission to utilize its database of health professionals in the country (3,800 members as of August, 2012). The survey was thus disseminated to the entire SLMA member database.

Third, the Sri Lanka College of Community Physicians (SCCP) is another organisational body operating within Sri Lanka’s health sector that represents the nation’s public health leaders and practitioners. Community physicians work at the nexus of community health, preventative medicine and health promotion. Meetings were held with the president and secretary of the SCCP to secure collaboration on the research project, and access was provided to the member contact list database (354 members as of September 2012). The same screening question used in the SLMA survey was employed to determine eligibility.

Finally, lists of health policy advisors of international organisations such as WHO, the World Bank, and the IOM based in Sri Lanka were compiled (n=6) and the survey was disseminated.

Page | 113

4.6.3 Overview of survey instrument

A number of studies that surveyed key decision makers for priority setting in the health services were utilized in tailoring the design of the survey. These were also analysed for their content and structure (347-349). An 18-item survey questionnaire was developed under 7 components listed in Table 7.

Table 7. Overview of the seven components used in the health policy survey 1: Socio-demographic and professional profile (gender, years of professional service, current employment status) 2: Responders’ policy involvement profile (the name of policy processes engaged in, the degree of contribution to the policy development process and the stage of involvement in the policymaking cycle) 3: Stakeholder participation in policy making profile (involvement of communities, consumers, patients, civil society organisations, peak bodies, lobby groups and advocacy agencies) 4: Methods enabling community participation in policy making profile (methods used to enable community participation in policy process) 5: Community Participatory Compass (P-Compass) profile (overall extent of community participation in the health policy process) 6: Participation in Policy Development (PPD-Tool) (extent of participation from within the Ministry of Health staff/ units/ departments in developing and formulating the policy) 7: Perceptions profile (value and attitudes placed on involving communities/health consumers in health policy making)

Component 2 sought to identify the survey responders’ degree of involvement and contribution to the policy development process. This was important to determine, since higher levels of engagement in policy making cycle indicate a greater degree of knowledge of the stakeholders and policy process which are important aspects of survey questions in components 3 to 7. Component 4 explored how communities were made aware of the policy-making process and how community participation was enabled, for instance via

Page | 114

community representatives on policy advisory committees. Component 6 presents the Participation in Policy Development (PPD-Tool) which sought to assess the overall extent of participation by the various units of the health system in the policy-making process.

4.6.4 Pre-testing survey instrument

The survey was pre-tested: an interviewer administered the questionnaire to four senior policy makers from the fields of disability services, child health and migration health in Sri Lanka, and feedback was obtained. The first iteration of the e-survey instrument was sent to two medical officers attached to the SLMA and two public health specialists. The main feedback following the pilot survey dissemination focused on the need to improve clarity with respect to the tool presented in component 6, and the policy-cycle question (component 3). Following the pilot phase of testing, a graphical interface was added to improve the conceptual framework of the policy cycle and depict the PPD-Tool.

4.6.5 Administering the survey

An online survey model is one in which invited participants are provided a link via email, through which they can then access the survey instrument after fulfilling a step on eligibility criteria (350,351). The survey is based on a ‘dynamic’ webpage design architecture, which allows the responder to ‘click’ options or select using drop-down menus and graphics. In reviewing the literature of survey administration, a recurring lesson was the frustration responders felt during the email submission process (352). A dedicated e-survey platform avoids the need for responders to save the completed survey form as a word file and send as an email attachments– a process which takes time and lead to errors in submission. I used a web-based survey method, using proprietary software entitled FluidSurveysTM Ver4.0 (2012) to construct the survey into modules, and paid a subscription fee to FluidSurveys to ‘host’ the online survey on its server for a period of 4 months. The FluidSurveyTM software excludes multiple entries from the same individual, and from a security perspective avoids lists being passed on to unintended recipients of the survey.

Page | 115

4.6.6 ‘Participation in Policy Development’ (PPD-Tool) used in e-survey

The PPD-Tool was designed with the aim of describing the participatory dynamic between the various levels within Sri Lanka’s health system (e.g. engagement of PHC units with district and national level units) in planning health policy development. The existing organisational structures within Sri Lanka’s health system were mapped and then labelled to reflect the respective levels/tiers (see Figure 21). The PPD-Tool used ‘static definitions’ of categories, where the survey responders had to select a statement category that best reflected the type of participatory engagement between health system structures in the policy development process. A series of statements were constructed to elicit a participation typology. For instance the statement: “The policy process was developed and formulated by central level authorities with the input of selected technical experts” yielded a participatory category: ‘Type B’. This meant that the stipulated health policy was developed mainly through the involvement of central level authorities and technical experts, and with no participation from other parts of the health system (defined in Figure 21).

Page | 116

Figure 21. Tool for assessing extent of participatory engagement within Sri Lanka’s health system (‘PPD-Tool’)

Page | 117

4.7 Analytical process adopted

4.7.1 Quantitative data analysis

Quantitative data emerging from the survey of policy makers were analysed in accordance to the good practice guide described by Kelly (2003) (353). Survey responses typically reflected qualitative differences in the construct being measured and were nominal in nature, with the exception of one question that utilized a Likert-scale (ordinal). Data were entered into a Microsoft Excel database and statistical tools embedded within the software were used to determined frequency distribution and means. Results were presented as one-way tables.

4.7.2 Qualitative data analysis

The qualitative data analytic process adopted was thematic analysis (TA), which drew upon principles of grounded theory (GT). TA is designated as the “foundational method for qualitative analysis” (354) and is used for identifying, analysing, and reporting patterns (themes) within qualitative data (355). According to Strauss (1990), classical GT analysis is described not so much as a specific method, but more as “a style of doing qualitative analysis” (356). GT analysis shares the same set of procedures for coding data similar to inductive TA, however its distinguishing feature is that it is always directed towards new theory formation, without reliance on any pre-constructed model (354).

I drew upon principles used in GT in order develop a substantive theory (a complex health systems model) around the determinants and dynamics that influence CP practices in PHC (Research Objective 4). It cannot be considered as a ‘true’ GT approach, since I utilized a pre-existing health systems model to anchor the overall research objectives (termed the ‘progenitor’ model described on page 72). The thematic codes that were constructed after an inductive process (of data immersion followed by constant comparison), were then superimposed on the progenitor model, reshaping it to form a new theoretical framework (detailed below). This is why this approach is considered an applied GT analysis. Braun

Page | 118

(2006) suggested that researchers need not subscribe to the implicit theoretical commitments of grounded theory if they do not wish to produce a fully worked-up GT analysis (357).

To summarize, TA using an inductive approach (drawing upon principles of GT) was the primary analytic strategy I adopted to analyse qualitative data that emerged from mixed method approaches. By harnessing the GT method, I aimed at distilling a set of determinants and dynamics that influence CP in Sri Lanka’s PHC system. The data harnessed from document analysis and the structured observational study were analysed through content analysis, a deductive form of TA.

4.7.3 Main steps in the data analytic process

Qualitative research analysis was conducted in an iterative fashion, with data collection, analysis and interpretation taking place within the one iterative process of interpretation (358). Guidelines for analysing qualitative research by Gilson (2011) (359), Carter(2009) (360) and Malterud (2001) (361) were followed in an effort to sustain a uniform approach to developing and discussing findings. Described in brief are the key steps in this inductive process of synthesizing and constructing the resulting themes, which were presented in the final results section of my thesis:

Step 1- Data immersion: As described in the previous section, an iterative process was utilized to cross check the written transcripts of audio recordings for accuracy of translation. The transcribed text was then independently read by a research assistant and lead researcher. Each interview transcript was read over again to identify emergent themes, with parts of text that have meaning to the study objectives also being recorded (as is consistent with a TA approach). This initial or pre-coding process may be considered ‘immersion in the data’.

Step 2 - Coding process: The first step of coding for me involved going over the text line by line, examining what was interesting in that chunk of data and assigning a label to this chunk, a process that is called open coding. For instance, as part of the open coding process some of my initial codes (with memos) in analysing the transcript data from group interviews discussing CP in health included: “CP definition - within medical model (population screening)”; Page | 119

“CP definition - as means of  service coverage (technocratic approach)”, “CP inhibitor - no support from supervisors to push CP agenda”; “CP inhibitor - job duty lists doesn’t include CP”; “Value on CP +ve”.

I then looked for patterns in these codes, moving onto axial coding (362). Here open codes are grouped into categories and subcategories, with some open codes becoming categories in their own right: for example, “Factors enabling CP”, “Perceived role of CP in NCD”, “Expectation by workers (of community)”, and “Barriers for implementation”.

In this process of moving from open to axial coding I followed Glaser’s (1992) constant comparison process (362) of continuously comparing within each code and between codes. As themes are organised under higher-level constructs, I asked rhetorically: “what categories could subsume all the lower level codes?” Writing memos on the possible relationships between codes allowed me to clarify my rationale for the nomenclature and revisit them later if I needed to re-code. Examples of such higher order categories developed were: “Administrative and stewardship functions”, “Resource determinants”, “Attitudinal factors” and “Technical skills and capacities”.

Some of the codes did not always ‘fit’ neatly into broad categories. I would often refer back to the original research question and ask: “What does this say about community participation in PHC?” This helped construct and anchor themes to the overall dynamic of health systems research. Not all such themes were so easily defined, with some having complex and unexpected associations: for instance, a finding that a lack of meaningful engagement between health professionals themselves (a diminished participatory culture within the health system) may contribute to the lack of CP approaches.

I assigned open-codes through line-by-line analysis and generated axial codes by using the QDA MinerTM qualitative research software platform (363). Although useful, I felt that one of the disadvantages of using qualitative research software is that it almost stimulates a ‘coding mode’ each time a transcript was opened. I felt at times a flurry of activity in coding, sorting and categorizing, leaving little for reflection! The data visualization tools in QDA MinerTM were also useful in making connections between themes, open-codes, axial coding and selective codes.

Page | 120

Step 3 - Hermeneutical process: A hermeneutical framework attempts to combine all or parts of the codified puzzle to one theoretical canvass (364). It forms the final analytic method, which integrates categories and explores their associations in the form of a single theoretical framework. I utilized the basic theoretical scaffolding of the complex health systems model presented earlier (the ‘progenitor’ model) (page 70) to build and ‘unpack’ the health system factors and determinants that affected and influenced the meaningful realization of CP practice at PHC level. I started by first anchoring the categories and high order axial codes that were associated with the broad environments associated with progenitor model (the ‘policy and structural’ environment and ‘practice’ environment). I then went through and added layer by layer the codes to the progenitor framework. If there was little association within the broad environmental domains, then a new domain was created, and codes linking these layered into the model. Though time consuming this process was surprisingly easy to perform. I used a freely downloadable open-source mind mapping and data visualisation software called CoggleTM (365) to assist me in this process. Uploading the broad themes and categories to CoggleTM enabled me to move around the constellation of factors and elements for best fit.

Page | 121

4.8 Strategies ensuring research rigour

In health research, the question is no longer whether qualitative methods are valuable but how rigour can be ensured or enhanced (366). Ensuring study results are a true reflection and honest account of the perspective of research participants is an imperative part of research quality and credibility (367). Although the influence of the researcher’s own preconceptions, beliefs and experience on the interpretation of qualitative data is accepted, and even regarded as necessary (366), such influence needs to be critically evaluated and recognised (368). Methodological checklists (369-371) and guides (367,372) play an important role in this process and suggest strategies such as triangulation, reflexivity and member checking to ensure the credibility of qualitative research.

I adopted Lincoln and Guba's Evaluative Criteria (367), which posit the trustworthiness of research to reflect credibility, transferability, dependability and confirmability.

4.8.1. Credibility

Credibility refers to confidence in the 'truth' of the findings. Techniques used for establishing credibility in my qualitative data included prolonged engagement, persistent observation, triangulation, debriefing and negative case analysis.

Prolonged engagement and persistent observation. The purpose of ‘prolonged engagement’ with participants and research settings is to ensure the capture of a diversity of views, influences and contextual factors that impinge upon the phenomenon of CP, whilst ‘persistent observation’ affords more opportunity to unpack the different strands and details of such views. As Lincoln and Guba state, “prolonged engagement provides scope, persistent observation provides depth" (367). The field research was undertaken over a two and a half year period across PHC settings in three districts. Health workers across six PHC units were engaged not only through interview processes, but also through observational follow up and in-depth case studies. This final component alone took between 4 to 6.5 months to complete within in each district setting. I was ‘embedded’ into the PHC system in each selected district during this time. I gained close familiarity with PHC workers and their practices through intensive involvement in their cultural environment over this period (Page 104). Indeed, a Page | 122

great strength of the focused ethnographic methods I adopted was that they cultivated close ties with health workers by documenting practices in real world settings. The intensive and prolonged observational periods were particularly useful in providing a rich understanding of worker perceptions and practices (323).

My research strategy also involved developing partnerships with central and district health administrators over a period of at least a year, and, in the settings where community interventions were profiled, up to one and a half years. A total of 12 PHC workers across 3 districts participated in a month-long structured observational study documenting community health interventions. I used a diversity of methods such as digital video-graphic recording of NCD prevention interventions utilizing CP approaches for a 6-month follow-up period (pp 109).

Observer Effect. Focused ethnography allows for an immersive experience of documenting the varied perspectives of a health intervention and provides layered contextual understanding of CP. However, a major criticism is that ‘observer effects’ will somehow bias research findings (373), though some scholars argue against this premise (374). Methodological papers suggest that observer effects may be minimized if an investigator can gain close co-operation and positive rapport during a longer and more intensive period of observation (374).

Sustaining an ‘embedded’ field-level presence allowed me to identify and rigorously document interventions that utilized CP approaches at primary care level. I felt that prolonged engagement with health professionals from central to village levels gradually developed into a genuine rapport, facilitated trust and understanding, and may have buffered the observer effect to some degree. As indicated in section 1.6 on reflexivity, there was one instance of resistance shown by a particular MOH officer. This was, however, attributed to issues of miscommunication than any breach of trust.

Whilst labour intensive, such efforts to gain familiarity and a positive rapport with policy makers, PHC workers and MOH units ultimately provided me with a rich data set from which to analyse CP interventions targeting NCD.

Page | 123

Triangulation techniques also strengthen research credibility (292) (298). Several kinds of triangulation were formally employed in this study: the first was across multiple sources of data (health professionals at central, district, village and PHC administrative levels, and those involved in policy formulation as well as direct service delivery); and the second was across a multiplicity of methods (from surveys, interviews, focused ethnography, descriptive analysis of work plans, policy documents). The process of corroborating research evidence from multiple methods and using different sources rather than a single method facilitates deeper understandings of meaning (372). Triangulation also enabled the layering of evidence and factors that influence CP to be progressively developed, elaborated and confirmed from findings from multiple research methods I adopted in my study.

Debriefing is a process of open reflection on observed phenomena with professional peers and/or researchers for the purposes of “exploring aspects of the inquiry that may otherwise remain only implicit within the inquirer's mind" (375). Beyond personal reflexivity, this can help uncover biases, perspectives and assumptions, and increase awareness of the individual’s position toward data and analysis.

Two methods of debriefing were used, formal and informal. Formal or structured efforts included organising a consultative workshop with participation from the director and technical staff at the MoH policy and planning unit; feedback sessions held following the presentation of preliminary findings at a national level workshop on revitalizing PHC in Sri Lanka6; and a scientific conference organised by College of Community Physicians7.

Informal debriefs were embedded within the observational study process. Most often, at the end of the ‘research day’ I would discuss and exchange notes with my field researcher, reflecting on the key lessons of the observed PHC intervention (recorded using EvernoteTM software on a smartphone). Also termed ‘investigator triangulation’ (376), critical discussions were also undertaken during the long car and train journeys from rural settings to ‘base’ in the capital. Editing and producing the two short documentaries on the NCD intervention case-studies also ‘forced’ me to critically engage with and debate the emerging thematic

6 Workshop: “Primary Care Model for Sri Lanka”, Palm Garden Village Hotel, Anuradhapura District, on 18- 19th March, 2011 7 Annual Scientific Sessions, College of Community Physicians, 2013 on Conference theme: “Defining Universal Health Access and Coverage in Sri Lanka”, 21-25th September 2013 in Colombo, Sri Lanka Page | 124

content with the video-production team. The short-film making was itself a methodological challenging process, where in distilling video graphic content to capture key themes I found myself revisiting assumptions and ideas.

A field research assistant played a key role in the interview as an assistive prompt (‘my brains trust’ was the colloquial term I often used to describe the field research assistant). Their role within the interview process was to ensure that I, as the principle interviewer, covered the scope of questions outlined in the interview guide. There were occasions where the assistive prompt would indicate that I should explore further a particular concept or theme that emerged during the interview session if they felt that needed further clarification or validation, or which may have been missed. Having a prompt certainly ensured rigor and quality in gathering data.

During the often long travel periods from one district setting to another, I was able to discuss with research assistants the content of the interviews and clarify what they felt were the key emergent themes. This inspired new avenues for analysis: one such example is questions about the culture of community volunteerism and how health institutions enable volunteerism, which emerged from such discussions. I used EvernoteTM software on a digital diary to document these insights and ‘musings’ throughout the research process. The iterative nature of the qualitative research process allows for questions that are not effective at eliciting the necessary information to be dropped and new ones added to the research guide as investigators learn more about the subject (308).

Negative case analysis involves searching for and capturing data that appear to contradict majority patterns or views that emerge from data analysis (377). As demonstrated in the literature review, CP in health is a complex phenomenon whose definitions are highly contested. From the outset, I realized that attention needed to be paid to harnessing and presenting ‘dissenting voices’: for example, when presenting perceptions of CP, presenting voices both in support for and in opposition to participatory practice.

Page | 125

4.8.2 Transferability

In adopting an ecological perspective for health systems analysis, an important goal of my research was to explore (and provide the reader) layer upon layer of contextual, attitudinal and historical data in exploring CP at PHC level. Rather than merely present only a single dimension of data, such as the attitudes and perceptions of PHC workers relating to the phenomenon of participation, I sought to capture the realities of their routine practice, how such practices may be influenced by regularized work-flows and daily reporting requirements, the effects (if any) of worker training curricula, or – at the highest level – the regulatory frameworks and policy formulary pertaining to PHC service delivery.

Rather than only documenting the implementation mechanics of those NCD interventions that utilized CP approaches, I aimed at presenting detailed contextual information on the setting in which the intervention occurred, the profile of the district health system in which it took place, and any historical or stewardship determinants which catalyzed its emergence. These layers are also important to ensure generalizability to other health system settings. The emphasis was therefore on building ‘thick descriptions’ of data (377), and on presenting these in a systematic way to allow a greater degree of transferability, where readers can also make their own interpretation and judgments.

4.8.3 Confirmability

Creating a detailed record of the research process (‘audit trail’) helps facilitate replication of the research process by other researchers – an important principle in scientific research (Mays and Pope, 1995, 2000). Lincoln and Guba (371) posit that audit trails are also a key part of trustworthiness. In the current study, records of observations, case note reflections and interpretations that were made throughout the data collection and analysis process were noted systematically using EvernoteTM software platform based on laptop and smartphone. At final analysis, my EvernoteTM research databank contained a total of 1,146 documents (e.g. case notes, research team reflections, interview transcripts, PHC work schedules, and scientific papers), 46 audio file (mp3) recordings, and digital images taken of the research process. In addition, 33.5 hours of high-definition video (raw footage) were captured of the NCD interventions catalysed by PHC workers. Dependability and Page | 126

confirmability of the data could also be demonstrated by running queries within QMinerTM. The software was able to locate passages that matched the criteria set in a query. The query process ensured that any issue described within the findings was not the perception of just one participant, but rather was confirmed by a number of participants of the same opinion.

Since I captured the case-study interventions video-graphically, I was able to re-play the digitally recorded video and assess it for interactions and elements that may have been missed if only photos or written notes on the activity were taken. By using multiple formats in documenting the event I was able to cross-check and verify information (methodological triangulation).

4.8.4 Dependability is often compared to the concept of reliability in quantitative research (378). It may be addressed through the way in which data were recorded, transcribed, and then analysed for coding (379). I support MacLean’s (2004) (309) indication that the process of transcription “is not merely verbatim copying from recording devices and ensuring correct semantics”, but rather capturing the intent of the participant and accurately describing ideas of what they said.

Two approaches for translating and transcribing data from audio-recorded interviews is shown in Figure 22. The first pathway indicates the ‘traditional’ linear transcription process. One problem affecting this linear process is how misinterpretations of language and meaning can be carried through to ‘downstream’ processes. This in turn affects content validity. Numerous methodological papers and guides have discussed the problems of relying exclusively on linear forms of verbatim transcription and back-translation, and have recommended the use of iterative approaches (380,381). The reliability of translation and transcription of participant response data was enhanced through a two-person iterative transcription process described in Figure 22 below.

Page | 127

Figure 22. The collaborative process adopted for data transcription

In the iterative process I adopted, the translator becomes a collaborator in the research process, strengthening the rigor of language-based inquiry. Larkin et al. (2007), in their paper on improving transcribing and translation in qualitative health research, suggest methods to ensure that “mutual reciprocity between researcher and translator offers greater possibility for construction of nuance and meaning” (382).

This iterative process also ensured that meanings conveyed by participants are adequately represented, including checking nuances of conversation and participant’s voice (such as tone), to allow for deeper contextual analysis. Cases of ambiguity in meaning, content or other inconsistencies were mitigated using this collaborative process. Joint consultation was undertaken for sections where there was ambiguity of meaning or inconsistencies. While it may appear protracted and time-consuming in implementing additional steps of involving two reviewers, it removed much of my anxiety in ensuring a rigorous capture of participant views from original interview recordings.

Page | 128

4.8.5 Reflexivity

In Chapter 1, Section 1.6, I described in detail my personal stance as a researcher and reflexivity within the research process. In qualitative research, the researcher is considered part of the research instrument. The credibility of a study rests not only on the procedures implemented but also the self-awareness of the researcher throughout the research process (383).

Maintaining a research diary (via EvernoteTM) was also an important tool in my process of ‘research introspection’. In an excerpt extracted from the Evernote e-diary, I describe surprise concerning the willingness of PHC workers towards exploring the topic of CP and NCDs during the formative phases: ‘There is always a fear they [PHC workers] will not be interested. Today proved that is just not the case. They wanted to discuss CP and NCD … at one point it was like catharsis where they just wanted to ‘vent’!!... it was as if they hadn’t been given the platform to do so. Challenge moving forward is how to harness this interest. Getting those documentaries done will be key’ [10.6.0 Evernote:12th March,12:8.01pm].

Page | 129

Chapter 5:

Study Findings

Image 5. A young mother discussing her involvement in a community based health NCD prevention intervention

Page | 130

5.0 Chapter overview

This chapter describes the key findings of the study. Research findings from qualitative research methods are presented first, followed by the results of the health policy maker survey. The themes generated through interviews are presented first, followed by the results of the observational and focused ethnographic study component (See Figure 23 below). As described in the methods section, this component is comprised of four parts (A to D), and findings from each part are presented discretely. Findings synthesised from the critical review of domestic health policies, legal frameworks, PHC duty lists and other relevant documents comprise the final qualitative results component. A summary of the key findings is presented at end of each methodological section.

Figure 23. Graphical overview of Chapter 5

Page | 131

5.1 Findings of key informant and group interviews

5.1.1 Participant characteristics

Group interviews. A total of 31 Public Health Midwives (PHMs) and 18 Public Health Inspectors (PHIs) participated in the six group interviews conducted across three districts8. The average number of years working as a health professional at the Ministry of Health (MoH) for each worker category was 10.8 years for PHMs; 9.5 years for PHIs, and 4.6 years for Medical Officers of Health (MOH). Only 8 of the 49 health workers who participated in the interviews were natives of the districts in which they worked.

The majority of PHMs (90%), PHIs (85%) and Medical Officers of Health (70%) were of Sinhalese ethnicity, however 98% were able to converse in the Tamil language9. The mean age of PHMs was 41 years (range 27-55), with PHIs being slightly younger at 38 years (27 – 51). MOHs were generally young medical officers with a mean age of 34 years (range 30-41).

Key informant interviews. A total of 33 in-depth key informant interviews were conducted: 6 with national level Directors in the MoH, 19 with district level MoH leaders and 8 with international agencies, academia and professional associations. A precise breakdown of their position and titles has been omitted to ensure confidentiality.

The themes (and subthemes) generated through qualitative data analysis are presented below. The quotations selected for presentation under each theme or sub-theme were based on the following criteria: the quotations were illustrative of a particular theme; quotations encapsulate a range of views where heterogeneity of views is present; and they are focused and succinct. Verbatim quotations from participants are presented as indented paragraphs in italics, within double quotation marks and with the pseudonym of the participant provided in parentheses. Within a paragraph, participants’ words are cited

8 It was important to have a meeting place that was neutral and allowed people to talk freely and openly about their opinions. All group interviews were held in the staff conference room at each MOH office with the exception of Thamankadwa MOH. At this location a community centre hall was utilized since the MOH office offered no private space for discussion. 9MoH policy stipulates that all health workers be proficient in both Sinhala and Tamil languages to a standard where there are able to communicate with patients/community members. Passing the language exam is also linked to higher remuneration and promotion. Page | 132

using double quotation marks “ “, and author words/reflections cited within single quotation marks ‘ ‘. In order to ensure participant confidentiality as stipulated in research ethics, data linking PHC workers to MOH location were removed, and only district and worker category (e.g. “PHI”) are provided.

5.1.2 Definition of community

The dominant view of ‘community’ shared by both PHC workers and administrators is of communities as beneficiaries of the health system under a health service catchment area. Rather than a community development or community capacity building model, this heuristic definition implies health professionals viewed community as ‘consumers’ having specific health needs, and PHC services as the ‘conduits’ to deliver services to users ‘at village level’. This is well reflected in the following quotes from PHC workers and health administrators.

“It’s obvious! Community for us is a group of people having some sort of health needs... some sort of problems that come under our health service area” [PHM,5N].

“I define community as the area my area…this is the practical view, yeah, this is the practical one we have for health services…community catchment area” [SMO, MOH1]

“People who receive our services is our community, they are the ones who benefit from our services” [PHM,4K]

In relation to community health interventions, geographical location seemed to be a practical consideration in defining a community: “Of course anyone will understand community can be as large as the entire district population or even the province! The main thing here is that we cannot use such definitions. It is just not practical. So that’s why community is divided to the district level, which comes under purview of the Regional Director of Health Services or as small as a village cluster communing under purview of the health ministry. So yeah, community is a ministry demarcation” [MO-HI, 1MOH].

The above definition of a community resonates with McLeroy’s (96) ‘Community as setting’ and Whitehead’s (97) ‘type VI and VII’ categories of community-based health interventions. The broader characterization of community beyond geographical or health service boundaries was articulated by only a few respondents whose definition included “students and teachers in the schools” or “the number of people who live in the area”.

Page | 133

While acknowledging their role as a government civil servant, some also recognized themselves as being “part of the very community we serve”. They saw their role as a ‘health leader’, someone “who communities can turn to” and “who can guide the community”.

5.1.3 Definition of community participation

Although community was defined predominantly through a programmatic or technocratic lens, as a means of achieving health goals, for the majority of PHC workers CP was about communities “working together with us to find solutions”. This reflects the notion of CP as a means of working ‘with’ (rather than ‘for’) communities: a. “Participation is not just about communities participating in health programs we run… It means seeing how much involvement we can have with people in preventing certain diseases affecting their lives…now this is a much more difficult thing to do”[MO-NCD,P].

b. “ (14) is when people help us in doing our health programs, in any way possible...it’s when parents come and want to get involved in my school medical inspection program” [PHI,5B]

c. “When our community helps us to find what their health problems are…you know, come to me and say ‘sir these are the issues, we need help with’ that’s participation for me. You see, that is when we can all work together to find solution for their health problems. I always feel that’s the best way” [PHI,2N]

d. “Community participation is not about us going to the community and doing things because you know, we know what is best, no, it is I feel when communities make demands to us, and request us to do things for their care, and we sort of help give them the opportunity to make such demand…” [PHI,1T]

e. “When communities participate in our programs, get involved with what we do, it has great value since it provide motivation for us to work harder, they look up to us, respect us, so it motivates us further” [PHM,3T]

f. “There are lots of benefits in CP, one is we can identify health problems in the community much easily with the support of the community” [PHM,2N]

The above quotes reflect the high value health workers place on working ‘with’ communities, as emphasized in the Alma Ata Declaration on PHC and Global Action Plan for the Prevention and Control of NCDs (24). Moreover, statements such as “we can work together to find a solution”, “when communities make demands on us” and “how much involvement we can have with people” strongly suggest that CP is seen as a means of driving ‘consumer demand’ and providing a platform for community advocacy. Such perceptions incorporate a view of

Page | 134

participatory action in problem identification and program implementation as an integral part of CP.

In reflecting on the majority of participant responses it is apparent that the value and importance assigned to community participatory approaches can be situated within the ‘managerialist’ approach to planning health service delivery. That is, CP is seen as a means of enhancing health interventions. These ‘interventional benefits’ may be broadly summarised under three categories (referenced in quotes above): enhanced problem identification (c,f); tailored interventions (c); and ensuring the sustainability of interventions (d,e).

A minority of respondents believed that iterative engagement with communities would bring benefits such as enhanced technical knowledge and skill enrichment. As one PHM stated, “We can all learn from communities more when there is constant interaction with community… I don’t have all the answers... So yes, this information flow is an advantage. This is how I think community participation helps us as health professionals”.

Some PHC workers cited motivations for participation as stemming from their sense of civic duty rather than any technocratic goals espoused by public health programs: “people of the community know us [PHC workers] by name, they trust us and frequently speak to us. That relationship is a factor which motivates us to engage. It is our duty to engage. It is our duty to get them [community] to participate in what we do, what decisions we make. Not always does it work this way, but this is the way” [PHI,3N]. A small number expressed the need for community engagement in order to maintain a historical tradition of commitment by the health service to citizenry, or as one Director stated, “CP is a core principal that Sri Lanka has a proud tradition of…we must maintain that”.

5.1.4 ‘Rights based’ approach to CP

Some respondents recognised that participation is a “basic right” and that CP in health underpins the “rights-based” approach. Despite this recognition, the approach is not integrated into routine health service delivery:

Page | 135

“I mean really from the beginning to the end of the intervention project, participation is about rights…their health rights. The thing is our [health] system doesn’t focus on health as a community right …although they may say it” [CCP, MOH2]

Participants also indicated that there must also be accountability and “shared responsibility” on the part of communities. As a PHI stated: “Yes, we must lead, utilize and mediate communities for better health. But there are limits, health is a shared responsibility…and we are limited by boundaries of our duty lists, time, funds…”. In pressing these respondents on what they meant by a ‘shared responsibility’ for ensuring a rights-based approach, they stated that while they “are always there to support communities”, communities “need to also build a culture of health demand”. This suggests the need for communities to play a role as health advocates and stakeholders in PHC service delivery.

5.1.5. The modes and means of CP

Whilst most PHC workers described the importance of effectively engaging with communities in disease prevention and health promotion programs, there were differences between PHIs and PHMs. In general, PHIs viewed CP through the mode of social mobilization processes, and considered this mode of participatory action as a useful means of problem identification or direct program implementation. “You see, when you’re looking at a problem like rabies control, it was important for me to involve community groups, not all, but some, to spread the message on the issue of stray dog populations in our area… I was quickly able to work with these community teams to undertake control measures…these included veterinarian groups and religious [Buddhist] youth teams. Not only did the community members identify the ‘hotspots’ but also helped tackle the practical issue of catching female dogs for sterilization!” (PHI 2T)

The preferred mode of community engagement for PHIs was through pre-existing civil society conduits and established representative structures within communities (such as religious social service groups and co-operatives). They also noted that “participation happens automatically” through family level connections established as part of existing health promotion programs. “We often link up through Graameya-Sewa Sahana-Kamitu [social service societies] and Buddhist Seva-sangams [religious based social network] when we think about engaging

Page | 136

communities. I can get their support to organize a campaign and mobilize the youth groups to help.” (PHI 3K).

5.1.6 Role of communities in NCD prevention

Involving communities in the fight against the growing burden of NCDs was considered by the majority of respondents to be ‘crucial’ in “implementing NCD prevention health programs effectively”. While most recognized the importance of CP, they acknowledged that meaningful community engagement within routine PHC practice is limited and “at the moment very low”.

Participants emphasised the need for “shared responsibility” and “leadership from within the community” if CP in health is to be manifested and sustained within the health system. Workers suggested that communities need to be involved in advocacy and lobbying efforts for “better health services” and “demand actions from health authorities to solve health problems”: “Communities need to get more involved by mobilizing existing community structures, such as village development societies, to demand programs from us…” [MOIC-RH,H].

There was, however, a recognition that inculcating a consumer ‘demand culture’ will not work for all chronic diseases. Within the field of mental health, for instance, due to stigma and poor service provision at primary care level, there was acknowledgement that the onus was on PHC workers to take the initiative on community mental health: “there needs to be [a] major shift to reduce stigma and change the way we [health professionals] engage on mental health with public, otherwise I don’t think there would be any demand for services from the people…we as health professionals and with police, teachers religious leaders need to do this …” [CCP, MOH 1].

In considering community leadership, some PHC workers noted the need to “be careful in engaging communities” due to what they perceived as the politics and power dynamics that surround community development initiatives. These are further explored under relevant sub- sections concerning community ‘gate-keepers’ (5.2.17).

Page | 137

A number of interviewees conceptualised CP in NCD prevention within a restrictive framework. Participation was seen as the “passive involvement” of people in health interventions. Such respondents felt CP to be an obligation of the community rather than an approach to be fostered by the health system. From this perspective, there appeared to be little scope, desire or need for iterative models of community engagement: “It’s simple. The role of the community in NCD programming basically is to participate in the screening programs.” [RDHS,P]. “As I see it community should and is obligated to participate in programs conducted by us [the health sector], this is the role I expect from the community” [MO-P,N]

5.1.7 Role of PHC workers in NCD prevention

PHC workers were considered to have an important role to play in NCD prevention at village level: however, respondents differed as to what this role was. Most health administrators considered PHC workers as facilitators for changing community attitudes and behaviours: “The role [of the] PHC worker in NCD prevention should be to trigger behaviour change” [HEO,N]. One senior administrator noted that workers should be actively “writing the script in the minds of community members at any given interaction with communities…large or small”. In contrast, others suggested that workers adhere to more biomedically driven population screening approaches for NCD prevention: “Clearly the role of the PHC workers in NCD prevention should be to increase screening clinic attendance rates from communities in their catchment area…they can be used to capture more and more people to attend the screening days” [RDHS,P].

Whilst explicitly recognising the value of behavioural change interventions, some respondents suggested that PHC workers promoting community participation in village level NCD screening clinics offered a more pragmatic, viable and “community accepted” approach to NCD prevention. They perceived communities as being more receptive to medically oriented interventions than health promoting ones: “…people like it [screening clinics] because they get the immediate feeling of a result, but we need to work on the prevention side. That’s the part that is missing” [PHM,1B].

Page | 138

However, some recognised that an emphasis on screening alone was insufficient if primary care workers were to meaningfully address chronic disease: “If we want to make this screening work, then we need to do more to ensure there is follow up of the people after screening…but this is not what’s happening” [PHI,3T]. The need to restructure existing work plans and duty-lists to allow for a greater degree of community engagement within routine practice were also highlighted: “If there is no major reform of the job lists, NCDs won’t be addressed at primary level. Simple as that. Even trying to roll out one-off training programs on NCDs, or starting pilot projects just won’t do…” [CCP, MOH].

5.2 Factors inhibiting community participatory approaches at the PHC level

Factors inhibiting community participatory approaches at the PHC level were divided into two broad categories: those that were related to and arose from the health system, and those that were determined by communities. Figure 24 below lists the higher order categories and sub-categories elaborated in this section.

Figure 24. Overview of factors inhibiting CP at the PHC level

Page | 139

Health system factors

The health system factors inhibiting CP are grouped under 3 sub-categories.

A. Operational barriers and technical capacities relate to those parts of the health system linked to such elements as program implementation, service delivery and capacity building of health care workers. B. Professional ethos and cultural barriers describe the influence of professional peers, workplace cultures and how decision-making power is shaped. C. Structural barriers relate to physical resources within a health system such as human resources for health, and how health services coverage are demarcated at community level.

A. Operational and technical capacities

5.2.1 Existing duty lists that do not emphasise CP approaches

PHC workers indicated that their current duty lists and job function schedules does not emphasise the use of CP approaches. As one PHI explained, “participation is not written into our job lists”. No clear guidelines existed on how they were expected to engage with communities, for instance in undertaking NCD prevention interventions.

Health administrators articulated that existing job functions needed to better reflect and be more relevant to current health priorities, disease patterns and health-seeking behaviours of local communities. For instance, current PHM duty lists stipulate supporting childbirth at household level as a ‘core activity’, despite almost all deliveries occurring in health institutions: “Home deliveries are still on their [PHM] job menus...it’s curriculum change that is needed. A new skill set is needed for the modern PHM in Sri Lanka. Counselling skills are vital.” [MOH,N1].

Structural reform of PHC worker roles and revitalizing their job functions were considered essential in addressing the growing burden of NCDs:

Page | 140

“We should also be encouraging self-care practices for those living with chronic diseases through our PHC staff...this has to happen now. Self-care, community care is a very important part of community participation for managing chronic illnesses” [MOH,H2]

5.2.2 Enhancing skills and knowledge in NCD prevention and management

Despite the commitment PHC workers demonstrated towards community participatory efforts for NCD prevention, they also noted a lack of formal training and expertise on the processors and strategies to enable such CP: “We haven’t received any formal training on how to look at diabetes prevention in communities. The chief nurse and medical officer made one or two awareness presentations last year, but that’s about the disease, not about how to work with communities on diabetes prevention” [PHM 1T]. According to one PHM, PHC workers are ‘well placed’ to address NCD risk factors at household and community level but “don’t have a clear idea of what to do technically”. Their current role was articulated by another PHM as to “simply encourage people to go get tested”. Beyond clinical knowledge, the ability to “effectively communicate messages on NCDs to general public and patients” was also highlighted by senior administrators as a key gap.

PHC worker training programs and teaching curricula (at both pre-service and in-service levels) place little emphasis on NCD prevention and relevant health promotion strategies. As indicated by the following quotes, the resulting technical competency gap inhibits the adoption of CP approaches to combat NCDs: “Now, if chronic diseases are the biggest disease threat for us, then don’t you think the curriculum of our medical staff should reflect this? It doesn’t!” [MOH,H2]

“Our PHI, PHM and MOH training curriculum doesn’t emphasize health promotion and community based strategies on dealing with cardiovascular disease and diabetes and so on.” [CCP, MOH]

During interviews, PHC workers identified a number of areas that may be useful in developing their skills and competencies with respect to NCD prevention and control. These areas included ‘risk communication’ and ‘counselling skills in lifestyle modification’. Due to word limitations these are not presented here, but have been published as a separate report and presented to the MoH as part of the research contribution to health systems strengthening.

Page | 141

5.2.3 Ad-hoc efforts in capacity building and adopting innovations

One-off in-service training programs aimed at enhancing PHC worker skills, often led by external organizations, were characterised by most PHC workers as being highly ineffective in being adopted into routine practice. A district health authority leader reinforced the importance of innovative and systematic training, given the need to empower workers to develop the skills and capacities to utilise CP approaches: “Over the years I have seen so many training programs, including on NCDs, from various organizations like WHO… They all have valid theories…so we host such programs…we sit, we listen and we are asked enthusiastically to implement…But as I said earlier, you cannot write the script in the minds of workers like that, it just won’t work” [MO-P,P].

In reflecting on this theme that crystallised during analysis, I constructed an analogy called the “filter and funnel effect” to describe attempts made by central level MoH program managers, donors or development agencies seeking to introduce new innovations, skills and approaches in primary care practice through ‘one-off’ in-service training programs. Such training interventions are conducted in an ad-hoc manner, without a sustainable mechanism for follow-up and monitoring. As indicated by the above responses, PHC workers may actively adapt or ‘filter out’ the expected work practices being promoted through such training efforts due to the lack of supportive management structures at MOH level, limitations on time, financial resources, and incentives, or the absence of monitoring and reporting requirements (Figure 25). Diffusion of innovations theory (384) suggests that a minority of PHC workers (the ‘early adopters’) synthesise such training objectives and navigate constraints to effectively implement community-based NCD interventions. The PHMs profiled in the case-study analysis (Section 5.7) exemplified such early adopters that implemented innovative and sustainable NCD interventions within routine practice.

Page | 142

Figure 25. The ‘filter and funnel effect’. Selective adaptation by PHC workers on innovations and practices introduced via one-off, ad-hoc training programs.

5.2.4 Monitoring and evaluation mechanisms that capture CP

Existing PHC management and stewardship functions place little value or emphasis on community participatory approaches for chronic disease prevention. Routine PHC worker reporting and monitoring mechanisms do not also capture work performance or activities in relation to NCDs: “As you know we have tracking indicators for MCH programs. We are pretty good at following up and committing to delivery of interventions when these are part of our performance report …right now we have nothing on measuring our work involving communities”[PHM 4B].

The fact that community initiatives were not embedded into routine PHC worker performance evaluations and reporting templates suggests that these activities were “peripheral” and non- mandatory. The lack of formal integration acted to discourage CP practice: “Although programs to fight chronic diseases are not directly on our duty list, we are keen to work on these initiatives. But the real issue is as she pointed out is that if we do, we are not supported…or appreciated by the system, by our superiors, for our initiatives” [PHI,3T].

Page | 143

5.2.5 Role of curative care institutions in enabling CP for NCD prevention

Most health administrators suggested hospitals and other curative care institutions were not conducive to building meaningful CP practice as there was little or no opportunity for long- term engagement with patients who “simply want treatment for an illnesses, nothing more”: “I don’t see how hospitals can build community participation...how can they? People show up for treatment when they are ill and they go back home or are referred. How can there be any participation?”[MO-MCH, N] “Rural hospitals here simply don’t have enough time to engage communities in health promotion” [MOH, L]

Participation in institutions like rural hospital services was seen as “attendance at screening programs”, follow-up visits to clinics and maintaining treatment compliance. There was no community expectation for anything beyond this: “People show up for clinics, that is the level of participation shown by the community…I think it will be strange if we start community development interventions from this hospital...that’s the job of the MOH office…” [MO–RH,H]

Another reason identified for lack of focus on CP within such institutions was that community members often ‘bypass’ lower level facilities. People are less likely to access curative care services at village level institutions (such as rural hospitals) for treatment and management of NCDs. They ‘bypassed’ these government PHC facilities to access tertiary care institutions and/or private sector hospitals as these were perceived to be offering better quality care and access to specialists. “There is a public perception that these [rural hospital] clinics…they don’t have the correct equipment...staff constantly rotating. And yes, this is true in many cases. You see, there are no auto-analyser machines in many rural hospitals, so even for simple blood tests, people have to go to higher level institutions, sometimes essential statin drugs are unavailable. So how can you expect people to come? This is why most patients bypass them and go to general or private hospitals directly” [CCP, MOH2].

A small number of responders did recognize the important role played by curative care institutions in enabling community engagement, especially when continuity of care needs are integrated with community-based care: “Village level hospitals should have a relationship with our preventive care units especially on how patients can be case-managed at community level… there is a role for all of us in case- management of the chronically ill…from preventive care, health promotion to curative management” [HEO, L]. Page | 144

B. Professional ethos and cultural barriers

5.2.6 Culture of participatory decision-making within the health system

The interview findings indicate a health system culture in which there is little or no participation between workers in PHC units at field level and their managers at more senior administrative levels. The existing health stewardship culture at PHC level provides a limited platform for worker engagement in health care decision-making: for example, with respect to PHC intervention design and planning: “The meetings we regularly have with our supervising MOH is only for information sharing…we don’t talk of what we have tried...it’s not about launching new ideas, it’s reporting only. I think we will get better motivated if they [District health administrators] could give more time for us to discuss new activities” [PHM,2T].

PHC workers highlighted the importance of a supportive management team that provides the stimulus for enhancing and scaling-up locally driven health interventions: “…even when there are good practices these are not really taken up” [PHM,3B]. Most PHC workers saw their role as “being the service provider”, and not especially involved in program planning, design and advocacy: “We are there to implement programs developed by our Ministry of Health. Of course we discuss program activities and issues at our monthly conferences. It will not work if everybody starts making their own program! No. There is standard to be followed” [PHI,2H]

Interestingly, most PHC workers and some district level PHC managers were unaware of the existence of a National NCD Policy or the presence of a dedicated post of the Medical Officer of NCD (MO-NCD) within their own district. Established under the National Directorate of NCD, the MO-NCD post was formulated in order to support strategic planning at district level for NCD prevention and control.

Inculcating a culture of CP appears challenging when collaborative decision-making processes within the fabric of the health system are elusive.

Page | 145

5.2.7 Fear of losing expert authority

The potential for reshaping power towards community members and of losing their ‘expert authority’ as a result of CP was rarely acknowledged by any of the respondents. However, a few workers pointed out, and in some cases even ‘feared’, how meaningful forms of CP may concentrate authority among participating community members or their representatives, thus potentially corroding the power and trust of the health authority: “Community participation is important…we all know that…but I fear, that some of the senior health officials won’t like the fact that when you have a community rep on a committee making decisions and having…how can the rep advise on the way say a dengue is to take place. Health officials at that sort of meeting will lose respect, and I don’t think that will be accepted at higher levels if that is reported back” [HEO, N].

A few PHC workers cited the implausibility of having community members ‘advise them’ on matters of public health: “How can I ask them [community] to get involved in something that is so technical?” [RD,N]. These respondents cited intervention complexity and lack of formal training as constraints on communities’ ability to understand and advise on health care issues.

5.2.8 Historical prioritization of MCH interventions

A traditional dominance of maternal and child health (MCH) programs within the core- package of preventive services was highlighted by health administrators as a factor in limiting innovations such as NCD prevention within the PHC ‘intervention space’. Respondents felt there remained a historical prioritization of MCH programming and ‘slowness’ by policy makers to respond to the rapidly changing epidemiological burden and demographic changes at village level. The ongoing focus was attributed to sustained advocacy from relevant departments within the MOH and to organizations like UNICEF: “You know, we all talk a lot about the ‘life course approach’, and UNICEF has massively promoted this too. But we are neglecting the men, the youth and the elderly of family. Participation in health is definitely lacking there” [MOH, 2N]

“I don’t blame the Family Health Bureau for constantly pushing to make MCH the main platform of our public health service. It’s the other departments that have not pushed the needs of missing community segments…” [MOH, CCP3]

Page | 146

While recognizing the importance of MCH, some suggested its sustained prioritisation had led to a limited focus on or even disenfranchisement of community and household members, namely men: “Why do we have only have national programs dedicated to ‘well woman clinics’ all over the island?...Why not have ‘well men’ clinics? Or involving youth for instance?”[MOH, 2N].

“Males are not linked into our village health services at all in the same way women are…right from pregnancy, young mothers develop partnership with our health service through our PHM ...but males do not form these bonds with health system” [D-MOH,1].

5.2.9 Role of General Practitioners in CP

PHC workers and HAs affirmed the dominant model of general practice in Sri Lanka as one which “treats patients for their individual conditions” and in which “we don’t really have a family medicine approach”. PHC units partnering with private primary care providers in enabling CP approaches were therefore deemed impractical. As one senior director stated “there is really no connection between the PHC units and private practitioners at village level, even though almost all private practitioners are Government medical officers that operate during non-office hours!” The emphasis of private clinics was on ‘moving on to the next patient’ rather than taking a health promoting approach: “Private GP set up in Sri Lanka is like a rotating OPD clinic…you definitely don’t get GP group practices setting up health promoting life-style modification sessions and undertaking familial risk factor assessment….no. That doesn’t happen here…it’s one patient and then next patient model, you know...” [CCP, MOH1]

Notably, senior HAs cited a few cases where doctors posted to work in PHC units in rural and remote settings had successfully established their own private practices, and were well supported and “nurtured” by the rural communities they serve: “You see… when a doctor wants to establish themselves in a rural hospital that community nurtures that doctor... Then there is real feeling of community spirit here. Not always, but there are cases.” [DC, MOH3].

According to one administrator, the integration of private practice in such cases actually enabled the promotion of CP and community development initiatives:

Page | 147

“Because he [the medical officer] was respected as their local area, they listened to him. Whenever we [the district] wanted to do a dengue control program, his area participated greatly…he had lots of influence in that community” [MOH, CCP3].

C. Structural barriers

5.2.10 Scaling down PHC service areas

District level HAs and PHC workers argued that the intensive nature of community-based health promotion practices, NCD case-management and patient follow-up demand greater commitments of time and resources. They argued that if PHC workers are to play a meaningful community development role with current schedules and clinical case-loads, then scaling down the size of existing geographical PHC catchment areas was needed. Workers argued that smaller service area demarcations would allow for more intensive community engagement and ultimately more effective chronic disease interventions within domestic settings: “Our catchment areas are too big, if the PHM and PHI areas can be reduced, we can spend more time on NCD prevention and we can assure good quality of our work. It’s not just diabetes…aged care, community mental health also need to be tackled simultaneously” [PHM,1T]

“If they [PHC workers] are expected to continue routine work but take on more programs with communities, it is impossible. This is why I say a reduction of the service area is a must” [MO-P, N].

A small number of PHMs invoked the true historical designation of their profession as ‘family health workers’ and suggested changing the focus of their work to engage more intensively with families rather the current emphasis on “maternal and child health issues”. They argued that shifting the focus to the family unit rather than the unit of mother and child offered a more inclusive approach compatible with NCD prevention work: “We [PHMs] are recruited really as ‘family’ health workers, but we only work on maternal and child health issues, the ‘family’ part is not really covered!... We know with chronic diseases, risk factors run in families, so it is family health that we need to do…So start by reducing our catchment area to work on a family approach” [PHM,1K]

Page | 148

5.2.11 Inadequate resource allocation and logistical constraints

Decentralization of Sri Lanka’s health sector in the 1980s created a ‘split’ in the health budget to both national and provincial health systems. Concerns were raised by HAs that funding allocations to the districts were ineffective and that resourcing curative care facilities was prioritized above prevention and health promotion programs: “There is lots of political manoeuvring of budgets given to the provincial health authorities each year…final result is allocative inefficiencies ...and fragmented services” [D-MoH,3]

There was a prevailing lack of clarity in communication and coordination between district level health authorities and workers at peripheral PHC posts in some settings in terms of financial and resource allocation, as exemplified by the following statements: “The funding for NCD screening clinics is allocated within the annual budget plan of our district...Unfortunately I don’t see many requests coming in from the field” [MO-, H].

A PHC manager located within the district stated: “a lack of financial resources is a major issue in tackling NCDs in our setting… we don’t have funds to do health promotion projects here” [MOH, H].

Limited transportation facilities and funding for PHC workers to undertake travel to remote communities was also highlighted as a logistical barrier for effective community engagement. Large geographical areas of coverage and difficult terrain also limited community outreach and follow-up activities: “It is very hard to reach some estates [communities] here… there are poor road conditions that makes it hard for us to reach them…it’s frustrating…it’s demotivating” [PHM, 3K].

“The petrol money allocated hardly covers costs of the outreach work. I have to pay from my own pocket” [PHM, 1T].

5.2.12 Addressing human resources for health gaps

Vacancies in PHC worker posts (especially in rural settings) have over-stretched some PHC units, making the completion of even core clinic schedules difficult. Addressing such structural issues in human resources at PHC level was highlighted as a priority before undertaking any ‘push’ towards increasing community engagement for NCD prevention: “…we cannot add or start any new approach until vacancy issues are resolved” [PHM 2T].

Page | 149

The continued application of an “out-dated mechanism” for allocating PHC posts within the district health system was also highlighted by a number of HAs as a constraint; they explained that the current system disregards population densities, geographical variables and specific terrain.

Community factors

Themes under this title concern factors and dynamics that HAs and PHC workers perceive at the community level, outside the health system, and which may influence CP. These include the challenge of negotiating power structures and ‘gate-keepers’ within communities, erosion of the culture of volunteerism, negative perception towards CP in health and time limitations.

A. Community attitudes

Under this theme a number of categories were defined involving negative perceptions, socio-cultural hierarchies and demand for curative rather than health promotion interventions.

5.2.13 Lack of perceived benefit in CP for communities

A small number of PHC workers and administrators expressed their lack of commitment to CP as a core element of primary health practice. This disenfranchisement was attributed to the ostensible unsupportive or negative attitude held by communities towards CP in health. According to such workers, communities are happy to be passive recipients of health care, rather than play a proactive role. These negative community perceptions and beliefs ultimately deterred health workers from pursuing meaningful community engagement strategies in their routine PHC practice: “…our communities really want to get involved in planning health interventions with us. These are rural communities. In our culture that kind of interaction is just not there [in health care]. You see, the public has that mentality that doctors must do the health care, not them!” [MO-P, N].

Page | 150

5.2.14 Demand for clinically oriented interventions

PHC workers suggest that communities prioritize clinically oriented interventions over health promoting ones. More specifically they felt communities have more “enthusiasm and engagement” for disease screening programs than for primary prevention. As this quote from a PHM indicates, creating community awareness of the benefits and need for health promoting approaches is essential before any sustained CP intervention in NCD prevention is to be facilitated: “The thing is, we need to first convince village communities on the importance of prevention…because you know our people, they want immediate benefits…people love to go for a free outreach clinic to see doctors...get medicine. This is why you will find our villages coming to screening clinics. They want to be tested! So show up for testing. Again it goes back to the point of our people’s attitudes towards prevention” [PHM, 1K, H].

Community attitudes were thus identified as a powerful determinant in obstructing or enabling the manifestation of meaningful engagement with health service providers.

5.2.15 Time constraints

Time constraints stemming from multiple commitments left community members little time to engage in health initiatives via community development processes: “As you know…the majority of the communities make their income by engaging in agriculture here. Long work days and times are tough. Cost of living is high. I think people are also doing multiple jobs. To volunteer their time a community health program or even obtain participation in outreach programs is proving very difficult” [RDHS,H].

Page | 151

B. Socio-cultural and political barriers

5.2.16 Socio-cultural hierarchy

Some respondents suggested that CP was influenced by a socio-cultural hierarchy embedded in Sri Lankan society that placed health professionals ‘above the community’, where health is the sole domain of medical experts. As this quote from a senior policy maker suggests, this prevailing social attitude that ‘health is a domain only for experts’ inhibits an otherwise vibrant civil society culture in Sri Lanka, where communities actively engage in other domains of social development: “There is no doubt that we as Sri Lankan citizens get vocal and get involved in issues affecting us… If you just look at the way our people are demanding for better educational reform and so on…but health sector is different. You don’t find the same type of protest, or push from people demanding health services or against the medical profession…Just open the newspaper, just see the debates our general public are having on health. It’s not there!...Debates are there, but these are by trade-unions and not between community and health development” [D-MOH,4].

5.2.17 Gate-keepers and political opportunism

Many respondents drew attention to the broader socio-political complexities involved in mobilizing communities and in facilitating community representative structures. The fear of political influence and negotiating “political opportunism” was cited as an impediment to PHC workers in developing and sustaining community mobilization activities: “I’m sure you realize the political opportunism this country has! …especially in rural areas. Nowadays, political people, get nervous when they see communities being organised even with something so positive as public health, it automatically becomes…a political act for them. Many of us are very careful to get involved” [MO-P, H].

“I have seen many good community projects just hijacked by local political groups or religious figures. Workers need to be careful they don’t get caught up in all that, especially in politically sensitive areas” [CCP, MOH4].

Another theme to emerge was the challenge of negotiating with and managing the dynamics of community leaders or “gate-keepers” who represent (or claim to represent) a particular community or constituency. Community representative processes are “not straightforward” Page | 152

and present a challenge for PHC workers, as they have to navigate through such community hierarchies and power-dynamics. This makes truly meaningful participatory engagement challenging: “What I see is that many villages enthusiastically participate in health programs, but a growing number who come with links to political big shots. They use health programs for their gain…so what I’m saying is that we need to be careful about engaging communities…we can’t just jump in just because it’s a good thing” [MO-P,P].

Difficulties were also articulated in engaging elders in close-knit communities around cultural aspects affecting health behaviours or female participation, with one HA stating, “I assure you, a major obstacle preventing young Tamil women’s participation in that estate is the attitude of some of the community elders”.

5.2.18 Corrosion of volunteerism

PHC workers expressed the view that there was a greater degree of monetary expectation by communities for volunteerism in the health sector, which has had some inhibitory effects on CP. The provision of excessive remuneration and incentives for community volunteers from private donors and non-governmental organizations in the aftermath of tsunami disaster relief operations in 2004 were hypothesised as a major contributing factor: “After the tsunami, there were just so many NGOs that worked in this area. You know they were providing major rupees to the workers from the community … it has been a real problem for us to get the participation of the same people on our community since they expect the same type of payment package…sort of a ‘volunteerism barrier’…” [HEO, H].

Workers in other districts not affected by the tsunami stated that donor funded development aid programs that offered monetary incentives also undermined the ‘spirit’ of health volunteerism and disenfranchised indigenous models of community action which offered no financial incentives: “We started a local village program called ‘Nutrition Village’ with local community support…this was a while back. Many community members volunteered their time… it was really starting to develop well. But when the government started ‘Divi-Neguma’(Village Rural Development) program, our program stopped because that program gave money to do certain community events and activities. So people went to [participate in] that program, leaving our program” [PHI, 3K].

Page | 153

5.3.1 Application of the P-Compass tool

The P-Compass tool was introduced to participants at the end of the interview as a means of gauging the level of CP in the PHC setting in which they worked. The overall degree ranged from no engagement with communities to more collaborative and meaningful partnership.

The tabulated scores of the P-Compass tool are shown in Table 8. As indicated, the dominant form of participation is one in which communities are at best ‘consulted’ or made ‘compliant’ in the implementation of PHC programs. A majority (79%) of respondents provided a P- Compass score ranging from 0 to 2, indicating an overall low level of CP in their practice setting. Here, the mode of participatory action is one where most PHC workers worked ‘on communities’, providing limited scope for iterative engagement.

Responses for each PHC worker category were also mapped, as shown in Table 8. The majority (90%) of PHIs reported low levels of participatory engagement (0 to 2 score range) than PHMs (at 55%). PHMs reported a relatively high degree of partnership and collaboration with communities, with 45% reporting scores ranging from 3 to 4.

5.3.2 Reflections on using the P-Compass

Interview participants found the P-Compass tool to be useful in capturing general reflections of CP practice within their PHC setting. As one HA indicated; “This has been a very useful exercise in reflection…I feel there is lot that is here [pointing to the card] in a simple and accessible way…it makes us practically think about how we link up with our communities”.

The use and placement of key terms with informative graphics also appeared to have aided responders to quickly deconstruct the participation gradient: “I think this is very simple. These here [pointing to the graphical icons on the tool] are very easy to understand and makes us think about where we are with this issue on community participation…can I take this card with me? I would like to read it more…” [PHM, 3B]

The graphical delineations on the cards therefore provided immediate and accessible cues to assist participants in identifying the overall dynamic and forms of participation.

Page | 154

Table 8. P-Compass findings

Score & Degree of Participants Mobilization Health Primary Summary level of participation action method administrators health care scores participation workers DL CL PHM PHI n=20 n=4 n= n=18 31 4 Very high Collective Has control Community action “By development 0% 0% 5% 0% 1% community” 3 High Co-learning Has Community delegated collaboration authority 20% 5% 25% 40% 10% 3 High Co-operation Plans jointly Community “With collaboration community” 2 Moderate Consultation Advises Community “For mobilization community” 55% 75% 55% 75% 65% 2 Moderate Compliance Is consulted Community consultation 1 Low Co-option Receives Community information compliance 30% 0% 0% 15% 11% “On community” 0 none none none none 10% 0% 0% 0% 3% DL = District level Health administrators (Regional Health Directors, Medical Officer of Health); CL= Central level Health Administrators

Section summary

Health professionals interviewed clearly articulated the need for and importance of CP in NCD prevention at the PHC level. Factors that limit meaningful CP practices in PHC settings ranged from structural health system factors such as the existence of large PHC worker service coverage areas and weekly clinic schedules that provided little time for meaningful community engagement, to attitudinal factors such as the belief by HAs that communities place little importance on health promotion and engagement with health professionals in reshaping service delivery. The anxieties in negotiating with community gatekeepers and navigating political power structures for harnessing community representation were also indicated.

Page | 155

If CP is to be meaningfully embraced, then an enabling environment needs to be created to address such factors. Some strategic measures appeared to be relatively simple, requiring programmatic or technical changes that affected PHC practice, whilst others required significant reform or cultural change within the health system. The technocratic notion indicated by respondents was that unless CP is enshrined within adequately resourced PHC programs, and is formally synchronised into worker duty lists and reporting templates, it is unlikely to be practiced.

There was also recognition of the need to address attitudinal barriers to promote a culture of CP within PHC units. ‘Resistance’ from some health professionals and collegial structures which do not value community engagement in health care decision-making or in enabling reforms within PHC practice was also recognised. As highlighted by a senior health administrator, “Of course there will be resistance! Whenever there is change there is resistance….But we need this change.” Enabling a ‘culture of participation’ has implications for not only the interaction between health workers and their communities, but also requires harnessing a participatory culture within the health system. Collaborations between PHC units, district level HAs and central level policy makers may be needed in order to develop effective indigenous models of participatory practice.

Page | 156

Findings from the focused ethnographic and observational study

Part A: District health system profiles

Page | 157

Figure 26. Overview of the type of data captured from the observational and focused ethnographic study results.

Page | 158

Overview of the observational and focused ethnographic study results

Figure 26 on the previous page provides a graphical overview of the results from the four parts of the focused ethnographic study, including details of the observations made, persons observed and records reviewed.

I utilized findings from structured observations and focused ethnographic methods to build a rich composite ‘map’ of the actual routine work environment and ‘intervention space’ of PHC workers at district level. This enabled me to gain a better understanding of how PHC workers organise and navigate their program interventions at PHC level.

5.4 Part A: District health system profiles

5.4.1 Demographic and chronic disease profile

Nuwara Eliya district is the most diverse of the three districts studied. It is populated by Tamils (51%), Sinhalese (40%) and Muslim/Moor (9%) communities. PHC workers demonstrated awareness that there were specific NCD risk factors along ethnic and socio- cultural gradients in the communities they served: “The estate Tamil community definitely has the highest risk [for NCDs] due to their unhealthy diet…we known this because of the level of under nutrition I see in the kids. For the Muslim community it’s the sedentary life styles and problems in the dietary pattern…”[MO-MCH,N].

“…unlike the male Tamil estate workers, the men in Sinhala community don’t take ‘beedi’ tobacco as much, but they definitely match their alcohol drinking habits equally... [PHI,N2].

“Domestic violence and injuries are reported more among the Tamil community in our area. Of course I can’t take you and show statistics but I know the pulse of the community. You see there are attitudes that are different in each group…” [PHM,N4].

A Chronic Disease Profile for each district was compiled using records provided by planning officers from each RDHS office. Ischemic heart disease was reported to be the leading cause of mortality in Nuwara-Eliya district, and cerebrovascular disease (stroke) for both Polonaruwa and Hambantota districts. Variations most likely stem from inconsistent reporting and missing data from the morbidity and mortality registries maintained at hospital level - a finding echoed in previous research undertaken into maternal mortality (385).

Page | 159

5.4.2 Human resource for health profile

Human resources for health (HRH) is an essential component of a health system. Without adequate health personnel at PHC level, undertaking comprehensive and effective community-based programming becomes difficult.

Data on health-related personnel for each MOH area were compiled from information derived from records provided by each respective MOH office, and complemented by interviews with PHC workers (see Table 9 below).

A total of 18% of all PHC posts remain vacant across the three districts. Nuwara Eliya district had the highest number of cadre vacancies across all categories of PHC workers. In analysing the type of PHC worker posts, the Assistant MOH (AMHO) position revealed the greatest percentage of vacancies (67%). This result was attributed to the fact that the AMHO post had been newly created in 2012, and that training and recruitment for postings were still being finalized. When the AMHO posts were excluded, the largest proportion of vacancies was reported for Supervising PHIs (38%), followed by PHMs (16%) and Medical officer of Health (MOH) posts (17%).

Since the MOH and PHM posts are essential for the delivery of the basic package of public health services in Sri Lanka (229), such vacancies adversely affect PHC service coverage. As a result, medical officers in neighbouring PHC units and PHMs within the same unit absorbed the service coverage and caseload allocations. The expanded service coverage areas and time pressures needed to conduct the basic package of outreach clinics affected the ability of many PHC workers to meaningfully engage communities: “It’s been about 12 months without…you know, the post in the [name of midwife area] being vacant. I think we all do our best to get all the work completed in time but you know in some months I can’t find the time to get through even my own cases…I know I can do more with the communities, it’s needed but schedule is packed” [PHM, 5N].

Page | 160

Table 9. Human Resources for Health District Mapping (2013): Active posts and vacancies by district, MoH area and specific cadre

District MOH area MOH AMOH PHM PHI PHNS SPHI HV Cadre vacancy (%) A V A V A V A V A V A V A Nuwara Eliya Kothmale 1 0 0 1 16 9 3 0 1 0 1 1 7 33% Nawathispane 1 0 0 1 17 4 2 2 1 0 0 1 3 28% Hambanthota Beliaththa 1 0 0 1 20 3 4 0 1 0 1 0 1 13% Tangalle 0 1 0 1 24 3 6 0 1 0 0 0 1 14% Polonnaruwa Thamankaduwa 1 0 1 0 23 1 6 0 1 0 2 1 0 6% Lankapura 1 0 1 0 20 3 1 0 1 0 1 0 1 11% Proportion of vacancies of 17% 67% 16% 8% 0% 38% specific cadre personnel (%) Legend: A = Active post; V = Vacancy. MOH = Medical officer of Health; AMOH = Assistant MOH; PHM = Public Health Midwife; PHNS = Public Health Nursing Sister; SPHI = Supervising public health inspector; SPHM = Supervising public health midwife; HV = Health volunteers. Cadre vacancy calculated by: (V/(A+V))*100.

5.4.3 PHC demarcation and effect on community engagement

Despite having a population twice that of Hambantota district, Nuwara Eliya district had an equal number of PHC units. The district profiles presented in Section 5.4.3 also indicate that the allocation of PHI and PHM areas was not clearly related to population distributions in each district. For instance, PHM work areas in Nuwara Eliya were double that of Hambantota, yet the number of PHI service areas remained the same. Interviews with health authorities at both district and central levels revealed that such discrepancies in human resource allocations and PHC demarcations stemmed from a lack of evidence-based approaches to health system planning. Allocative inefficiencies have also been created by a culture of political influence and interference from both local and provincial governments. Box 4 below presents evidence of external forces influencing PHC area demarcations. Pressure from health worker trade unions and a lack of robust policy and regulatory frameworks from the Health Ministry also contributed to ineffective health system planning.

The overstretched PHC units with cadre vacancies, low PHC worker to population ratios, and communities dispersed across vast geographical barriers (such as those living in the mountainous terrain of Nuwara Eliya district), may ultimately make intensive community

Page | 161

engagement and outreach activities difficult. Not surprisingly, data from the worker follow- up study (see next section) showed that the time spent for travel and the overall monthly expenditure on outreach activities were highest for PHC workers in Nuwara Eliya district. Such health system factors were also highlighted in interviews with PHC workers.

Box 4. Nuwara Eliya district: mini case study

A Health Ministry policy decision was taken in 2009 to increase the number of PHM area demarcations in the district from 234 to 480. It was seen as one means of “increasing the reach” that the district health system had within communities living in privately owned enclaves of the politically powerful ‘tea barons’. The decision was prompted by calls from ruling politicians and political lobby groups for the “state health services to better engage with marginalized estate populations,” as articulated by a senior MOH official. However, the rapid expansion of PHM areas was not matched by increased recruitment and deployment of PHC workers. No additional resources were provided to the PHC units following the ministerial decision. Many workers posted to the district would not remain for their allotted placements. The limited resources and complex social problems of estate populations inhibited the flow of workers to the remote hill-station postings. Nuwara Eliya district has the highest number of vacancies across all PHC worker categories, with up to 33% in Kothmale MOH.

PHC workers indicated in interviews that smaller service area demarcations would allow for more intensive community engagement and ultimately more effective chronic disease interventions at household level. They indicated that if PHC workers were to play a role in NCD prevention and promotion, then scaling down to smaller service areas would be needed. One PHM argued: “We have a huge area to cover in providing midwife services... the system should reduce our catchment area, get more workers so we can have more time with communities... [it] takes time to work on NCDs”.

Page | 162

Part B: Observational follow-up study

5.5.1 Characteristics of PHC workers

An observational study was undertaken to document the resources (time, financial, material and human) incurred by each PHC worker in undertaking health interventions at community level. As indicated in the methods chapter, I aimed at documenting what PHC workers considered to be community-based health interventions and documented these through observational follow-up.

Two public health midwifes (PHMs) and two public health inspectors (PHIs) from each selected district (total of 6 PHMs and 6 PHIs) were invited to participate in a structured observational study of 4 weeks’ duration. All 8 PHC workers consented to participate and agreed to provide details on resources and expenditures incurred in carrying out routine activities. One PHI withdrew on day 6 of follow-up and another on day 13 due to personal reasons10. The final analysis therefore involved 6 PHMs and 4 PHIs.

5.5.2 Mapping ‘community-based’ health interventions

There was little engagement with communities utilising clinic services in order to facilitate health promotion activities or collaborative discussions on mitigating general or specific health issues.

The interventions conducted by PHMs involved outreach clinics (an average of 6 in each MOH area every month) and domiciliary visits (an average of 16 households per month). The community clinics were conducted in rural hospitals, school buildings and community halls located at village level. In terms of the type of clinics conducted, antenatal clinics accounted for 17%, family planning clinics 8%, mental health outreach clinics 8%, child nutrition and anthropometric (growth) clinics 25% and ‘well-woman’ screening clinics for 25%. Interventions at community level conducted by PHIs involved school medical inspections (an

10 Both PHIs contacted field researchers and their respective MOH supervisors to inform them of their withdrawal. One had an unexpected family emergency and other took leave of absence. Page | 163

average of 4 each month for each MOH area), environmental health education programs (an average of 5 per month) and school vaccination program (2 per month for each MOH area).

Tables 10 and 11 provide details of datasets on the average time spent by both PHMs and PHIs in conducting community-based outreach clinic activities. PHMs spent an average of 6.5 hours undertaking the community clinics, with an average of 19 minutes being spent on a health education talk with attending mothers. The average time spent on the school health program by a public health inspector was 5 hours 15 mins. The average time the PHI spent on engaging with communities via a health education talk was 36 minutes.

The monthly data analysis reveals how PHC workers devote a significant proportion of their time to undertaking community-based clinics. None of the PHC workers that were followed had engaged in community-based health promotion interventions or those having a major community capacity building component. CP approaches may require significant investment in time with communities, at least at the formative phases of the intervention where intensive engagement may be needed. This would certainly make it challenging for PHC workers to embed such intensive community engagement efforts within existing work programs (also reflected in Section 5.2). As one PHM indicated “we barely have time to meet the monthly clinic targets...”.

Table 10. Community outreach activities by PHMs (clinic-based and household)

Community outreach clinic visit Household/domiciliary Average time Average time spent on Average time spent for visit taken to delivering health the travelling from (Avg. time spent with conduct a education talk (T) residence to outreach mother vs. other clinic* clinic members of household during household visit) PHM1 6 hrs. 25 min 14-20 mins (T) 1 hr. 55 mins Data excluded due to recording errors PHM2 5 hrs. 25 min 19-23 mins (T) 1hr 49 mins 61 mins with mother, 11 mins other. PHM3 6 hrs. 05 mins 10-15 mins (T) 25 mins 56 mins with mother, 16 mins other. PHM4 6 hrs. 55 mins 17-25mins (T) 40 mins 45 mins with mother, 18 mins other. PHM5 5 hrs. 50 mins none 55 mins 58 mins with mother, 14 mins other. PHM6 5 hrs. 28 mins 12-26 mins (T) 20 mins 52 mins with mother, 12 mins other. Page | 164

*includes meal and rest periods

Table 11. Community outreach activities by PHIs

Average time Average time spent on Average time taken to Average time spent for taken to delivering health conduct food the travelling from conduct a education talk at each inspection personal residence to school health setting program location program PHI 1 5 hrs. 10 min 20 mins 55 mins 55 mins PHI 2 5 hrs. 05 mins 36 mins 1 hr. 05 mins 1 hr. 35 mins PHI 3 4 hrs. 55 mins Not specified 1 hr. 14 mins 10 mins PHI 4 5 hrs. 20 mins 33 mins 58 mins 40 mins

5.5.3 Resources and expenditures incurred

A general perception among PHC workers was that there were insufficient resources provided by health authorities to undertake community health interventions. Some suggested inefficiencies and inequalities in resource allocation. For instance, PHM 2 from Pollonaruwa district is provided with a MoH motorbike, while PHM 1 in the same district has to utilize public transport via a bus or three-wheeler. The overall monthly out-of-pocket expenditure on undertaking outreach clinics was nearly double for PHM 1 than for PHM 2.

The PHC workers incurred an average cost of 1,150 rupees per month ($9 USD) in undertaking community outreach activities. The most costly items were transportation and fuel (Table 12). With an average monthly income of 19,070 rupees (386), some workers were paying upwards of 10% of their monthly salary (non-reimbursed) to cover the costs of conducting community outreach activities. Some PHC workers had no formal transport facilities allocated to undertake outreach interventions whilst others were supported with an official health ministry vehicle. When vehicles were allocated, fuel allowance allocations were not consistently provided. The irregularities in resource allocations such as the provision of a vehicle by the MoH were attributed by PHC workers to inefficiencies, corruption and nepotism that prevailed within approval processes of the health bureaucracy. The role played by powerful health worker unions in obtaining entitlements was also cited.

There appeared to be no transparent and equitable system established to allocate resources to PHC units, leading to disenfranchisement of some workers. This affected the ability of some PHC units to effectively undertake community outreach interventions. As one PHM indicated, “without the appropriate travel allowances I struggle to conduct outreach

Page | 165

activities…especially when doing follow up activities. I take from my own to pay for this but it’s not right. It should not be like this…”.

Table 12. Resource expenditures incurred by PHC workers in undertaking community health interventions

MOH Mode of transport to Ownership of Overall monthly IIEC material and other

offices conduct outreach vehicle expenditure on resources for NCDs (PHI/PHM) activity undertaking outreach activity District [in Rrupees] PHM 1 Public transport None 1,400 to 2,000 Yes. flip charts and (bus/three wheeler) posters from an NGO. PHM 2 moped bike provided by 800 to 1,100 None MOH PHI 1 motor bike provided by 450 to 900 Yes. flip charts and MOH posters from an NGO.

Polonnaruwa PHI 2 motor bike provided by 720 to 1,100 None MOH PHM 3 Public transport None 1,250 to 1,950 None (bus/three wheeler) PHM 4 moped bike provided by 405 to 850 None MOH PHI 3 motor bike provided by 250 to 900 None MOH

Hambanthota PHI 4 motor bike provided by 450 to 1,100 None MOH PHM 5 moped bike None 1,040 to 1,500 None

PHM 6 Public transport None 1,150 to 1,750 None (bus/three wheeler) PHI 5 motor bike provided by 750 to 1,140 none MOH

NuwaraEliya PHI 6 motor bike provided by 730 to 1,050 none MOH

Page | 166

Part C: NCD prevention and control interventions

5.7 Mapping NCD prevention and control interventions

As a result of the limited number of community health interventions aimed at addressing NCDs that were identified from the six selected MOH study settings, I extended the search ‘net’ to cover all MOH areas within the three selected districts. The expanded search included all 7 MOH areas in Pollonaruwa district, 12 MOH areas in Hambantota district and 13 MOH areas in Nuwara Eliya district. The HAs and PHC workers that participated in the interview process and observational study component also asked their peers during formal district level meetings and informal worker gatherings to contact the research team if they had knowledge of any community-based health interventions. The resulting intervention ‘map’ is presented in Table 13.

As indicated in table 13 below, the community based NCD initiatives that were implemented with support of the PHC units were characteristically ‘one-off’ interventions that were undertaken once without any follow-up, and were not sustained over extended periods of time. Most interventions were not integrated into routine PHC practices and were not sustained. Of a total of six interventions, only two were found to explicitly utilize CP approaches to address NCD prevention goals, and to do so within the routine PHC practice setting. The first, catalysed in Nuwara Eliya district, involved increasing community awareness of healthy diets to combat childhood obesity, ensure proper nutrition for early childhood development and provide healthy nutrition options for participating families. The second intervention was identified in Pollonaruwa district and involved an oil, salt and sugar reduction program for households to enable healthy diets. Both these interventions were initiated by PHMs, and are profiled in the next section as in-depth case studies.

Page | 167

Table 13. Mapping NCD prevention and control interventions Intervention details Intervention frequency and Intervention sustainability catalyst

1. National authority on tobacco and alcohol One-off NGO/Donor (NATA) program targeting school children in selected schools to promote messages on negative effects of alcohol and for tobacco prevention. The NGO funded program undertaken in partnership with district health authorities ended with closure of funding.

2. Alcohol prevention program in the village of One-off RDHS Sooriyawewa in response to a high profile case of a victim of domestic violence. Subsequent media attention of alcohol abuse and domestic violence cases in area prompted authorities to work in partnership with both PHC and social workers to be trained in violence prevention and mitigation. After training was provided to PHC workers within the MOH area at onset, the program had no sustained follow-up or support mechanisms for the PHC unit.

3. “Healthy lifestyle centres” is a MOH initiative to The clinics were not Central level offer population based screening for identification conducted on a monthly MoH (NCD of NCD risk factors for those aged between 35-65 basis as per ‘healthy Directorate) via years. The ‘healthy lifestyle centres’ pilot project lifestyle centres’ plan (34). WHO funding and the “Package of Essential Non-communicable . Disease Interventions for primary care” (PEN) funded by WHO was found to be active in 3 MOH settings. The ‘drop-in’-style community clinic-based intervention aimed at screening local populations for hypertension, diabetes and cardiovascular risk factors (34). PHMs and PHIs were requested during their routine home and school visits to encourage adult family members to be screened at the healthy lifestyle clinic. Individual MOHs organised clinics according to their availability and capacity at field level.

4. Health promotion program conducted by a PHM Intervention involves a PHM – Nuwara to increase awareness of healthy diets to combat community Eliya childhood obesity, ensure early childhood care and empowerment/skills transfer development and provide healthy nutrition options component and an ongoing for the whole family. support mechanism. The intervention component is conducted at least once a month, sustained for a period of over 2 years.

5. Oil, salt and sugar reduction program for Same model as above. PHM – households identified through PHM. Polonnaruwa

Page | 168

Part D: In-depth case study analysis

An overview of the two case studies of NCD prevention interventions that utilized CP approaches is presented below. The context in which the intervention first emerged, its goals and development processes, the profile of participants involved and resources utilized are described. The P-Compass and Spidergram tools were then applied to describe and assess the dynamic of community participation within the intervention.

5.7 Case-Study 1: Salt, Oil, Sugar (SOS) Reduction Program 5.7.1 Setting: A PHM catchment area in Polonnaruwa District. 5.7.2 Background In conducting household visits of pregnant and lactating mothers in her catchment area, the PHM had noted a growing number of adult family members with obesity and chronic diseases such as diabetes. In the Well Woman Clinic she conducted with the MOH on a bi- monthly basis she noted an “increasing number of women presenting with hypertension and obesity”. She determined that unhealthy diets were a major determinant in this observed pattern: “I knew it was partly because of their diets. It has changed a lot, so much fatty food now…cheap too, unlike the increasing price of vegetables in our town”.

The PHM revealed that her interest in chronic disease prevention had been cultivated early in her career, when she had undertaken an assignment on metabolic syndrome at her midwifery training school. “Yes, there was very little teaching we got on NCDs…but I knew I wanted to do something even at that time…of course I had no idea how to!”.

The PHM initiated the program after obtaining permission from her MOH. PHC workers usually seek permission from their supervising MOH before engaging in any ‘new’ interventions: “When our medical officer in charge returned from a training program he attended in Colombo he gave a session to us all on aspects of chronic disease. I thought ‘this is my opportunity to start something’! So I approached him after the training and got permission right away….it started like that”.

Page | 169

The PHM decided to start a program to address the NCD risk factor of unhealthy eating through reducing added salt, sugar and oil in diets. The PHM first approached families who were attending the maternal and child health clinic (MCH) for which she was responsible in her area. She shared the idea of such an intervention and sought the advice of community members in designing it. The mothers she consulted suggested the formation of a community health club (‘Grameeya Suwa Sahana Kamituwa’) to catalyse intervention as they felt this had better organising potential and to enhance participating of families. Taking their advice, the PHM formed the community club within her PHM area.

In the first week, 6 households enrolled in the program. Encouraged by the feedback of those early adopters in the intervention, other community members also joined. By the 5th month after commencing from the PHM catchment, 38 households were actively enrolled in the PHM catchment.

In the formative phase of the intervention, the PHMs would visit the household and use child nutritional status as the ‘entry point’ for initiating the discussion. She would then discuss with family members their personal food choices, daily food habits and cooking methods. Householders were prompted to reflect on their food baskets and eating behaviours. She would then provide a brief overview of the SOS intervention and ask the family to reflect on the program goals. On a follow-up visit, the PHM would provide more in-depth information on the role that excessive fat, salt and sugar consumption in diets have in contributing to risk factors for stroke, diabetes and cardiovascular diseases. Different types of sugar and carbohydrates in locally available foods were differentiated. The intervention she would explain “aims not only at controlling added refined sugar, but also food products that had high concentrations of sugar, sodium and triglycerides”.

The families who agree to participate would then be advised to record the amount of salt, sugar and oil kept in the household, and then track the use of these on a daily basis. Those with a measuring apparatus (such as a kitchen balance and measuring spoons) were also encouraged to calculate the actual amount of daily sugar, salt and oil used. The PHM also assisted families to draw-up a ‘wall chart’ calendar to be hung in their kitchen, and taught

Page | 170

them how to track the consumption pattern over time11. The wall chart was used by family members as a reflective tool for behaviour change.

The PHM would visit the households periodically and assist them with goal setting and discuss strategies specific to each household to assist in promoting healthy lifestyles. Households that were early adopters to the intervention improved the tracking sheets and collected additional data such as household savings as a result of reduced consumption.

The PHM adopted a three-point strategy to ensure monitoring and feedback for those participating in this community-based intervention within her busy work schedule:

 At the conclusion of each MCH clinic, mothers participating in the SOS intervention were encouraged to share their experiences and needs and to discuss the personal strategies used in enhancing healthy diets and reducing excess consumption. This provided an opportunity for other mothers attending the clinic to ‘listen in’, stimulating interest and motivating others to participate. In observing such discussions in four clinic sessions, I found that the intervention group mothers had great influence in attracting others attending the clinic to participate.  The PHM used her routine domiciliary visits to spend extra time to discuss the intervention with participating households.  The community health club organised small group meetings on weekends for participating families within the MOH area and invited the PHM to provide feedback on their practices. This was an initiative led entirely by community participants. The PHM also invited other health experts to present at the health club sessions.

5.7.3 A short-film I developed on this Case Study (15 mins duration) can be viewed online at: http://case-studies.strikingly.com (Please note: the narration is in Sinhalese and sub-titled in English. The Tamil-language version, also sub-titled in English, is available on DVD).

5.7.4 Intervention impact and challenges

Whilst an empirical evaluation of the intervention outcomes and impact was beyond the scope of this research, an analysis of weekly tracking sheets and interviews with eight participating families showed the intervention had resulted in reductions in salt, sugar and oil

11 Basic data captured included the amount (in grams) of sugar, salt and oil purchased for household and utilized in consumption (tracked on weekly basis). Page | 171

use at household and increased household savings as a result of reduced spending. The savings factor appeared to be a major motivator driving sustainability of the intervention amongst male participants.

Evidence of positive behaviour change was shared:

“Tea was the one thing I needed to have with lots of sugar. As you know we [Sri Lankans] love our tea! When my wife reduced this, I was upset. But you know what happened? I started to drink herbal teas as a result and now got used to them. They don’t need much sugar…in fact they taste better without it! Herbals are better for you than sugar and packet tea [commercially packed] leaves, cheaper too” [Father, 32 years, Household 4].

In some families, the intervention also extended to reducing other risk factors such as alcohol and tobacco consumption, especially among adult males in the household:

“Madam [PHM] knew that I’m using alcohol, she saw me one day when I came to pick up my wife from her clinic. She talked to me.…you know, I felt bad about myself but I still continued. When my wife started this [intervention] I also got involved…then when she [the PHM] came to our house, we talked more. Slowly I also started to participate in this… we saved money. Alcohol was eating more money than all of this. I needed to cut out my alcohol. So I started doing it slowly too. I’m happy about this program and Madam. One day I will quit alcohol” [Father, 28, Household 3].

Community members revealed that traditional cultural practices reinforced by older women living in each household made implementation difficult. For instance, there was resistance from many elderly women to changing the practice of adding salt when cooking rice.

The PHM expressed genuine surprise at the way communities not only embraced the project, but engineered new additions and instigated program expansion by inviting other community members. However, she did acknowledge that future expansion would require dedicated resources in the form of time in order to manage the growing participant caseload and technical expertise to develop screening tools. The PHM emphasised her vision that it is CP that is the key enabler and sustainable force for the intervention:

“This intervention is about their own lives, so they should gradually take the responsibility on this intervention and continue it for a long time, they should change their attitudes as well and that will safeguard the long term success of this intervention”.

The PHM felt that she needed to undergo further training on addressing specific NCD risk factors and on home-based care of patients with heart disease and diabetes. She also Page | 172

expressed a desire to learn from behavioural change interventions that have proven to work in other settings. She indicated that, due to this effort, she had earned recognition from her MOH team and colleagues. However, she stated that many of her peers had been reluctant in replicating the model in their catchment areas due to the additional workload involved.

5.7.5 Resources used for the intervention

Resources are divided into four groups: a. Financial resources: All travel expenses (including cost of fuel for her scooter) to undertake fieldwork and home visits were incurred by the PHM using her own funds. b. Time: The formative phase was usually undertaken during one or two household visits, where the PHM spent approximately 50 to 90 minutes. Follow-up visits for monitoring and evaluation ranged from 15 to 30 minutes and occurred on a fortnightly basis for the first 3 months. The PHM used time after work hours and on weekends to conduct educational programs and undertake planning meetings with the community health club.

The continued expansion of the project and time demands on community follow-up prompted the PHM to advocate for a greater degree of flexibility in completing routine clinics from her supervisors (MOH and supervising PHM). Whilst the supervisors supported her efforts, they also insisted she maintain the usual clinic schedule. Although this placed time pressures on the PHM to continue to manage the development of the intervention, she was able to manage her clinic schedule. c. Educational resources and technical support: The PHM had liaised with the District Child Rights Protection Officer responsible for child protection and the Health Education Officer (HEO) of the district responsible for health education to share community development strategies and obtain advice on teaching materials and methods for NCD prevention. She also obtained the support from a senior clinician working at the District General Hospital who contributed time to deliver health education talks (on two occasions) at the village health club.

There was a limited amount of health educational resources on NCD prevention (consisting of four leaflets and a poster) developed by the Ministry of Health in 2008, which the PHM

Page | 173

obtained from the HEO. The PHM expressed frustration in not having access to more educational resources12. d. Human resource/technical contribution: The PHM is supported by two health volunteers from the village who assist her with organising the logistics for undertaking MCH outreach clinics and organising community meetings for the NCD intervention program.

Stakeholders involved in the intervention and their responsibilities was also mapped:

Stakeholder Stakeholder type and role: domain: a. Health PHM-Responsible for conceiving and formulating intervention. Involved communities system in the implementation, monitoring and further enhancement of the intervention via an iterative process. MOH-Provided initial endorsement for the PHM to undertake program. No direct involvement in planning, support or assessment of intervention. HEO-Assisted PHM in providing available (although out-dated) health education materials relating to NCD prevention and control. Acted as a resource provider. b. Community Families attending MCH clinics, health volunteers and community health club members actively participated and worked with the PHM to develop intervention. c. Non-health Child rights protection officer acted as resource provider. Grama Niladari (village sector leader) promoted program within constituents.

5.7.6 Monitoring and evaluation strategies

Assessment of salt, sugar and oil consumption was measured through weekly tracking sheets. The PHM, in consultation with community members, is planning to standardize the reporting formats in a common template. The PHM indicated she was also in the process of lobbying health authorities to obtain donor support to provide plastic measuring cups and scales to participating households. Such measures, she explained, are essential for evidence-informed advocacy: “Moving forward, if we are to make this known to others, we need better science around the measurements people do”.

12 With the consent of the PHM, MOH and RDHS, I lobbied relevant central level authorities to provide the PHM and RDHS with recently developed educational materials in the form of flip-charts, posters and leaflets on NCD prevention. These materials were prepared by the Director of Policy and Planning, HEB and the Directorate of NCD in 2014. Page | 174

Tools to explore the CP dynamics in the described intervention

5.7.7 Applying the project cycle model

The classic project cycle was used to map the points at which CP was manifested in the development of the intervention (Figure 27). While the initial ideal for the program emerged from a need identified by the PHM, the community was actively involved in piloting the intervention, planning strategies and promoting partnerships, and was integrally involved in providing feedback via tracking sheets and feedback sessions. CP emerged in Steps 2, 3, 4 and 5 (shaded areas).

Figure 27. Identifying manifestations of CP based on the project cycle model

5.7.8 Applying Rifkin’s Spidergram model The specific methodology of the Spidergram described by Rifkin (2007) for CP in nutrition programmes was used as these were directly relevant to intervention goals (387). This version of the model involved exploring gender dimensions of participation, as the role both men and women play a critical role within community based nutrition. Figure 28 presents the Spidergram for case-study 1.

 Planning and management. Community and family ownership was fostered with participants’ own ideas, time and investments. The PHM respected community views and

Page | 175

contributions, and created a culture of shared responsibility for the new phase of program.

 Gender. Whilst the program had its origins in an MCH program, there was participation from both males and females in implementing the intervention at the household level. The intervention was designed through a family centred approach, and the female head of household did take greater responsibility for monitoring since they had greater control over food preparation and purchase. The PHM provided specific information and attention to catering for the dietary needs of pregnant and lactating mothers or those with chronic diseases. No specific gendered aspect was considered.

 Resource mobilization and support for program development and sustainability. The relatively low resource costs in implementing the intervention at a household level enhances scope for sustainability. The PHM indicated that any scaling up of the intervention to larger catchment areas would require significant time and financial investments and administrative support from the health system. If this was not forthcoming, she indicated that the members of the local community health club remained a platform to harness and negotiate funding for community development activities. It was clear that the enabling environment that was created provided confidence to the PHM that resources could be obtained. There was, as indicated, encouragement to diversify community self-help efforts.

 Leadership from the community. Although the PHM was the main catalyst in formulating the intervention, community members worked intensively with the PHM in its development process. PHM built partnerships with the community and provided them with opportunities to enhance the program.

 Monitoring and evaluation. As discussed earlier, the community is integrally involved in assessing and analysing intervention effects and the PHM has mechanisms to respond to requests. The current modus operandi appears to maintain program effectiveness and provide future sustainability.

Page | 176

Figure 28. Spidergram for Case-study 1 (Score 1 = narrow CP; Score 5 = wide CP)

5.7.9 Applying the P-Compass tool

The NCD prevention intervention was driven in close partnership and collaboration with communities, where communities played an integral role in leading the intervention processes. A very high level of CP and collective action was registered in the P-Compass (shaded area):

Score Level of participation Degree of participation Participants action

4 Very high Collective action Has control “By community” 3 High Co-learning Has delegated authority 3 High Co-operation Plans jointly “With community” 2 Moderate Consultation Advises “For community” 2 Moderate Compliance Is consulted 1 Low Co-option Receives information “On community” 0 none none none

Page | 177

5.8 Case study 2: Family nutrition program

5.8.1 Setting: A PHM catchment area in Nuwara Eliya District

5.8.2 Background

By observing anthropometric data obtained through the Child Health Development Records (CHDR), the PHM noted that a growing number of children under 5 years of age in her catchment area were malnourished or at risk of malnutrition. She also noted a growing number of children who were obese. This was also true for many women that attended her midwifery clinic. The PHM hypothesized that the underlying issue may not be due to lack of access to healthy vegetables, protein and fruits13, but may stem from poor dietary habits and sedentary lifestyles. She decided to explore opportunities to develop a community health intervention to promote ‘family nutrition’. Empowering families to prepare nutritious meals was not normally an activity undertaken by PHMs within routine programs. Enhancing child development was articulated as an underlying intervention goal for the PHM: “child nutrition is so important… something like eighty-precent of brain development happens during this time before 5 years. Shame on us all if we fail our children”.

She began discussing with mothers attending the MCH clinics and living in her PHM catchment area the types of meals they prepared at home. She found that many relied on food prepared at roadside food and beverage outlets or ‘fast-food’ stands in the village. When preparing food at home, “many mothers simply took a small portion of vegetables in their diet”, and when they did cook vegetables “they added huge quantities of coconut oil and over cooked them to add more taste”.

The PHM first discussed the idea of establishing the intervention with mothers attending her MCH clinic and then with community members when conducting household visits. Many suggested collaborating with the local preschool to provide a space in which to ‘anchor’ the intervention. The PHM met with the district educational authority to obtain formal endorsement of the project.

The community-based nutritional intervention that was established had two components:

13 Nuwara Eliya District is well known in Sri Lanka for its fresh fruit and vegetables due to fertile soil and environment. Page | 178

 Component 1 – involved preparing nutritious meals for children at the preschool premises with the participation of mothers and schoolteachers. The intervention aimed at empowering mothers with nutritional education, healthy eating and food preparation skills. Creative methods of improving children’s knowledge and understanding of health food types, such as promoting consumption of locally produced vegetables and fruits of different colours, were embedded within this process.  Component 2 - engaging adult members of households, including adult males, in nutrition education and methods to prepare traditional food in a “healthy yet tasty” format.

5.8.3 A short film I developed on this case study (12 mins duration) encompassing elements of background, participant profiles, evolution of the intervention and challenges faced can be viewed online at: http://case-studies.strikingly.com

5.8.4 Intervention impact and challenges

No formal evaluation of the intervention had been undertaken. However, the PHM noted some evidence of child nutritional status improvement in some children from participating households by assessing anthropometric data prior to and during the intervention period:

“I haven’t planned any scientific way to test [the intervention] using research, but I can show you by the children’s log books…see here [shows a copy of the child nutritional reports], you see how this child has achieved improvement... twenty percent improvement in weight gain, now you know that is big!”

A number of participating households had altered their family cooking preparation and diets as a result of the intervention. The healthy cooking practices were also adapted by male participants: “Only when I started this program and talked to others I realize how wrong I was in the way we had always cooked and was depending on lots of oil. Not having it changes meals. But now I have lots of ‘options’ and the recipes I learnt are done without losing taste!” [Husband, Household 2].

An important finding on the rationale for CP was the finding that many community members expressed their willingness and motivation to participate due to the trust and respect they had for the PHM:

“She has known my family for 3 years now, she helped me through my first pregnancy and now it’s my second baby. She is like a mother to us, we support her a lot in this program” [Mother, household 2]. Page | 179

The PHM felt that her knowledge of NCD prevention and control was limited and should be updated. She also sought training on home-based management of chronic disease and updated knowledge of drug treatments for cardiovascular disease and diabetes.

5.8.5 Resources used for the family nutrition intervention

a. Financial resources: Each family contributed a nominal amount of 50 rupees per month to be used to purchase vegetable oil, protein and consumables used for the communal cooking intervention. All fruit and vegetables needed for cooking were sourced from the home gardens of each participating community member and brought to the school. This represented a community driven co-operative approach to meeting financial costs. b. Educational and materials resources: The PHM felt that there were no available resource guides at district level on ‘Community programs for improving nutrition’. With the support of communities the PHM is now compiling a ‘healthy recipes’ book. c. Time: The school intervention took an average of 3 hours to complete. The PHM visited the households of participating families to assess food basket and meal preparation at least once per month. She stated that managing the intervention had not significantly affected her routine work schedules with field monitoring aspects integrated with household visits: “...time is not an excuse to me…I can manage this intervention and my other midwife activities together”. However, she suggested a reduction of PHM catchment areas would result in greater opportunities and more intensive community development opportunities: “If I had a smaller area with a smaller population, I can you know, go much deeper to community health issues with them…they have lots of ideas, what limits me from doing more is this large catchment area”.

Page | 180

d. Human resource/technical contribution: Stakeholder mapping showed that by harnessing the support of community, teachers and a health volunteer, the PHM was able to catalyse the intervention.

Stake holder Stakeholder type and role: type: a. Health PHM - Responsible for conceiving and formulating intervention. Communities were system engaged in an iterative manner for program implementation and further expansion via participatory processes. b. Health volunteer from community assisted PHM in undertaking communication and Community administrative support for the intervention; assisted with MCH outreach activities. Mothers of the children of preschool –both beneficiaries and participants to the intervention that actively worked with PHM to implement/develop program. Preschool children- beneficiaries of the intervention Male and female heads of households from Families participating in Component 3 of the intervention c. Non-health Preschool teacher- assisted the implementation phases of component 2 of sector intervention by mobilizing mothers and pre-school children, providing school resources such as space and kitchen area for cooking intervention and sourcing local produce/food items for the intervention. Child rights protection officer- supervising overall activities of the preschool and enabling legal support to PHM to access school.

Tools to explore the CP dynamics of the family nutrition intervention

5.8.6 Applying the project cycle model

While the program emerged from a need identified by the PHM, the community was integrally involved the in programme implementation, resourcing and development phases. Therefore CP emerged across Steps 2, 3 and 4. This manifestation of CP within the project cycle differed from the previous case-study where communities jointly contributed in formative phases, right through to the monitoring and evaluation phase (Figure 29).

Page | 181

Figure 29. Identifying the manifestation of CP based on the project cycle model

5.8.7 Applying Rifkin’s Spidergram model

 Planning and management. The PHM was the catalyst for the program and conceptualized the overall strategy. However, the intervention was designed via CP approaches. The involvement of communities was harnessed during the implementation phase rather than in the formative development phases (Figure 30).

 Gender. Engaging mothers was central to the CP approach adopted in the intervention. Male participation was minimal.

 Resource mobilization and support for program development and sustainability. Resource mobilization was ranked at 5. The community’s sustainable contribution of time, material, human and food resources to develop and maintain the intervention is crucial to its ongoing success.

 Monitoring and evaluation. The intervention lacked a robust monitoring and evaluation mechanism and participants had no role in engaging with evaluative practice with the PHM.

Page | 182

 Leadership from the community. Although the PHM remained a driving catalytic force at the start of the intervention, as we continued to profile the intervention, the empowered community group at the pre-school sustained the intervention. The PHM enabled community leadership capacities with some community members taking ownership of key aspects of the program: “We will take this program forward for sure…I mean even if madam is not here, we know how to do it now” [Mother, Household 4].

Figure 30. Spidergram for Family Nutrition Program Case-study 5.8.8 Applying the P-Compass tool

A high level of CP and collective action was registered in the P-Compass (shaded area).

Score Level of participation Degree of participation Participants action

Section summary

The two case studies demonstrate that meaningful engagement with communities can build their individual capacity as well as collective capacity (within family units and community health networks) to address risk factors for chronic diseases and promote health. Community members worked with the PHM to refine the intervention strategy after an initial awareness and training phase, and gradually increased their role and responsibility in facilitating the community health intervention over time. In some instances participating communities added their own innovative trajectories to the intervention. Both interventions also harnessed locally available and sustainable resources. No rigorous evaluation of the interventions has been conducted by health authorities.

Page | 184

5.8.9 Overview of focused ethnographic study results and reflections on methods

Understanding the context and conditions in which PHC workers operate is essential for any meaningful assessment of their capacities and scope to provide services and intervene within a particular health domain. The work settings and work practices for each PHC unit vary widely in relation to the structural, socio-economic and vulnerability gradients of the populations they serve.

The findings highlight a number of the structural and organisational characteristics that have an impact on the extent and ability of PHC units to meaningfully engage with communities. The importance is apparent of understanding the social context and conditions in which PHC workers operate before planning complex system reforms.

Adopting a focused ethnographic approach using a four-part process not only allowed for an immersive experience of documenting the varied work practices of PHC workers and community health interventions, but provided a ‘layered’ contextual understanding of CP.

Rather than defining what community-based interventions are, I aimed at capturing what PHC workers and administrators considered such interventions to be, and documented these through a focused ethnographic approach. Although time-consuming, this approach allowed exploration of a greater range of interventions within the routine practice of PHC workers.

It was clear from the observational results that almost all interventions shown by PHC workers to the field researcher and I as being ‘community-based’ were predominantly clinically oriented ones which focused on individual screening devoid of community participatory or health promoting approaches. Interventions centred primarily on screening interventions in which ‘community’ was defined as a population attending a nutritional assessment clinic or school setting for health assessment. Thus the community health intervention definition adopted by PHC workers largely relate to McLeroy’s (2003) ‘community as a setting’ model, and to Whitehead’s (2002) type VI and VII model, of interventions performed ‘on communities’ by a health authority within a defined service area. During the 4-week observational follow-up study of ten PHC workers (Part B), only two Page | 185

interventions were observed that adopted community participatory approaches in NCD prevention. These two interventions (examined in Part D as in-depth case studies), show truly meaningful collaboration and partnership between health care workers and community groups. The case studies that were also documented through video, reflect some of the innovative strategies adopted by the PHC workers and how they harnessed indigenous resources within local communities to initiate and sustain interventions to address NCD risk.

Even when opportunities to develop collaborative partnerships with community groups arose, time and resource commitments to clinical casework inhibited the development of such initiatives. In one example observed, community members (mainly young mothers) approached the PHM to be a resource provider to guide them in organising a community awareness program on alcohol abuse and domestic violence in their village. They were especially concerned about a rising level of domestic violence that had resulted in abuse of five children. The PHM assisted community members in organising a training program with participation from local police, religious leaders, medical officers and village elders. With an engaged community coalition she wanted to further develop the intervention from simply being an ‘awareness raising’ program to one which addressed underlying determinants, established community reporting mechanisms and promoted practical case-management skills for community volunteers. Despite formulating what seemed like a sophisticated and robust intervention model for that community, the PHM was unable to build upon its success due to the demands of maintaining her routine PHM clinic schedule. Advocacy with health administrators had also been ineffective in catalysing change. There were no formal support mechanisms instituted by district health authorities to stimulate or enable the development of CP approaches.

On other occasions there were ‘missed opportunities’ for CP by PHC workers. I made the following observation in my field notes after completing an observation of a PHM’s interaction with community members: ‘There is no doubt [about] the dedication she [PHM] has and the respect these participating families attending the clinic gives her... She gave a short 10 min health education talk on importance of exercise during pregnancy but also on gestational diabetes. There were definitely some confused faces in the room with the latter topic! Few women asked what they can practically do about it and whether there could be some sort of an exercise group or session organised with their participation. Many Page | 186

agreed. I felt this formed a strong entry point and opportunity to do some real community engagement!, but no, wasn’t really picked up or responded to – another missed opportunity?’ [31.2.0,Evernote:11.5.13:8.01pm].

Mandatory reporting on their clinic activities and data requirements for routine health surveillance activities consumed a large proportion of administrative ‘desk’ time of PHC workers. The standardized reporting formats were also used by MOHs in their performance review. There were no reporting nodes for community development or community health promotion within these standardized reporting formats.

A conference involving PHC workers chaired by the MOH of their area is held each month in order to ensure strategic planning of PHC activities (388). On attending four such conferences in both Nuwara-Eliya and Pollonaruwa districts, I found them to be mainly structured around logistical and statistical issues in conducting routine MCH and PHI activities. There were opportunities within the agenda for PHC workers to share insights and intervention ideas. During one conference, a PHI described the need to develop a community approach to addressing road traffic accidents after a number of fatalities involving school children in the locality. While many expressed their support for such a program, there were no real strategic-level decisions taken on following up and developing the CP intervention. No willingness was observed to provide structured support to incubate such ideas, invest in their development or provide critical debate on their strategy and effectiveness.

In conducting the observational and focused ethnographic studies, I found the PHC workers to be highly committed to their work, sacrificing their personal time on weekends to conduct community meetings or follow-up patient visits. As indicated in Table 12, many PHC workers contributed their personal funds toward fuel and transportation costs to successfully complete outreach clinics. If community based approaches are to be embedded or take root within routine practice, then PHC workers need to be better supported to ensure an ‘enabling intervention space’ with adequate provision of time and resources.

Page | 187

Study findings

Results of document analysis

Page | 188

Section outline

A key goal of Research Objective 1 was to examine the extent to which CP and community participatory approaches have been defined, described and contextualized within domestic health policies, legal documents and health sector master plans relevant to community health service provision in Sri Lanka. Documents relating to PHC workers’ roles and duty lists, health worker training curricula and monitoring reporting templates were also reviewed to examine how CP has been characterised and formulated within routine PHC practice. The insights derived from examining these documents will be useful to determine if there is an enabling domestic policy, legal, procedural environment, as well as adequate worker training and monitoring mechanisms to support CP engagement, and for undertaking community based NCD prevention and control programs.

5.9.1 Domestic policy and legal framework analysis

A total of seventeen national policies; eight health sector strategic planning documents; thirty-one domestic legal frameworks; and the governments’ strategic governance plan were identified as having direct relevance to addressing health care needs and services for Sri Lankans (Annexure 7).

The chronological timeline of the emergence of each health policy in Sri Lanka was plotted alongside global policy frameworks that promoted community participation and inter- sectoral action in health (Figure 30). Sri Lankan health authorities have been proactive in adopting and aligning domestic health policy frameworks with global health agendas since the Alma Ata Declaration on PHC in 1978.

The vast majority (90%) of policy instruments pertaining to health care in the Sri Lanka explicitly articulated and espoused the importance of CP and community engagement within their policy goals and objectives. The only two policies that did not espouse a role for communities was the National Oral Health Policy for Sri Lanka of 2012 and the National Medicinal Drug Policy for Sri Lanka of 2005.

Page | 189

In undertaking any policy mapping exercise it is important to recognise the various types of policies that exist and their intended purpose (389). Some policies such as the Health Promotion Policy of 2010 offer a formulary that may be classified as ‘broad statements of intent’, whilst others such as the National Nutrition Policy are ‘prescriptive’ in achieving policy targets and setting goals.

Although most policies defined their scope and objectives, only a few outlined the policies’ implementation processes. For instance, the 2008 National Nutrition Policy of Sri Lanka elaborated the use of CP as part of a “settings based approach”. However there was no elaboration on the meaning, purpose and rationale of such an approach within the main body of the policy document, or within the reference documents and technical guides that accompanied the policy. In contrast, there were clear articulations of the mechanism and rationale for CP in other policy documents. For instance, clause 2.1 of Sri Lanka’s National Mental Health Policy detailed the importance of creating of an enabling environment for community engagement to improve mental health service provision and highlighted the expected manifestation at community level: “Each district will have the following network of services: a community support centre for every MOH area, a community resource centre, and other community based care appropriate to the needs of local communities to ensure their participation”. Such definitions of CP and how they are framed within policy documents are summarized in Table 14.

The aspiration and expectation of each policy in terms of engaging communities and the context to which participation was articulated is presented in Table 14. The results revealed that the majority (52%) of policies articulated CP as a means of achieving policy goals. Five policies emphasized CP within the service planning phase (28%) and three policies (17%) described the utility of engaging communities when evaluating policy intervention efforts. Only one policy, the National Health Promotion Policy, articulated the need to ensure community involvement within the ‘strategic planning’ phases of policy programs and actions. The policy suggested that the health system should ‘enable’ networks of community based volunteers to “take the leading role in promoting health in their settings”. The policy also advocates for dedicated budgetary allocations to ensure participatory planning of interventions. From these results it appears that the predominant ‘policy expectation’ within Sri Lanka’s health policy landscape is one that values CP within health care services delivery in which communities participate primarily to assist health workers in program Page | 190

implementation. Analysis of health policy frameworks reveals the rationale for CP as being largely ‘technocratic’, one that views participation as a means of ensuring successful interventions and through involving ‘beneficiaries’ in implementation.

In analysing policy formulation, a lack of policy coherence was evident. Policy coherence can be defined as a policy whose objectives are consistent and attuned to objectives pursued within various elements of the stipulated policy. For instance, how policy goals, implementation measures, monitoring indicators, financing and resource allocations are calibrated and internally consistent within a policy document, and externally consistent with other health policies domestically (390). None of the policy documents indicated any specific program indicators targeting CP in health within monitoring frameworks, and only a few promoted the need for general policy accountability frameworks. For instance, the National NCD policy which articulated periodic reviews and establishment of result based evaluation systems, even though specific measures on CP were not discussed. The National Policy on Tobacco and Alcohol (2007), emphasized how the policy will be evaluated and financed and what measures would be taken to monitor its implementation. More critically, the need for practice evaluation and the technical competencies and resource allocations needed to catalyse CP for the achievement of health policy goals have not been explicitly indicated in the 17 policies analysed.

Page | 191

Figure 31. Sri Lanka health policy timeline in relation to key global health policy developments

Page | 192

Table 14. Policies articulating CP approaches as way of meeting policy goals

Health policy & Policy statements relating to CP as way of meeting Role of relevant MoH policy goals and objectives community or authority responsible consumers for policy articulated in coordination policy Charter for Health  ‘To promote community participation and  service planning Development (1980)* involvement’ (core function 5)  service delivery (Prime Minister and  ‘To create greater awareness among people of Health Minister, under the importance of health’ (core function 4). the National Health  The action plan which accompanied the NHC Council) iterated the health system to ‘empower communities towards more active participation in maintaining their health’ National Health Policy  Clause 10. “Ministry of Health will promote the  service planning of Sri Lanka, Ministry involvement of the community in health care”. (limited of Health (2002) information (Health Minister and provided in National Health policy executive) document). Sri Lanka National  Policy objective: “To create nationwide health  strategic Health Promotion promotion actions by mobilizing and empowering planning Policy (Draft 2010) communities toward active participation in  service planning (Directorate of the comprehensive health promotion continuously  service delivery Health Education throughout the life course”. Bureau)  Clause 2.2: “Strengthen health volunteer system and all kinds of groups and networks in the society to take the leading role in promoting health in their settings”  Clause 2.3: “Periodically organise national campaigns to motivate and maintain countrywide continuity of health promotion awareness”. National Policy on  Clause (l) “encourages active community  service delivery Tobacco and Alcohol participation in the implementation and  service (2007) monitoring of the National Policy on Tobacco and evaluation Alcohol”. National Mental  Policy Principle 3: “To provide services that will  service delivery Health policy (2005) be organised at community level with community, (Directorate of Mental family and consumer participation” Health, National Institute of Mental Health) National Maternal  Strategies for achieving Goal 5c) “Foster  service delivery and Child Health community empowerment and mobilization to policy (2005) sustain conducive behaviours in support of (Directorate of the Maternal and Child Health”. Family Health Bureau) National Policy for  Goal (iv) “Empower the community for promotion  service delivery Prevention and of healthy lifestyle for NCD prevention and Control of Chronic Page | 193

Non-communicable control. Models and mechanisms will be Diseases (2010) developed to empower communities to ensure (Directorate NCDs) their participation in multi-sectoral activities related to health promotion, and NCD prevention and control. In order to develop healthy communities, the following measures will be taken”. National Nutrition  Policy Statement 5.2.3: “Community  service delivery Policy of Sri Lanka Empowerment. Empower the community by  service (2008) reorganising grass root level community evaluation (Directorate of the organisations, in planning, implementation and Family Health Bureau) monitoring of nutrition intervention programs”. National Policy And  Guiding Principles: “A multi-sectoral response is  service planning Strategy on Health of required to address their needs and meaningful  service delivery the Young (2011) participation in arriving at decisions related to  service (Directorate of Youth, economic and social development of the country. evaluation Elderly, Disabled and In major activity areas: c. Ensure integration of Displaced persons) health of young people into existing health services and programmes. f. Promote active participation of and leadership by young people in planning, implementation and monitoring of programmes” National e-Health  Section 01.01.09: “Sub-systems and module  service planning Policy of Sri Lanka developers will ensure a participatory approach (2013) that allows relevant stakeholders (and end users) (Directorate of Health to contribute to the development process”. Information)

Sri Lanka code for the  The code "recognises that families, communities,  service delivery promotion, protection women’s organisations, civil society groups and and support of breast other non-governmental organisations have a feeding and special role to plan in the promotion of marketing of breastfeeding and in ensuring the support designed products needed by mothers who are breast feeding". (2002)

Page | 194

5.9.2 Results of content review of PHC duty lists and training curriculum

PHC worker duty lists were obtained from the Family Health Bureau and Directorate of Policy and Planning at the MoH. The health worker training curriculum was obtained from the MoH’s National Training Institute in Kaluthara (2013).

In-depth analysis of duty lists, training curriculum and monitoring mechanisms of PHC workers revealed little articulation of community participatory practices within the scope of routine practices. There is also no explicit monitoring and reporting requirements for capturing community engagement or for community-based health promotion practice. Table 15 summarizes the content review of these documents.

Only the job duty list and training guide of the PHM defined a clear role in relation to NCD prevention and control within the scope of conducting ‘well woman clinics’ and for close monitoring of pregnant and ‘at-risk mothers’, women diagnosed with chronic diseases such as hypertension and diabetes. Well woman clinics are community-based clinics that target women over 35 years of age that provide screening for breast malignancies and cervical cancers. These are classified as ‘non-mandatory’ and are non-routinely performed. Well woman clinics have also been shown to have very low implementation status (391).

Page | 195

Table 15. Duties and work functions of PHC workers relevant to CP and NCD interventions

PHC worker Duties and work functions relevant to NCD Duties and work functions which category prevention & control stipulate community participatory approaches

Medical No specific duties assigned or articulated in relation The following work areas are defined Officer to NCDs. under the objectives of the MOH Office: of Health* •Health promotion: organizes and directs health education work; •Solicits inter-sectoral collaboration; •Solicits community participation. No details are provided on the actual tasks, procedures, monitoring mechanisms on these objectives.

Public Health Same as above. Duty list stipulates PHI: “shall plan and Inspector** implement a programme of health education in his area and ensure community participation in health activities"

Public Health General statement on addressing NCDs stated in PHM Guidelines state, “In addition to her Midwife*** Clause 4(x) of PHM training guide: “Shall assist the domiciliary care, she also participates in health team in the prevention and control of both the area clinics linking the community with communicable diseases & non-communicable the health care system. diseases”. PHM Programme Objective 9 states: “To PHM Guidelines highlight the importance of close enhance community participation for the monitoring of NCDs in at-risk mothers. For instance, promotion of school health activities”. in presence of Diabetes Mellitus home visits are to Other clauses relevant to CP: be scheduled ‘every two weeks’, where the PHM is  4.(xi) shall participate in community expected to “Check urine for sugar via dipstick tests, health programmes organized within check for Fetal Heart Syndrome, monitor maternal her area with the approval of the weight gain and link referral with specialist”. MOH/DHO.  4.(xii) Shall assist in the training The Guide also encourages the PHM to undertake programmes of field health staff and (non-mandatory) ‘Well woman clinics’ for women other voluntary health workers with over 35 years of age to screening for hypertension, the approval of the MOH. diabetes, breast malignancies and cervical cancers:  Clause 2.C. mandates all PHMs to “If any abnormality is detected clients should be “live in their area and have their referred to the health care system for necessary offices in the area”. The rationale for management”. The PHM is expected to record the this is also based on the training following data from such clinics: number of cervical requirements. cancer (pap smear slides) sent for lab investigation, breast examinations performed, positive notifications of these tests. Notes: *Department of Health Services (1999) Manual on the Management of Divisional Health Services Based on Primary Health Care, Sri Lanka. Part IV: Duties of The Medical Officer Of Health. **Duties And Responsibilities Of Public Health Inspector (Range) General Circular No: 29th September, 1989 ***Family Health Bureau, Maternal Care Package (2011) A Guide to Field Healthcare Workers. MoH Publications. p.62

Page | 196

5.9.3 Health Sector Master Plan (HMP)

Eight national health-related plans and/or strategic health sector development documents were examined, including the Health Sector Master Plan (HMP) for Sri Lanka (2007 to 2016). Adorned with the slogan ‘Healthy and shining island in the 21st century’, the HMP remains the key programmatic guide for the health sector. It aims to facilitate quality, equity, productivity and ensure better resource allocation of the health service by ‘responding to the people's needs by working in partnership with them’ (392). The overarching aim of the HMP is encapsulated in five strategies, which place community empowerment as their foremost goal: “empower communities (including households) towards more active participation in maintaining their health”. The rationale for patient participation as a major focus of enhancing rights, satisfaction, and quality of health care is also articulated in the HMP: “There has been a global awakening to the patient's right and equity. Looking at the characteristics of the disease itself, it is now well recognised that the patient’s participation and satisfaction are important in the success of treatment. Greater efforts are needed in educating patients as well as health service providers to make better choices. This calls for reorientation of people's cultural norms around health care in association with the following principles of improvement of Quality and Safety; securing of Patients’ Rights; and enhancement of Client Satisfaction” (HMP, Pillar 2: Responding to Patients' Expectations, Culture and Car. pp18-19).

5.9.4 Guidelines for strengthening health care at primary level

The strategic document "Approach and Guidelines for strengthening health care at primary level” was developed by the MoH’s Directorate of Policy and Planning (2012), and sets out an ambitious agenda to reform Sri Lanka’s health system (258). The strategic model and policy statements call for a reorientation of PHC to address the growing burden of NCDs and the need for community-based care. The document was formulated over a five-year period through multiple rounds of stakeholder consultation which involved scientific forums, policy workshops, field-level observational studies and pilot projects. A major thrust of the proposed reform strategies was for increased community participatory approaches at primary health care units: “there is a need for de-institutionalizing primary care and promoting community participation”. The rationale for CP is articulated as a means to ensure and enable NCD prevention and promotion interventions, and to support the evolution of an

Page | 197

effective ‘referral and back referral system’ between communities and the PHC system at village level (258). To ensure the realization of these goals, technical capacity building of health care workers and administrators, new tool kits for NCD screening, health promotion campaigns, professional linkages and enhanced budgetary facilities are proposed (259).

5.9.5 Domestic legal frameworks pertaining to community participation in Health

A nation’s constitution forms the supreme law of the country, and the inclusion of the right to health, with the concomitant right of participation, is central to the recognition of health- related human rights (393,394). Despite progressive public health reforms and policies, the ‘right to health’ is not enshrined as a fundamental right of a citizen in the Sri Lankan constitution (395). A review of 31 domestic legal documents presented in part D of Table 6.10 found none that specified, advocated or espoused the value of community participation in health.

Domestic laws do, however, stipulate decentralization of health sector governance to the provincial, district and village levels. The 13th Amendment of the Sri Lankan Constitution devolves major elements of health care service delivery to provincial and district-level government authorities, as stated in the constitution, to ‘promote more efficient administration by the local authorities in relation to public health’ (395). Health care is included as an ‘obligation’ devolved to provincial councils (394).

Within this devolved health sector governance framework, it may be assumed that sub- national governments’ proximity to their constituent communities may enable them to respond better to local needs, advocate for health finances and enable a better and more accountable information flow between citizens and local government. The process of decentralization can itself enhance the opportunities for participation by placing more power and resources at a closer, more familiar, more easily influenced level of government. However, despite the devolution of power to peripheral organs of the health system, health sector decision-making, program formulation, budgetary and resource allocation remain at central level (396).

Page | 198

5.9.6 An erosion of participatory space in health care decision making in Sri Lanka

In exploring Research Objective 1, I extended my analysis beyond examining policy and legal documents, to also conducting key-informant interviews and exploring the relevant literature, commentaries, and technical reports concerning Sri Lanka’s domestic health policy development framework. I was able to interview three former MoH Directors of Planning with a cumulative leadership span of fifteen years (two have subsequently retired). The case-study presented below (Box 5), was developed during the process of undertaking this research, and highlights the gradual erosion of participatory decision making within Sri Lanka’s health system. It reflects the decline in the stewardship apparatus within the health system to promote a joint decision-making space between relevant health system units and for engagement with communities in planning health interventions.

Box 5. Disbanding the National Health Development Network

As highlighted in Table 14 the Charter for Health Development embraced an inter-ministerial and inter-sectoral approach to health action in Sri Lanka. Chaired by the Prime Minister, the National Health Council (NHC) ensured a rigorous mechanism to ensure inter-sectoral participation in health action at the national, provincial and district levels. In the five core functions that governed the NHC the commitment to promote CP was paramount. Its core functions were highlighted as: ‘To create greater awareness among people of the importance of health’ (core function 4) and ‘To promote community participation and involvement’ (core function 5).

The establishment of the Nutrition Coordination Committees (NCCs) was an example of a practical community-based program emerging from the NHC, which embraced a participatory approach within the health system and at inter-sectoral level. The NCCs were established at village level to maintain essential food items in the market, facilitate food security and provide access to sufficient and quality food within reasonable price limits for consumers. At village level, community co-operatives were established to ensure food security, and health workers partnered with these community groups to ensure distribution to families with under-nourished children under 5 years of age. The work of the NHC program received international support and endorsement and was commended by WHO as a ‘good- practice model’ in espousing PHC values through meaningful strategies to achieve ‘health of

Page | 199

all’ (397), as well as being a model for inter-sectoral action and decentralized health governance (221).

In 2004, under the chairmanship of the then Prime Minister Mr Mahinda Rajapaksha, the NHC ceased all functions. There were no available reports or formal dissolution notifications to identify the rationale for cessation of this high-level coordination mechanism. Policy analysts indicated a shift in macro-level government policies in favour of more centralized approaches to governance since 2004 (398) (399) (400). In 2013, the Government of Sri Lanka reduced many of the powers vested in provincial health councils towards central level control by promulgating the 13th amendment of the constitution. This further eroded the decentralization of health services and participatory models in national health governance. In the current model of health governance, the Health Development Committee (HDC) is chaired by the Director General of Health Services, and attended by provincial (PDHS) and District (RDHS) heads of health across the country. Unlike the NHC, it is not an inter-sectoral forum and affords no opportunity for participation by other government agencies, civil society or stakeholder groups. A review of meeting minutes revealed that the HDC focused on addressing mainly administrative functions of the health system14.

Figure 32 depicts the disbanding of the National Health Development Network and the current governance structures for inter-sectoral health development in Sri Lanka. The graphic depicts the potential erosion of ‘participatory health care decision making’ in Sri Lanka. In the pre-2004 model the involvement of other sectors and partners are shown. Stakeholders in health also included community-based coalitions. Health workers from village, district and provincial levels could participate within national health care decision-making processes.

A 2012 report on public policies and their impact on health concluded that Sri Lanka did not, at that stage, have an ‘integrated governance tool’ which dealt with health in other public policies during formulation, implementation and evaluation at national level (401). The Presidential taskforce on Dengue prevention and control was cited as an example of an inter-sectoral collaborative model which not only involved an inter-ministerial process, but extended to village level coordination mechanisms. Participation from schools, businesses and civil society organizations was crucial for effective programme implementation. In the absence of an NHC, the report suggests there are little or no community level health coalitions. Those remaining operate as ‘fragmented committees and task-forces’ working independently and sometimes in competition with each other for resources and health coverage across national and village levels (401).

14 In 2013, the HDC met 3 times. An examination of the meeting minutes revealed the content focused primarily on administrative appointments, functions and health care financing. Page | 200

Figure 32. Disbanding the National Health Development Network and the current manifestation of governance for inter-sectoral health development in Sri Lanka

Page | 201

Despite the political rhetoric on decentralization of health sector governance, over the past decade provincial and district level governance structures have been excluded from meaningful engagement with central level health planners. There also appears to be an erosion of institutional mechanisms promoting inter-sectoral action in health. Pursuit of such strategies may lead to a disabling environment for citizen engagement in health services, and disenfranchise meaningful efforts at CP practice.

Whilst this thesis explicitly focused on exploring the dynamic of CP between the health system and communities, there also appeared to be a lack of a participatory culture within the health system. With the disbanding of the National Health Development Network, no formal collaborative planning mechanisms have been established between primary care providers and health administrators at provincial and central levels.

A Participation in Health model was constructed to highlight the above described dynamics of participation that may exist within a health system and between a health system and communities (Figure 33). Inter-sectoral action between the health system and other government sectors such as education, are indicated by the red arrow. The extent to which various units of the health system coordinate with each other in developing health programming (Intra-sectoral coordination) is depicted as the black region in middle of Figure 33. Participatory approaches at the nexus between PHC units and communities are indicated as the black and orange regions.

Figure 33. Participatory dynamics within and between the health system. Analysis of interactions for inter-sectoral action, participation within units of health system and between health system and communities usually via the PHC system

Page | 202

As highlighted earlier (sections 2.1.10 and 2.3.2), the force of decentralization and devolution of health sector governance forms a crucial element influencing CP in health (170,193,194,196). Organisational commitment and devolved decision-making contributed to positive attitudes towards CP among health staff (191).

Decentralization has long been advocated as a desirable process for improving health systems (138). A key element of decentralization is the amount of choice and engagement that is transferred from central administration to institutions at the periphery of health systems (126). A missing link in literature on participation is the extent to which policy makers at central levels of the health bureaucracy engage within the various levels of the health system. The role and involvement of primary and secondary level care in facilitating specialist referral has been poorly examined. Such intra-health system coordination is an essential part of a decentralized health system. There is also evidence that if workers are not meaningfully engaged in health care decision-making, then worker motivation is adversely affected (402). Figure 33 indicates the participatory dynamics within units of the health system and between aspects of the health system and communities. Partnerships between health care units (‘inter-organisational working’) and within professional teams (termed ‘inter-professionality’ in health care) of a health system, has become a central plank of public policy in the UK, especially in the field of health and social care (403).

Section Summary

Overall, the results show that the theory and rhetoric of CP, especially at PHC level, are clearly articulated in health policy and planning documents in Sri Lanka. Despite this, however, there is little coherence in existing policies, health sector master plans and other domestic frameworks.

Decentralization and devolution of health services to provinces and districts are constitutionally enshrined, and provide a strong legal framework to enable participatory engagement within the health system. However the mode and articulation of such participatory engagement between the various levels of hierarchy within the health system for health service planning is unclear within the policy and planning discourse.

Page | 203

The ‘renewed’ Alma Ata of 2011 called for a re-commitment by Governments to promote CP approaches in health systems. The results of this section reveal a robust and enabling health policy environment for CP in Sri Lanka; this provides a range of entry points for the health system to utilize participatory approaches in health care service delivery. However, the realisation of these policy goals is inhibited by the lack of clarity within the functional arms of curative and preventive health services across the central, provincial and district health governance structures.

Page | 204

Study findings

Results of the survey of policy makers

Page | 205

5.10 Results of the survey of policy makers

The survey examined the extent to which communities, PHC workers and district level health authorities have been involved in health policy development processes in Sri Lanka.

5.10.1 Demographic characteristics

The criteria for eligibility was defined as those having direct involvement in the process of health policy formulation in Sri Lanka. Thirty-four completed surveys were received. Four survey response forms were excluded due to incomplete data across more than half the question domains. The respondents were predominantly male (72%), and are long serving members of the health system, with an average of 21 years of professional service (Table 16). The majority were employed within government agencies (65%) at the time of survey, with the largest representation from the Ministry of Health.

Table 16. Demographic characteristics

Characteristic Response Count % Sex Female 9 28% Male 21 72% Number of years in professional service <10 years 2 7%

10 to 20 years 12 40% >20 years 16 53% Average 21 Current post/employment setting Government (Ministry of Health) 12 41% Government (other than Ministry of Health) 7 24% Non Governmental Organisation (international and local) 5 17% Private sector 1 3% Academic 5 17%

Page | 206

5.10.2 CP in the policy making process

The majority of policy makers (93%) valued CP and supported the need to involve communities in health policy-making. However, less than half of all policies (42%) had formally engaged communities during the policy making process (Table 17).

The most common method employed to harness CP was by undertaking community feedback surveys (18%), facilitating the participation of community representatives in policy advisory committees (14%), and by undertaking research on community perceptions (11%).

The three leading methods chosen by the policy makers to raise community awareness of proposed health policy development processes was through newspaper articles (27%) and through official notifications by government circulars presented on designated public information days (25%).

5.10.3 Results of P-Compass and PPD tool analysis

Respondents were asked to reflect on their experience of health policy making and to rate the overall degree to which communities were engaged within each policy making process (Figure 34). The results revealed that the most dominant form of participation in policymaking is where communities are ‘consulted’ for policy development (38%), with only 18% of all policies adopting modes of community collaboration and cooperation.

Over half (55%) of all health policies were developed by central authorities without any involvement from provincial, district or PHC units of the health system (Figure 35). According to survey respondents, only 3% of all health policy development processes had engaged PHC workers. These results reaffirm previous findings indicating a disparity in grassroots health worker involvement and participation in health policy formulation and agenda setting in Sri Lanka.

Page | 207

Table 17. CP in the policy making process.

Yes No a. Were communities or patients consulted as part of the policy making process?* 21 (42%) 29 (58%) b. Were non-governmental stakeholders consulted as part of the policy 40 13 making process?** (75.5%) (24.5%) c. How were the communities made aware of the policy making N % process? (n=108 responses) Newspapers 29 27% Television 3 3% Radio 4 4% Word of mouth 9 8% Postal letter 8 7% Gov. admin. system notice (public info. days) 27 25% Media briefings 9 8% Public meetings 15 14% Internet/e-portal 4 4% d. How was community participation enabled? (n=91 responses) Community representation on policy advisory committees 13 14% Group discussions with communities 7 8% Community feedback surveys 16 18% Community consultative forums/ debates 11 12% Written public submissions via regular mail 13 14% Written public to submission via online portal 5 5% Through NGOs facilitating participation 2 2% Community participation was not enabled 11 12% By undertaking community based research 3 3% Don’t know 10 11% e. What value do you place on involving communities in health policy making? (n = 30 respondents) Very important to engage communities 15 50% Important to engage communities 13 43% Makes no difference to engage communities 1 3% Not very important to engage communities 0 0% Unnecessary to engage communities 1 3% * Non-response: 4, ** Non-response: 2

Page | 208

Figure 34. P-Compass analysis - Plotting the level of community participation in health policy making

Page | 209

Figure 35. Participation in Policy Development Tool (PPD) analysis - survey of policy makers

Participatory Participatory approaches within the various units of health system for health policy making typology* Type A Central level 23% administrators (CL) Type B 32% CL + TE Type C Provincial 9% CL + TE  level (PL) Type D PL District 24% CL + TE   Level (DL) Type E  PL  DL  Medical officer CL + TE of Health (MOH) 9% level Type F  PL  DL  MOH  Primary CL + TE Health Care 3% workers Note: A series of statements were constructed to elicit a participation typology. Type A: “The policy process was developed and formulated entirely by Central level authorities without participation from any level/actors of health system” Type B: “The policy process was developed and formulated by Central level authorities with input of selected technical experts such as leading academics” Type C: “The policy process was developed and formulated by Type B, including participation from Provincial level health administrators” Type D: “The policy process was developed and formulated by Type C, including participation from District level health administrators and program managers” Type E: “The policy process was developed and formulated by Type D, including participation from MOH officers” Type F: “The policy process was developed and formulated by Type E, including participation from PHC workers”

Page | 210

Section summary

Despite the importance policy makers placed on CP, the majority of health policy development pathways had minimal involvement from communities. A key theme to emerge from the exploration of CP in health systems is that policy makers and planners need to examine not just the interactions between the health system and community, but also assess the dialogue and engagement within the health system.

The survey results showed that only 3% of health policy formulation processes had actively involved PHC workers. The extent of participatory engagement between ‘front-line’ PHC workers and higher tiers of health governance has been relatively neglected in the literature surrounding CP and health. Health system analysts argue that health worker motivation has enormous influence in effecting changes in service delivery, organisational culture, reporting structures, human resource management, channels of accountability and types of interactions with clients and communities (404) (405). Indeed, the reverse is also true - consultation, collaboration and inclusion of PHC workers in health care planning and decision making enhances their motivation at the same time.

211 | P a g e

Chapter 6: ANALYTICAL FRAMEWORK SYNTHESIS AND DISCUSSION

Image 6. Children participating in a community based nutritional program at a pre-school setting

212 | P a g e

6.1 Chapter outline

The aim of this chapter is to examine and present the key findings in relation to the research questions and show how they contribute to the formation of a composite health system framework to better understand community participation within routine primary health care practice. This is followed by a discussion of the implications for policy and practice and summary comments regarding study rigor and its strengths and limitations.

Section 6.1 Answering the research questions - presents the main findings in relation to the objectives and research questions and relates these new insights to what was already known regarding these issues.

Section 6.2 Bringing it all together - synthesis of the analytical framework to explore CP using a complex health systems approach.

Section 6.3 Model elements and dynamics – The composition of each domain and the linkages between the factors within each domain are elucidated in this section.

Section 6.4 Conclusions - Summarizes the key conclusions and discusses their significance in relation to current PHC reform processes and efforts to combat NCDs. Implications for policy and practice and recommendations for further research is also presented.

213 | P a g e

6.1.1 Answering the research questions

This thesis sought to explore the knowledge, experience and perceptions of field-level primary health care workers, health administrators and policy makers toward CP at the PHC level, to examine its utility as a strategy in tackling the growing NCD burden, and assess the extent to which participatory approaches emerge within routine PHC practice. The factors that limit, enable or enhance meaningful CP at PHC level are synthesized within a complex health systems model.

The primary research question that was deconstructed into four inter-related research objectives are addressed here, drawing on the evidence generated by the multiple method research study. Each research objective is addressed in turn, within a discrete text box.

Health policy landscape and policy making processes

Stimulated in part by the historical declaration of ‘health for all’ in 1978, health authorities in Sri Lanka have consistently demonstrated policy commitment towards CP approaches in health. The MoH has been proactive in adopting global health frameworks emerging from forums such as the World Health Assembly that promote the integration of CP into PHC, as well as health promotion and inter-sectoral action in health in domestic health policy. A chronological mapping of the emergence of national health policies in Sri Lanka and analysis of their policy content showed close alignment with global health frameworks.

The concept of CP in health is clearly articulated and its value firmly enshrined in the majority (88%) of policy documents governing health service provision in Sri Lanka. The National Health Sector Master Plan, which sets the government’s priorities in improving the health of all Sri Lankan citizens, explicitly articulates the goal of empowering communities towards more active participation in maintaining their health. Within the 2010 National

214 | P a g e

Guidelines for Quality improvement of Health care, CP is viewed as “paramount to the annual action and monitoring plans for all programs and directorates of the Ministry of health” (388). Findings of the policy document analysis revealed the need to ‘empower communities’ as a core policy directive across many domestic health policy documents. Indeed, the results revealed that fewer than half of all health policies emphasized CP as a means of achieving their policy goals. The ‘policy expectation’ and rationale for ushering in CP was that the process should be initiated and led by health professionals, with communities supporting workers at the ‘service delivery’ phase.

The policy maker survey revealed that an overwhelming majority of policy makers (93%) valued CP and supported the involvement of communities in health decision-making processes. A third (33%) of all Sri Lanka’s health sector policies had also engaged communities in the policy making process, although participation was mainly facilitated through ‘consultative’ modes, research surveys (18%) and by having civil society representatives on the policy advisory committees (14%).

Although CP is enshrined as an integral part of the fulfilment of many health policy goals and objectives, its definition, scope and practice are not clearly defined and remain ambiguous. There are no guiding frameworks describing strategies for community engagement, nor efforts made by health administrators to systematically document community-based interventions at the PHC level into a database or repository. For CP practice to exist, an enabling programmatic environment is needed with adequate resourcing and time-allocations within PHC work plans. In the absence of such enabling factors, only innovative PHC workers and health administrators are likely to catalyse interventions by harnessing local resources.

However, as indicated in the case-study analysis, even when such innovative efforts do manifest, they are not nurtured and enhanced, nor scaled-up to other settings by health authorities. The interview findings indicated that the existing health stewardship culture at PHC level provides a limited platform for worker engagement with senior management structures to enable health care decision-making with respect to PHC intervention design and planning. Inculcating a culture of CP appears challenging when collaborative decision-

215 | P a g e

making processes within the fabric of the health system remain elusive.

Policies such as the National NCD policy promote population screening as a means of community empowerment and place little emphasis on promoting healthy lifestyles at the community level. A primary goal of the National NCD Directorate’s ‘healthy life style’ program, which has received financial support from WHO, is centred around NCD prevention based on undertaking population wide health examinations. ‘Intervention success,’ as defined by the program, is measured by the ‘total number of persons in the community screened’. There is limited articulation of the importance of working in meaningful partnerships with local communities to address NCD risk factors through community development processes.

Other parts of the National NCD action plan that espouse the importance of addressing the social determinants of health linked to NCDs are unrealistic in relation to current health system capacities and resource constraints. For instance, in Strategic goal 4, objective 2.2 the action plan suggests that PHC units seek to work with other partners to directly minimize health inequalities by engaging in “village level poverty-alleviation initiatives”. This involves intensive community development action and partnership with poverty alleviation actors. Whilst this approach represents a rigorous effort to address the underlying social determinants of poverty and ill health, they represent actions that are unrealistic given the limited time and resources of PHC units.

Analysis of the health policy mosaic also revealed a lack of clarity regarding the role and coordination of the curative and preventive arms of the health system in addressing NCDs, especially at PHC level; and the role, if any, of private sector health institutions.

Domestic legal frameworks and policy making processes

Decentralization and devolution of public health decision-making is constitutionally enshrined. Sri Lanka’s decentralized health system allows for relative autonomy in health service decision making at provincial, district and village (divisional) levels, and is primed to support the fulfilment of local community aspirations and needs. Despite this, health sector decision

216 | P a g e

making, program formulation, budgetary and resource allocation remain concentrated at central level. It appears that the prevailing political climate does not support such devolution of health sector governance to provincial, district and village level authorities. As indicated by the policy maker survey and the group interviews, there is also limited engagement of health professionals at the PHC level in the design and delivery of community-based health programs. Only 3% of health policies and planning documents involved field-level PHC workers in their formulation.

There has, in recent years, been an erosion of the intra-sectoral coordination mechanisms within Sri Lanka’s health system, with the disbanding of a number of key institutional mechanisms established to promote participatory engagement of village-level actors in health care decision making processes. This includes the dissolution of the National Health Council (NHC) and the disenfranchisement of the Health Development Committees (HDCs) and other civil society groups at village level.

CP as a priority strategy for PHC practice is not embedded in existing PHC worker job duty lists and monitoring and evaluation requirements. Operational procedures and routine work plans do not emphasize the need for health administrators and workers to engage in health promotion strategies for NCD prevention and control at the community level.

Senior managers interviewed also cited external factors such as the resistance from health worker unions towards structural reform of PHC worker roles and revitalization of job functions. Some elements of the existing job functions also do not reflect current health needs and priorities. For instance, despite nearly all births (99%) taking place in a hospital or maternity home setting (406), PHMs continue to be trained in undertaking home-deliveries. The qualitative data revealed the reluctance and concern expressed by some district health administrators and departments within the MoH (such as the Family Health Bureau) to redefine the role of the PHM for fear of ‘losing the traditional and symbolic focus’ on MCH (see section 5.2.8). Ongoing disputes between health worker unions with the MoH, and between the unions themselves, about ‘encroachment’ on traditional worker roles, may engender resistance to reform (407).

217 | P a g e

The cumulative effect of these factors is that the design, development and evaluation of community health interventions take place in a ‘top-down’ fashion. Innovations that emerge from PHC levels are also not readily absorbed or supported by higher-level structures, thereby inhibiting the development and expansion of such CP approaches.

If health professionals do not value CP, then one may argue that there may be little scope for participatory approaches to emerge within routine PHC practice. Identifying the values, perceptions and commitment of health workers, planners and policy makers is fundamental to any effort by health authorities to revitalize CP within the health system. The perspectives and assumptions around CP of health decision makers and grass-root level health workers are likely to influence or undermine the ability to manifest or realize CP within routine practice settings and affect the contribution made by communities (408-410).

The rationale for CP in health is viewed by almost all PHC workers and administrators through a technocratic lens. They argue that participation is of value when it serves to fulfil health program goals and assists workers in program implementation processes and to achieve public health outcomes. ‘Community’ also becomes defined within the technocratic paradigm as the potential beneficiaries and consumers of the health system living in a population catchment within a defined PHC service coverage area. The PHC units are therefore viewed as a conduit for delivering PHC services to the health consumers at village level. Such a technocratic approach may also be less-intensive (with minimal distribution of power and influence required) than an approach which views CP as a process for ensuring equitable reshaping of power and facilitating the empowerment of community members to actively participate in their own development (15).

The literature reviewed earlier identified a limited number of studies which examined how health professionals in statutory health organisations also viewed ‘community’ through a consumerist lens. Although it is still unclear how definitional understandings of community can directly influence or affect the delivery of services, Ryan (2001) argued that when statutory

218 | P a g e

organisations treat citizens as consumers it transforms the interactive space between a consumer and health provider to a “passive commercial transaction” rather than an interactive and meaningful engagement (411).

Despite the overwhelming majority of Sri Lankan policy makers (93%) valuing CP in health, my research findings revealed that less than half (42%) of all health policy development pathways had ever involved communities. The results also showed that only 3% of all policy and planning instruments had engaged field-level PHC workers in their formulation. In the group interviews, PHC workers described their lack of participation in community health program design and reform processes (section 6.3.2). Further research is needed to explore the true extent of participatory engagement with health workers, and within the tiers of Sri Lanka’s health system.

Evidence from the literature suggests that if health care workers are not meaningfully engaged in health care decision making, then worker motivation, sustainable changes in service delivery and accountability at PHC level will be adversely affected (402). Figure 33 presented earlier shows the participatory dynamics within units of the health system and between the health system and communities. Partnerships between health care units (‘Intra- organisational working’) and within professional teams (termed ‘Inter-professionality’ in health care) of a health system have become a central plank of public policy in the UK, especially in the field of health and social care (403). While the specific relevance to other settings has not been explored, these concepts are likely to be important in all settings.

The attitudinal landscape of primary care workers and administrators influences the realization of CP and is presented as a dedicated domain within the complex health system model which I develop further below.

Despite the robust support of CP by health professionals, study findings revealed a scarcity

219 | P a g e

of interventions in routine primary care practice settings that utilized CP approaches. Despite extending the mapping exercise coverage area to encompass all MOH units within three districts, only six interventions were identified that harnessed CP approaches in my study. Most initiatives that were implemented were one-off vertically driven interventions, usually aimed at screening for selected diseases such as diabetes and risk factors such as hypertension. The mapping results indicate that most of these vertical interventions were undertaken as part of a centrally driven initiative by the MoH or donor organisations and were poorly integrated into routine PHC practices and therefore not sustained. Of the six, only two were found to utilize CP approaches that sought to address NCD risk factors and to do so within a routine PHC practice setting with indigenous resource capacities. The two interventions were studied in depth (described in sections 5.7.1 and 5.8.1) and presented as video vignettes.

In-depth case study analysis of these interventions demonstrated that despite resource constraints and rural and remote contexts, meaningful engagement by PHC workers can indeed build community capacity to address lifestyle-related risk factors for chronic diseases, which may also lead to a positive health impact in terms of NCD risk reduction.

The willingness of district and central-level health authorities to provide structured support to incubate and catalyse innovative ideas by PHC workers in tackling NCDs, invest in their development and eventually evaluate their effectiveness is also crucial. The case study analysis revealed limited support from district authorities and technical focal points in supporting the PHC workers in catalysing (and sustaining) the community based NCD prevention intervention. The success of the interventions may be attributed to the effort and commitment of the individual health workers and community participants. Despite knowledge of the interventions, there appeared to be little or no effort by local district health administrators to enhance or formally evaluate such practices (p.184).

Scaling-up locally evolved, resource and cost-effective interventions is a hallmark of adaptive and responsive health systems (128,129). The prevailing culture within NCD prevention and control is one which relies on one-off training programs aimed at increasing knowledge and awareness of PHC workers, or donor-driven initiatives piloted through

220 | P a g e

vertically controlled programs from the central MoH level.

The ‘filter and funnel effect’ was constructed as an analogy to encapsulate the effects articulated by PHC workers and health administrators when external agencies and central level authorities attempt to introduce new innovations through ad-hoc projects or one-off in- service training programs.

The existing PHC worker job duty lists and monitoring mechanisms need to embed CP approaches to achieve PHC program goals, as articulated in numerous health policy documents (Table 14), if such practices are to diffuse into routine practice. The need for dedicated teaching modules within in-service and PHC worker training programs was also indicated by both PHC workers and policy makers.

The next section is devoted to answering this research objective and presenting a complex health systems model. Framed firmly within a socio-ecological perspective, the model suggests that for CP to be meaningfully manifested and sustainably integrated into PHC practice, health decision makers will need to understand the determinants of CP in PHC practice in terms of five domains (or ‘environments’ as they are referred to in the model). For instance, health systems need to inculcate skills and capacities and provide guidance for PHC workers to better engage with communities, challenge dominant professional cultures within certain settings that do not value community or patient collaboration, and encourage innovation.

The model therefore emphasises the stewardship function and leadership of the health authority in inculcating the participatory rhetoric, and not, as one health administrator indicated, to “merely leave it to field level workers to decide what to do”.

221 | P a g e

6.2 ‘Bringing it all together’: Synthesizing an analytical framework to explore CP using a complex health systems approach

In accordance with the methods of thematic analysis using principals of a grounded theory approach, emerging themes from the research studies were progressively and inductively derived, with concepts grouped under increasingly higher order categories. As Bradley (2007) suggests, this analytic method integrates the themes stemming from the research to explore higher associations and formulate an overall theory (the 'hermeneutical approach') (412). Finally, the emergent theory is incorporated into an overall composite health system framework to make it easier for policy makers and researchers to explore associations both within and between health system domains relevant to CP.

Figure 36 provides a visual representation of the resultant substantive theory in the form of a composite health systems framework, consistent with the ‘classic’ WHO health system model (260). It builds upon the basic health systems model presented earlier that was constructed to explore the research objectives (Figure 12). The hallmark of this final model lies in its capacity to present the complex dynamics and interplay of factors that may influence the manifestation of CP in Sri Lanka’s PHC system. It has been constructed, as described below, by integrating research insights and analyses into the progenitor framework presented at the start of the research. It represents a key ‘product’ of this research, and could be adapted for use in other settings.

6.2.1 Building the model

The framework entitled “CP in PHC analytical framework” identifies five domains (also termed ‘environments’) that influence the emergence and diffusion of CP within Sri Lanka’s PHC system. These domains have been categorised in short as: the ‘Administrative and Resource Environment’ (light-blue hue); ‘Policy and Regulatory Environment’ (dark-blue); ‘Technical competency and knowledge environment’ (yellow); and the ‘Attitudinal Environment’ of PHC worker and health administrators (orange).

222 | P a g e

Figure 36. CP in PHC Analytical framework. Conceptual framework presenting the five broad domains/environments that influence the meaning, realization, manifestation and diffusion of community participatory practices to address NCDs at primary health care level in Sri Lanka.

223 | P a g e

The fifth environment is an ‘external’ one (external to the health system), encompassing factors outside the formal health system that may influence the realization of CP practice at PHC level, such as communities’ own commitment towards participation, health volunteerism, socio-cultural factors and other more general economic and political influences.

Factors within each of these environments may work independently or synergistically in various ways to influence the manifestation or utilization of CP within routine PHC practice. This PHC ‘practice environment’ is indicated by the dark circle at the centre of the diagram. The solid arrows stemming from each environment filter towards the PHC intervention space at the centre on the circle, also symbolising the multiplicity of factors influencing CP.

The theoretical framework builds upon the basic model used in exploring the research objectives (Figure 12). The policy and structural environment of the health system is expanded to encompass stewardship (functions such as regulation, legal and administrative functions), whilst the practice environment comprises technical capacity, competency and the attitudes of PHC workers.

For ease of analysis, the routine administrative and operational aspects of health stewardship such as human resource management, program design, monitoring and evaluation are categorized separately from the normative and political aspects of the stewardship function such as policy formulation and inter-sectoral coordination. Those ‘external’ elements to the health system, such as the politicization of community health development processes leading to the disenfranchisement of workers, are also included in the model.

A note on stewardship within the model

Stewardship refers to the wide range of functions carried out by governments through their respective statutory health authorities to achieve national health policy objectives (413). Stewardship is also highly political since it involves agenda setting, prioritization of resources and formulating rules and regulations for health system functionality (414). Stewardship encompasses not only regulatory functions but also the management of both administrative and resource gradients within the health system. This includes distribution of human resources for health, health care financing and establishing effective accountability mechanisms with communities (414). Based on this definition, the stewardship function of the

224 | P a g e

health system extends to the administrative and resource environments as well as to the more obvious health policy and regulatory environments. For ease of analysis within the model, the stewardship function is encompassed within both environments. The rationale for treating stewardship in more than two domains of a health system is also echoed in WHO’s health systems framework model (260) and by other scholars (264).

6.3 Model elements and dynamics

The composition of each environment/domain and the linkages between the factors within each of the 5 domains are elucidated in this section. A graphical format (Figures 21 to 26) provides an overview of the key themes and nodes within each health system domain. The model outlines the multiple health system determinants and conditions that ultimately influence the participatory capacity and participatory potential at the PHC level.

6.3.1 Administrative and resource environment

Key themes influencing the realization of CP within routine PHC practice in relation to this environment are listed (Figure 37) and briefly discussed below.

 A lack of collaborative engagement of stakeholders within the health system

Interview, survey and observational study results revealed limited scope for participatory decision-making between stakeholders within the health system (PHC workers, middle and executive level management) for community based health planning, policy formulation and intervention design (Figure 33 and 35). Enabling participatory approaches to engage with communities at the PHC interface becomes challenging in an organisational culture in which there is a lack of meaningful collaboration and iterative engagement between PHC workers, administrators and executives.

A closely related theme was a general lack of awareness and communication by PHC workers in relation to established strategies and policies identified for NCD prevention and control. For instance, the field level workers and MOHs were unaware of the existence of a medical officer of NCDs (MO-NCD) working at district level who was responsible for 225 | P a g e

stimulating and supporting NCD programming in the field. They were also unaware of national guidelines and policy frameworks on NCD prevention and control in Sri Lanka.

The WHO has highlighted the need for health systems to support “integrated community- based primary health care services” for health promotion, prevention, early detection and treatment of NCDs (21). Currently, however, no such integrated care plans exist at the PHC level in Sri Lanka. Community bypassing of local secondary level health care institutions such as rural hospitals, even for the most minor ailments (highlighted in section 2.3.7), has resulted in an over-utilisation of OPDs at secondary and tertiary care health institutions (253). If a truly integrated community-based PHC service and a co-ordinated case-management approach to tackle NCDs are to exist, then it is necessary to recognise such community dynamics and health seeking behaviours, and to establish partnerships between primary, secondary and tertiary level structures.

 Current PHC work-plans place little emphasis on CP or NCD prevention approaches

Historically, the programming ethos propagated through public health institutions such as the Ministry of Health’s Family Health Bureau and Epidemiology Unit has served to maintain the traditional focus on Maternal and Child Health and communicable diseases surveillance as core functions of PHC units. The goal of prevention relevant to NCDs has not been emphasized.

Some informants indicated a degree of resistance to innovation and reform of the traditional roles and functions of primary care workers. As one district health administrator warned: “one can talk about using primary care workers to work on chronic disease prevention programs, but as long it doesn’t distract the worker from their sacred role of protecting the mother and child”.

Current work-plans, reporting templates and performance monitoring matrices emphasize clinically oriented screening and communicable disease surveillance activities rather than health promotion and community development initiatives. The focus on secondary prevention with limited emphasis on primary (and primordial) prevention activities was also identified by PHC workers and health administrators in the group interviews. NCD prevention and

226 | P a g e

control is not seen as part of the core activities of a PHC worker. There are also no mandatory reporting templates for PHC programs.

Although the overwhelming majority of PHC workers showed willingness to utilize CP approaches, they also cited incompatibility with existing work schedules, management and monitoring practices. CP was characterized by most PHC workers as being a ‘time-intensive’ approach. Over-stretched work schedules inhibited meaningful community health development.

 Innovative practices that utilize CP approaches to address NCD risk factors emerge within routine PHC practice, but have limited support and diffusion

The CP mapping exercise revealed two indigenously driven NCD prevention interventions across three district settings (Table 13). The analysis of these case studies catalysed by two PHMs showed they adopted CP strategies that enabled a high degree of community ownership and agency in driving interventions to address NCD risk factors. Whilst a dedicated assessment may be required to ascertain the impact and effectiveness of such interventions, the case study analyses demonstrated robustness and rapid adoption of these community driven NCD prevention programs in households within these regional and resource limited settings. Despite the existence of such interventions, there was no formal support, adaptation or take-up by the district health system.

As described in the two case studies, meaningful administrative and technical support to further develop such interventions was not provided, not even by the PHC units in which the innovative interventions emerged. Scaling-up of locally driven, evidence-based and cost- effective interventions is a hallmark of responsive health systems (282). The fact that both interventions continue to thrive and expand in the absence of structured support and stimulation from the formal health system is indicative of the high degree of community engagement, dedication and commitment of the PHC workers who catalysed the program.

However, the sustainability of such innovative interventions appears dependent upon the ongoing energy and commitment of the PHM. A lack of support may lead to worker disenfranchisement and limited opportunities for expansion or scaling-up to other settings.

227 | P a g e

 Reducing PHC service area demarcations (with concomitant increase in PHC cadres) will allow for more intensive community engagement

There was strong support for revision of PHC service area demarcations/catchment areas to take into account population densities and ensuring increase in PHC cadres to facilitate more intensive interactions between the PHC system and the local communities. The need for demarcating a reduced PHC service area was advocated by both health administrators and PHC workers. Responders also identified a number of challenges when articulating the need for low PHC worker to community ratios to enhance CP. Significant gaps in HRH that existed in some district health systems resulted in PHC workers needing to provide outreach services to these areas, thereby limiting opportunities to engage communities. Other challenges relate to health care financing for district health systems, with smaller area demarcations requiring a concomitant increase in PHC cadre allotments and HRH resource investments.

 Limited investment for health promotion at PHC level

Prioritization of health sector financing for clinical care, with limited investments for health promotion and primary prevention activities, acts as a barrier to stimulating CP. While a small grant scheme for NCD prevention and control programs had been initiated by the Ministry of Health at central level for selected districts in year 2012, the findings from the district health system profiling indicated that the facility was to be underutilized by both district level administrative staff and PHC workers. There was also evidence of ‘political manoeuvring’ of these discretionary budgets given to the provincial health authorities, leading to resources not filtering down to the PHC level or for PHC units being unaware of such opportunities (sections 5.2.6 and 5.9.2). The community-based NCD prevention interventions led by two PHMs profiled in the case study analysis received no financial and resource support from the district health administrators, despite their knowledge of the programs.

Ambiguous policies on the provision of monetary and resource support for PHC workers, by district and divisional health administrators, to enable transportation and outreach support costs, limit the effectiveness of PHC worker access to rural and remote communities (see 6.7).

228 | P a g e

 Lack of incentives and rewards for undertaking community health interventions

The PHC worker reporting and performance review mechanisms placed little emphasis on community health promotion interventions. There was also no formal reporting requirement for NCD prevention and promotion interventions by PHC workers.

The guiding principles by the MoH outline the purpose of routinely held monthly review conferences involving health administrators and field level PHC workers as ‘strategic planning of PHC activities’ (388). Yet very little strategic planning or reporting of NCD prevention interventions, or indeed interventions harnessing CP approaches, occurred at district level.

PHC workers across the three district settings indicated that these forums primarily functioned to report on disease surveillance data and clinical activities such as antenatal clinics. Since such activities are integrated into routine reporting templates that are also monitored by central level authorities (such as the Epidemiology unit), they are adhered to. No such monitoring system exists for tracking CP processes, nor NCD interventions.

As highlighted in the GIs and the observational follow-up study, most field-level workers spend a considerable time completing epidemiological surveillance reports mandated by the relevant vertical programs. An examination of PHC worker monitoring and review templates indicated no reporting nodes for NCD prevention and health promotion interventions, nor any scope for workers to report findings or describe lessons from community development initiatives. Embedding such components within monitoring matrices, encouraging PHC worker innovation for locally driven chronic disease intervention approaches, whilst inculcating leadership and accountability among district HAs is needed.

229 | P a g e

Figure 37. Key factors within the administrative and resource environment

230 | P a g e

6.3.2 Attitudinal environment

As discussed earlier, CP in health is seen by most PHC workers and administrators from a technocratic perspective in which CP serves to fulfil health program goals, to assist workers in program implementation processes and to achieve public health outcomes. HAs and workers framed communities as ‘beneficiaries’ within this paradigm. This concept of CP is quite distinct from a pedagogical approach in which CP aims to reshape power and facilitate the agency of community members in taking forward their own development.

The themes distilled from the exploration of the attitudes of PHC workers and administrators towards CP and its role in addressing NCDs were broadly categorized into factors that inhibit the realization of CP at PHC level (‘pull factors’) and those that promote and enable CP (‘push factors’). These factors reflect the influences of a multiplicity of determinants, features, and dynamics that impact on the manifestation of CP in PHC practice (see Figure 38).

Factors inhibiting CP

The pull factors inhibiting CP encompass factors such as the lack of perceived support from communities and the professional, structural, cultural and operational barriers within the health system.

 Community attitudes and societal barriers

PHC workers suggested that communities might be disinterested in CP action due to a number of factors. Many consider that communities wish to remain passive recipients of health care rather than play an active role in influencing health services and collaborating with health professionals. Socio-cultural hierarchies embedded in Sri Lankan society confer high status on medical professionals. Such attitudes may inhibit and undermine lay health worker efforts at CP engagement. PHC workers also expressed a view that communities often prioritize and demand biomedical interventions such as clinical screening programs, rather than participating in health promotion interventions. Workers suggested that PHC interventions such as screening for diabetes at outreach clinics were more valued by

231 | P a g e

communities than longer-term programs such as active lifestyle campaigns that seek behaviour change.

Health workers were also challenged by the question of who actually represents (or claims to represent) ‘community’. Some resisted efforts to harness CP practice due to the challenge posed in managing local political dynamics by groups with vested interests and influential local community leaders. Beyond political opportunism, workers also expressed their concerns that community gate-keepers with vested interests may interfere with or inhibit effective coordination of community health programs (5.2.15).

Community volunteerism in the health and social service sector has inhibited CP due to excessive monetary and resource incentives offered to community volunteers. For instance, the remuneration and incentives offered to community volunteers in the aftermath of the 2004 tsunami disaster and with some donor funded health development programs was cited by health workers as a major reason why community volunteerism had declined. Some suggested this monetary and resource expectation from communities had, in recent years, led to a decline in voluntary participation in community-based health interventions.

 Professional ethos and cultural barriers

A minority of PHC workers did not support community participatory approaches to health care. These workers felt communities did not have the capacity to meaningfully engage in health programming and health care decision making. A form of professional elitism was maintained by some health administrators who felt that the views of medical experts should be privileged over community health programs since PHC, in their view, is an ‘expert driven’ enterprise.

As indicated in sections 5.1.7, 5.2.2 and 5.2.3, there is a lack of knowledge among PHC workers and HAs of what constitutes effective CP approaches, and how these are to be implemented. Confusion in the coordination roles of members within each PHC team was also noted, especially with regards to chronic disease prevention. The confusion in coordination roles among members of some PHC units in undertaking community health initiatives extended to fuelling anxieties among some MOHs, many of whom were engaged in private sector general practice clinics in the local area. They resisted any empowerment of allied

232 | P a g e

health professionals within PHC teams and opposed efforts to enhance even basic clinical skills and knowledge. Suggestions that front line PHC workers such as PHMs and PHIs be upskilled to offer basic clinical examinations such as blood pressure assessments at household level, and to provide knowledge on commonly used chronic disease medications such as statin drugs, were not supported.

 Operational barriers

Existing PHC work plans and reporting requirements do not prioritize CP approaches, nor do they stipulate community-based NCD prevention and health promotion interventions. Even locally driven NCD prevention initiatives that were found to utilize highly successfully community participatory mechanisms with indigenously sourced resources (such as those documented in the case-study analyses) were not readily harnessed by district-level health systems. Community-based health promotion initiatives have been integrated into routine PHC practice. Time constraints stemming from excessive workloads make sustained community engagement difficult, even for those PHC workers committed to addressing NCDs using meaningful community-based engagement strategies.

Factors enabling CP

 Provides legitimacy and status

CP further legitimizes the role for PHC workers as the primary conduit between communities and health services. PHC workers emphasized how CP may enhance and reinforce their own status and leadership at community level.

 Enables health program success

CP is viewed as a useful strategy to achieve programmatic goals in public health and for health promotion. Investing in resources and training for PHC workers to utilize CP strategies may lead to better population coverage and effective uptake of existing PHC interventions. Since CP enables the formation of active community groups or citizen coalitions, these may be harnessed by the health system to initiate other health sector interventions.

233 | P a g e

Intensive engagement with communities provides a better understanding of current community health needs, practices and health behaviours. This may assist PHC workers and health administrators not only to tailor health interventions according to local capacities, but also to promote health system accountability to local communities and service users (150).

Figure 38. Key enablers and barriers operating within the attitudinal environment

234 | P a g e

6.3.3 Technical competency and knowledge environment

Skill deficits and inadequate technical capacity to catalyse and sustain community-based NCD prevention and control interventions using CP approaches stem from a number of factors (Figure 39).

 Gaps in knowledge and technical capacity of health administrators and curriculum designers responsible for training PHC workers

Training curricula, at both pre-service and in-service levels, place little emphasis on empowering PHC workers with the required skills in community development and health promotion (section 6.3). Workers also expressed the need to enhance their capacities in community development and health advocacy, which they saw as crucial if underlying determinants of NCDs were to be addressed (section 6.3.4). Boosting capacities for inter- sectoral efforts were also highlighted as a priority by district health administrators. One example was related to mobilizing educational authorities, parent groups, local chambers of commerce and fast-food outlets near schools and tuition centres to provide healthy options for school children.

Due to limited knowledge of chronic diseases, health promotion strategies and techniques for NCD risk factor reduction and practical aspects of home-based care, field level PHC workers were reluctant to engage with communities and respond to their demands. The PHC workers profiled in the in-depth case-studies and observational studies articulated their need to enhance their health promotion strategies and knowledge of community-based models for NCD prevention. They expressed the need for evidence-based methods and technical skills such as patient counselling and the provision of effective home-based management of chronic diseases (6.3.3). Such technical knowledge had not been provided in pre-service and in-service PHC worker training programs (6.8). Some field-level workers expressed reluctance to engage in community participatory practices since they were not confident of being able to effectively respond to such demands from community members and those with chronic disease (sections 6.2.11 and 6.3.1).

235 | P a g e

Current worker training programs emphasize strategies directed at individual risk factors rather than seeking to address social determinants of health. As highlighted in Table 13, pilot programs currently designed by the MOH for addressing NCDs such as “Healthy lifestyle centres program” and “Package of Essential Non-communicable Disease Interventions for primary care” (PEN) are driven by need to systematically assesses those between 35 and 65 years of age on physical risk factor assessment. There has been little scope for promoting CP practice. Rather, the emphasis appears to be on clinically oriented approaches. For instance the PEN project in Sri Lanka focused primarily on providing medical equipment such as blood glucometers, peak flow meters and essential drugs to PHC settings. Whilst such measures are vital for empowering PHC facilities with the capacity to treat conditions like hypertension, technical and resource support to catalyse health promotion interventions with engagement from local communities for combatting NCDs have not been included in the “Package of Essential Non-communicable Disease Interventions for primary care” (Table 13).

 The need for evidence-based guidelines for CP approaches

Health authorities have not documented (systematically or selectively) the practices and lessons learned from community-based health interventions. There are no locally developed evidence-based guidelines on CP approaches. As highlighted by the responses of PHC workers, an accessible repository or compendium of community based interventions that may serve as a reference guide for workers and administrators would be useful (section 6.3.4).

The district profiling study also revealed that Information Exchange Communication (IEC) materials for PHC workers to undertake health education interventions related to chronic diseases and their risk factors were not readily available across the 3 MOH settings (Table 6.9).

 The ‘filter and funnel’ effect

"One-off" training programs aimed at enhancing PHC worker knowledge of NCD prevention or stimulating initiatives at the PHC level leads to poor diffusion and integration within routine practice (section 6.3.1). At best, such one-off training efforts result in a selective application of practice that is not sustained. The “filter and funnel effect” (section 6.3.3), as articulated by PHC workers and health administrators, encapsulates these effects. This is 236 | P a g e

especially in relation to external agencies and central level authorities attempting to introduce (and diffuse) new innovations, skills and approaches for PHC practice through ad- hoc pilot projects or one-off in-service training programs.

 Job function reform should accompany efforts to enhance technical capacity

The willingness and commitment articulated by PHC workers to be empowered through in- service training programs in learning requisite skills concerning community health development should be viewed positively by health decision makers in the fight against chronic diseases and within the context of revitalizing PHC services (section 6.3). However, as highlighted in previous environments, workers indicated that changes were needed to existing job duty lists and reporting factors if meaningful diffusion and sustainability of such CP efforts were to take place (Section 6.5). District and central-level health authorities may need to be willing to provide structured support to incubate innovative ideas such as those initiatives described in the case studies, as well as investing in their development and evaluating their effectiveness (Section 6.7.2).

To summarize: the interplay of the above mentioned factors has revealed two key health system results. Firstly, there is a lack of technical skills and capacities at the PHC level for workers to undertake community-based NCD prevention interventions; secondly, the requisite monitoring and evaluation mechanisms needed for capturing emerging innovations and good practices and then ensuring their diffusion across health system have not been established.

237 | P a g e

Figure 39. Key factors and dynamics within the technical competency and knowledge environment

238 | P a g e

6.3.4 Policy and regulatory environment

CP is clearly articulated as a strategy in the majority of domestic health policies in Sri Lanka. The role that PHC workers have in working with communities to achieve health policy goals has also been explicitly articulated in a number of policy documents, including the overall guiding framework for the health system, the National Health Master Plan. Despite the policy rhetoric surrounding CP, clinically oriented interventions dominate the PHC intervention landscape and are emphasized as a key means of community empowerment within the national NCD prevention and control program (‘healthy lifestyle clinics’). Qualitative findings also indicate limited participation of field-level health care workers and administrators in policy formulation and in the design and evaluation of community health programs.

Decentralization of health decision making to district and village-level units is constitutionally enshrined. Under the 13th amendment of the constitution, Provincial Ministries of Health have the responsibility to formulate policy, inter-sectoral action in health and coordinate context- specific public health programs such as dengue control. The prevailing political culture in Sri Lanka has resisted the devolution of health sector governance to district and village-level structures, and health planning and decision making remain centralized. The disbanding of the National Health Development Council (NHDC) and village development committees since 2004 due to the incumbent governments’ centralization of power has led to an erosion of civic engagement.

Grass-roots level inter-sectoral coordination mechanisms that were supported by health and local governance systems to enable greater community and civil society participation were eventually dissolved in 2005 under the presidency of Mr Mahinda Rajapaksha. The Rajapaksha presidency from 2005 to 2014 was characterised by a greater concentration of powers to central authorities, effectively reducing many of the decision making powers vested in the provincial and district level administration (including provincial health councils).

As highlighted in Section 5.9.7, the village-based Nutrition Coordination Committees (NCC) established at MOH level in 1994 were an example of a practical community-based program that was catalysed and resourced by a policy decision emerging from the NHDC. The NCCs comprised community members, civil society co-operatives, health workers,

239 | P a g e

farmers and local chamber of commerce representatives that met on a regular basis to maintain essential food items in local markets, facilitate food security programs and provide access to high quality fruit and vegetables. Such approaches not only embrace CP and community empowerment ideals but also address deeper social determinants important to ensure adequate nutrition, healthier lifestyles, and food security and availability.

Resistance by large and powerful health worker trade unions against the re-demarcation of PHC zones (see Box 4, Section 5.4.2) and by some departments within the MoH (such as the Family Health Bureau), which resisted altering the traditional roles of PHC workers (Section 5.2.8), makes revitalizing job functions and structural reform difficult.

Although CP is enshrined as an integral part of the fulfilment of many health policy goals and objectives, its definition, scope and practice is not clearly defined and remains ambiguous. Despite the goals of the national NCD policy for Sri Lanka, explicitly prioritizing the need for health systems to engage in ‘community empowerment’ processes, the accompanying policy action plans are devoted primarily to training workers to undertake screening of individuals for NCDs. Little importance is assigned to building the technical capacities of PHC workers to undertake health promotion and community development initiatives as part of the ‘roll-out’ strategy (Section 5.9.2).

Findings suggest that even when innovative practices in CP in health are catalysed by PHC workers, these community-based health initiatives are not readily absorbed or supported by health administrators, thereby inhibiting their further development and potential diffusion within other PHC settings. District profile mapping and analysis of PHC worker training curricula found limited training resources and reference material to guide PHC workers and health managers on participatory approaches to address NCDs.

Efforts should be made to document community driven interventions to combat NCDs such as those outlined in the case studies presented in this thesis within other district settings. Case study analysis forms a practical and persuasive method that may guide indigenous efforts in promoting CP practices at the PHC level. A rigorous evaluation of interventions may then be carried out to distil the key strategies used and to identify processes of engagement and the ‘lessons learned’ in adapting participatory practice to the local context. Creative presentation of such interventions to health workers, in the form of short films (as developed in this thesis) and accompanying study guides, may serve as a useful starting point to diffuse 240 | P a g e

innovative practices. If such concerted efforts are not possible, then, at the minimum, peer-to- peer empowerment approaches may be harnessed. For instance, those PHC workers who are the key protagonists of innovative interventions can share their experience and expertise in catalysing NCD prevention at the monthly conferences or district level workshops. A deliberative skills building and planning workshop may then be carried out to enable PHC units to explore strategies for CP in health.

Finally, the literature review found very few studies on CP in health interventions in Sri Lanka, and none were found to examine its manifestations within the nation’s health system. In order to promote an evidence-based approach to policy and intervention planning, a sound evidence base is indeed. Research areas into CP in Sri Lanka are further elaborated in section 6.5.3. Figure 40 aims at encapsulating these factors within the policy and regulatory environment and outlining the linkages between such factors.

241 | P a g e

Figure 40. Key factors and dynamics within the policy and regulatory environment 242 | P a g e

6.3.5 Use and limitations of the model

“To date, CP has most often been seen as an intervention to improve health outcomes rather than a process to implement and support health programmes to sustain these outcomes. To understand the relationship between CP and improved health outcomes, new frameworks are needed…” Susan B. Rifkin (51) As highlighted in the statement above by Rifkin, and also reiterated by others (153), little effort has been made to move beyond lists of relevant factors that influence CP to more comprehensive health system models that help guide more effective CP approaches by Ministries of Health. A greater understanding is needed of how factors within organisations interact to inhibit, enable, and sustain more effective community partnerships.

The complex systems model presented here is not intended to ‘measure’ the degree of CP within a specific health intervention or setting, such as Rifkin’s Spidergram model (15). Rather, the model seeks to identify various health system components and dynamics that may work synergistically to enable or inhibit CP within routine PHC practice. By examining each critical health system function in this manner, the framework offers both a degree of macro-analysis of the overall health system alongside the relevant micro-analyses (for example, the policy environment), which help elucidate what is actually happening, and the potential for change, at the primary care level.

Identifying the relational dynamics within each environment may enable health system analysts and decision makers to better understand how participatory practice may be harnessed (and sustained) within routine PHC practice settings. The model also contends that each environmental domain may exert influence independently as well as in relation to other factors that ultimately affect the PHC intervention space. For instance, despi te the existence of a conducive health policy framework toward CP and a well-financed chronic disease screening program, participatory approaches are unlikely to emerge or thrive in routine practice unless the ‘pull’ factors within the attitudinal environment of PHC workers are addressed. The analysis reveals some of the constraints on the emergence or integration of CP, and the factors which promote CP flourishing and being sustained. These insights may assist policy makers to develop local options and preferences for future action.

Attempts to realize CP may also be thwarted by a failure to address the underlying barriers to CP. For instance, limited health worker competencies in effectively engaging with communities can be linked to the absence of training tools and PHC worker curricula that fail 243 | P a g e

to provide guidance to workers regarding techniques and strategies to facilitate participatory engagement. This in turn is linked to a health stewardship culture in which CP is not prioritized or monitored within routine work plans and interventions at the PHC level.

The model offered here encourages analysis of the interactions and interconnections between elements of the environments, allowing a systematic and holistic analysis of the adoption and diffusion of participatory health interventions in general. Whilst the model was developed through a grounded analysis of issues specific to the Sri Lankan health system, it may also be relevant to other settings and systems. The model may also be used to guide an applied health system research agenda around each domain, to further ‘unpack’ the factors and issues highlighted. Annexure 8 hypothesizes the potential linkages between these different domains. It must be noted however that not all such associations described have been elicited through the empirical work. The analysis presented in Annexure 8 should therefore be considered an explanatory frame that reflects work progress, but requires further research for establishing definitive links/associations between the domains. Such additional work was beyond the scope of this thesis.

Increased understanding of the importance of actively involving communities in health care and the potential of communities to successfully direct health interventions has emerged over the past 50 years (415). If health systems are sincere about embracing CP as a viable strategy in PHC – as promulgated by numerous global health declarations and, in the case of Sri Lanka, enshrined within the national health policy framework – then a number of measures need to be taken if long-term strategic changes are to occur. Statutory organisations will need to develop more sophisticated skills and techniques for engaging with communities in order to change the dominant professional cultures within their organisation and to develop a more participatory culture through encouraging innovation and tackling risk aversion.

As indicated by Pickin and colleagues (153) and also by Neuwelt (13), the ‘style’ of health sector stewardship is vitally important in this process of revitalizing PHC. Understanding and reshaping the power balance between individuals and communities, and within the health system more generally, is vital to CP. Effective approaches require recognition of the complex influences within organisations that affect their ability to develop and sustain such relationships.

244 | P a g e

The inhibitory forces outlined in the model will need to be tackled before the policy drivers can take effect. The model may also provide the basis for a process of organisational learning in which assumptions are tested at various PHC units at district level, and health administrators are assisted in identifying where and how they need to build their own capacity to engage effectively with communities.

Finally, for the policy maker and the practitioner, any analytical framework is only as good as the extent to which it is recognisably applicable to real-life problems (282). Use of this framework model at country level will lead, no doubt, to its further refinement over time. A number of additional study challenges/limitations have been elaborated in Annexure 9.

6.4 Conclusion

For participation to truly ‘work’ within health systems, an enabling environment conducive to its meaningful translation into routine primary care practice is needed. In undertaking this research, I have recognised that there is little evidence of the actual practice of CP at the PHC level, despite it being enshrined as a core strategy in Sri Lanka’s national health policy framework, promoted in the current PHC revitalization and NCD prevention action plans and largely supported by health professionals across all tiers of the health system.

A lack of support mechanisms at the PHC level to catalyse CP approaches, technical, institutional and managerial competency to facilitate meaningful community engagement, and the lack of commitment of some health managers and workers towards CP, have inhibited its realization. As the results of the intervention mapping across the district health system setting have shown, only a few committed and dynamic PHC workers successfully managed to navigate such constraints and harness indigenous capacities to catalyse and sustain CP practices in health.

A key element of decentralization in health systems is the reshaping of power, including the degree of choice and engagement that is transferred from central administration to institutions at the periphery of health systems (126). In Sri Lanka, the National Health Development Network, which directly forged channels for participatory engagement in health

245 | P a g e

sector governance processes for field level health institutions, PHC units and civil society groups, was dissolved in 2004. This is likely to have contributed to an erosion of participatory space in health care decision making in Sri Lanka. Nurturing partnerships and valuing consultative processes may promote a participatory culture at the PHC level that may in turn enhance CP.

A missing component of the literature on participation was the extent to which policy makers at central levels of the health bureaucracy engage within the various layers of the health system. Figures 33 and 34 presented earlier showed the limited participatory dynamics within units of Sri Lanka’s health system and between the health system and the communities with which it should interact. Such intra-health system engagement and coordination is an essential part of a decentralized health system. There is also evidence that if workers are not meaningfully engaged in health care decision-making, then worker motivation will be adversely affected (402). Active partnerships between health care units (inter- organisational) and within professionals in PHC teams (‘inter-professionality’) have been identified as an important public policy priority strategy in national health systems (403).

Although this thesis focused on exploring the dynamic of CP between health professionals and communities, a lack of a participatory culture within the health system was identified. If CP is to be meaningfully embraced as a viable public health strategy, then health authorities in Sri Lanka need themselves to promote a culture of iterative and collaborative planning between the health executive, district-level management and grass-roots PHC workers.

PHC workers showed a sound understanding of the growing burden of NCDs, and indicated their commitment to reversing this trend at community level via effective participatory approaches. However, a number of pre-requisites were highlighted as being important to realize these goals: reform of existing work duties, management and reporting practices to include NCDs, skills development and training; reducing work areas to ensure a more meaningful interaction with consumers and enable a more intensive intervention space; dedicated resource support to build community capacities; and conducive regulatory frameworks to enable healthy choice options for the public. Despite the rhetoric, there are only a few NCD prevention initiatives currently operating at the PHC level in the study sites, and only two were identified that involved CP approaches.

246 | P a g e

To bridge the gap between what health workers want and are committed to doing, and what existing policies aspire to achieve, requires strategic efforts to connect the ‘middle ground’. These include efforts to revitalize worker duty lists, training curricula, management structures and stewardship functions at the primary care level. Other factors include skills development for workers and administrators in how to construct a CP strategy; reforming PHC work areas to smaller catchment sizes to ensure a more meaningful and intensive interaction with communities for community health development; dedicated resource support to build community capacities, and linking primary care practitioner networks (both public and private) with community health development initiatives.

The development and application of a practical instrument such as the P-Compass showed that it is indeed possible for health workers and managers to quickly deconstruct the complexity of understanding CP and, through personal reflection, grade the level of CP in their own practice setting. The instrument was designed as a reflective tool and not as a rigorous means of measuring CP. With further testing and development, it may be adapted and utilized to gauge the dynamics of CP within a health system program or institutional unit.

The implementation of community participatory interventions to combat NCDs within Sri Lanka’s health system requires innovation and careful adaptation. Managing organisational, financial, technical and attitudinal aspects will remain a challenge for health policy makers. Policy makers need to understand that, in complex systems, interventions may also have unintended consequences. For instance, attempts to introduce new innovations into routine PHC practices through one-off training programs undertaken without appropriate adaptation to local contexts and in the absence of sustainable follow-up mechanisms may result in ‘the filter and funnel effect’ (section 6.3.3). To reduce resistance to the uptake and assimilation of innovations at the PHC level, a ‘systems thinking’ approach should be adopted, as highlighted in the overall analytical framework presented here. Such adoption requires partnership with PHC workers and communities to facilitate an ‘enabling environment’ - as espoused by the Ottawa Charter for health promotion.

A greater emphasis on community engagement has been echoed in recent MoH efforts to revitalize PHC services to meet the challenges of a growing burden of NCDs, and the associated need for community-based chronic disease management. CP also continues to occupy a prominent place in current global health dialogue, with UN member states at the

247 | P a g e

30th Anniversary of Alma At Declaration in 2008 reaffirming their commitment to its principles, with CP promoted as a key factor in discourses around the social determinants of health and health as a human right (416). If CP is indeed viewed by governments as a vital component for re-orienting PHC health services, then a rigorous evidence base is necessary to inform decision makers and practitioners.

6.5 Relevance for Policy and Practice

Based on the study findings, a “report card” on the current status of CP within PHC services in Sri Lanka is presented (Figure 41). The surface area of each domain corresponds to the extent to which each domain supports and enables CP in health. i.e. largest size representing a domain conducive and supportive of CP. The ‘Report Card’ also serves to highlight the potential areas for health system intervention in enhancing capacities for CP in PHC, and more specifically for NCD prevention and control.

Figure 41. Report card on the current status of CP within PHC services

248 | P a g e

Despite a strong domestic health policy framework promoting CP in health, and supportive attitude by health workforce across all tiers of Sri Lanka’s health system (from ‘front line’ PHC workers to middle level managers and health executives) towards CP, there is poor programmatic inclusion, inadequate resource provision, health administrative oversight, and financial investment for the meaningful engagement of CP in PHC in Sri Lanka. In addition, mechanisms on ensuring policy accountability and monitoring where CP has been explicitly articulated for the achievement of various health goals, including those on NCD prevention and control, have not been established/nor followed through by the Ministry of Health. Despite the enabling policy and attitudinal environment, there has been little effort to develop technical competencies, training curricula and investment on empowering PHC workers, NCD prevention programs and PHC units with the required skills for CP in health. The lacuna between stated policy objectives and organizational and programmatic practices and concomitant deficits in technical competency and investment for CP in health requires critical evaluation. As a result there is little or no CP practices within the PHC system, and more specifically within NCD prevention and control programs – this is represented by the shrunken black sphere of ‘CP in health’ at the centre of Figure 41.

Implications on health policy and practice

It is remarkable that this study is the first to comprehensively assess and systematically document CP at PHC level in Sri Lanka, given the historic accolades bestowed on Sri Lanka’s PHC system since Alma Alta in 1978. This research has offered important insights into how health professionals define and perceive the importance of CP in health, the domestic policy and legal frameworks pertaining to CP, the manifestation of CP at district level and the barriers/enablers for its utility within routine PHC practice. A number of issues for health policy and practice are distilled in light of the study findings:

1. The in-depth exploration of the meaning of key terms such as ‘community’, ‘community participation’ and “CP approaches” in this thesis is important in the context of health policy formulation and practice in Sri Lanka. Hitherto, no such exploration has been undertaken despite Sri Lanka’s early commitment to PHC. The assumption concerning definitions of community and CP is critical to the investigation of the link between this concept and evidence concerning proven health outcomes. The literature reviewed 249 | P a g e

continually points to the fact that without a clear definition of these terms it is not possible to set and evaluate program objectives. Where CP has been clearly defined and targets and indicators identified, linking concepts with outcomes becomes more feasible within health program development/PHC practice. 2. Sri Lanka has not effectively translated its policy rhetoric on CP into PHC practice. The results of policy analysis indicated that although CP has been well articulated in the majority of national health policy documents and accompanying action plans, a dedicated monitoring framework and relevant set of program indicators on CP is absent. More broadly a policy accountability framework and mechanism (for instance an audit or review board) needs to be established. 3. Resource allocations for CP intervention activities have not been established for PHC units or for programs managed by PHC workers. No technical guides exist to outline the diverse strategies, skills, guidelines, technical capacities and resource requirements to undertake effective CP interventions within PHC settings (Annexure 7). 4. Whilst the study clearly indicated the commitment of senior experts, front line health workers, and middle level managers at the Ministry of Health to promote CP, there appears little knowledge or interest in learning from existing or developing new CP interventions. The video-graphic case studies and analytic methods developed in this thesis provide practical learning tools for health planners to assist in developing NCD prevention interventions that embrace CP approaches. Rifkin, (1996) states that the expertise of PHC workers develops through a cyclical process of “experiential learning and teaching involving indigenous knowledge” refined by consistent contact with local communities. PHC worker expertise as documented in the case-studies presented can therefore be used to inform the development of on-site learning and further development of the public health workforce. 5. The set of health system analytic tools developed and tested in this research project, namely the P-Compass and PPD-tool were innovative. Further validation of these tools across multiple settings of the health system will further enhance their viability and applicability. There is a critical need for technical support and capacity building for mapping other CP practice in health, assessing the various modes of CP and in- depth evaluation of such programs. Building a repository and resource pool of PHC workers from such evidence-informed approaches that are sensitive to cultural

250 | P a g e

contexts and resource realties will be critical for advancing CP. A National research group on CP in health could be established with the aims of undertaking such a task. The group may be steered by the MOH with support from health partners in order to coordinate research at country level, develop evidence based guides, and for documenting and evaluating practices. The ways in which health systems engage with communities are best understood by using multi-level approaches. 6. During interviews, PHC workers identified a number of areas that may be useful in developing their skills and competencies with respect to NCD prevention and control. These areas included ‘risk communication’ and ‘counselling skills in lifestyle modification’. These have been published as a separate report and presented to the MoH as part of the research contribution to health systems strengthening.

7. There needs to be a critical review of the role of the Medical Officer of NCDs, who has been designated as the district health system focal point for NCD related programming. Currently there is poor delineation of the functions of this post, and indeed the level and scope of the position in supporting community based approaches for NCD prevention and control.

6.6 Recommendations for future research

In advancing an evidence-based health systems research agenda, the following may be useful in ‘completing the knowledge circle’ to assist Sri Lanka’s health policy makers and planners to devise innovative and sustainable intervention strategies to address NCDs at the primary care level. The findings of such studies may be useful in the current efforts by the Ministry of Health towards PHC revitalization in the context of the growing NCD burden.

 The current study on CP in primary care was framed from the perspective of the statutory health authority and of health care professionals. Documenting the experiences and perspectives relating to CP of communities, patients and health consumers is essential to ‘complete the circle of understanding’ regarding the complexity of the participatory dynamics and factors that influence and cultivate meaningful engagement between communities and health systems.

251 | P a g e

 Analysing the diffusion of NCD prevention and control interventions that harness CP approaches across a diverse range of community settings, and/or address one or more NCD risk factors becomes important if health system planners intend to positively influence the routine practice culture in at PHC level. Qualitative research assessing organisational readiness for change at district, provincial and central levels would also be useful.

 If involving the public in health service development is to lead to better services and improved outcomes, then evidence will be required to show the effects of this on the quality and effectiveness of services. Professionals greatly influence how service users are to be involved in service development and this may limit the change that can be achieved. Although methodologically challenging, studies elucidating the causal links and pathways between the intensity or degree of participatory practice and measure of community engagement with actual health status indicators and/or health seeking behavioural outcomes will be important to understanding the underlying influence of CP in health.

 Financial assessment and costing studies that aim to estimate the costs (both direct and indirect) involved in formulating, planning, implementing, sustaining and evaluating community participatory intervention approaches in primary care would be valuable. Modelling the costs of interventions that emerge within the health system (such as those documented as video-ethnographic case studies in this research project) and those that are vertically driven may be useful. Assessing the level of internal (health system) versus external (development partner or donor) contributions within each intervention will yield useful data on how best to allocate resources to these interventions and with what anticipated effects. Modelling costs of interventions that seek to address risk-specific factors for NCDs, those that are specific to chronic diseases, or those that address general life-style and wellbeing factors, will be important for health system analysts seeking to develop practical and sustainable strategies that could be scaled up to national level.

252 | P a g e

 Finally, a comprehensive investigative ‘mapping’ of the NCD prevention and health promotion interventions that utilize CP approaches across the remaining 6 districts in Sri Lanka may be useful to identify desirable strategies, required resources and necessary operational changes. By adopting the analytical framework and evaluation tools utilized in the intervention case-study component of this research, a more rigorous and comparable set of data may be derived. A strong intervention evidence base is likely to be required to enable policy makers, planners and health care workers to plan effective CP in primary care practice.

253 | P a g e

REFERENCES

1. Chan, M. (2008) Return to Alma-Ata. The Lancet, 372, 865-866. 2. Macaulay, A.C., Commanda, L.E., Freeman, W.L., Gibson, N., McCabe, M.L., Robbins, C.M. and Twohig, P.L. (1999) Participatory research maximises community and lay involvement. North American Primary Care Research Group. BMJ, 319, 774-778. 3. Zakus, J.D.L., & Lysack, C. L. (1998) Revisiting community participation. Journal of health policy and planning, 13, 1-12. 4. Bossert, T. (1998) Analyzing the decentralization of health systems in developing countries: decision space, innovation and performance. Social Science & Medicine, 47, 1513-1527. 5. Ramiro, L.S., Castillo, F.A., Tan-Torres, T., Torres, C.E., Tayag, J.G., Talampas, R.G. and Hawken, L. (2001) Community participation in local health boards in a decentralized setting: cases from the Philippines. Health policy and planning, 16, 61-69. 6. Kilpatrick, S., Cheers, B., Gilles, M. and Taylor, J. (2009) Boundary crossers, communities, and health: exploring the role of rural health professionals. Health & place, 15, 284-290. 7. WHO (1991) Community involvement in health development: challenging health services. Report of a WHO Study Group. World Health Organization technical report series, 809, i-iv, 1-56. 8. Meleis, A.I. (1992) Community participation and involvement: theoretical and empirical issues. Health services management research. 5, 5-16. 9. National Institute for Health and Clinical Excellence (2008) Community Engagement to Improve Health. London, UK. 10. Lawn, J.E., Rohde, J., Rifkin, S., Were, M., Paul, V.K. and Chopra, M. (2008). Alma-Ata 30 years on: revolutionary, relevant, and time to revitalise. The Lancet, 372(9642), 917-927. 11. Irwin, A. and Scali, E. (2005). Commission on Social Determinants of Health. World Health Organization, Geneva. 12. Schwab, M. and Syme, S.L. (1997) On paradigms, community participation, and the future of public health. American Journal of Public Health, 87, 2049-2051. 13. Neuwelt, P.M. (2012) Community participation in primary care: what does it mean 'in practice'? Journal of primary health care, 4, 30-38. 14. Ziebland, S., Evans, J. and McPherson, A. (2006) The choice is yours? How women with ovarian cancer make sense of treatment choices. Patient education and counseling, 62, 361- 367. 15. Rifkin, S.B., Muller, F. and Bichmann, W. (1988) Primary health care: on measuring participation. Social Science and Medicine, 26, 931-940. 16. Mendis, S., Davis, S. and Norrving, B. (2015) Organizational Update: The World Health Organization Global Status Report on Noncommunicable Diseases 2014; One More Landmark Step in the Combat Against Stroke and Vascular Disease. Stroke, 46(5):e121-2. 17. Samb, B., Desai, N., Nishtar, S., Mendis, S., Bekedam, H., Wright, A., Hsu, J., Martiniuk, A., Celletti, F. and Patel, K. (2010) Prevention and management of chronic disease: a litmus test for health-systems strengthening in low-income and middle-income countries. The Lancet, 376, 1785-1797. 18. Beaglehole, R., Epping-Jordan, J., Patel, V., Chopra, M., Ebrahim, S., Kidd, M. and Haines, A. (2008) Improving the prevention and management of chronic disease in low-income and middle-income countries: a priority for primary health care. The Lancet, 372, 940-949.

254 | P a g e

19. WHO (2005) Preparing a healthcare workforce for the 21st century: the challenge of chronic conditions. Chronic illness, 1, 99-100. 20. Atun, R., de Jongh, T., Secci, F., Ohiri, K. and Adeyi, O. (2010) A systematic review of the evidence on integration of targeted health interventions into health systems. Health policy and planning, 25, 1-14. 21. WHO (2008) World Health Assembly Resolution (WHA 61.17) on Health of migrants. 22. UN (2011) United Nations General Assembly Political declaration of the high-level meeting of the general assembly on the prevention and control of non-communicable diseases. New York. 23. Diem, G., Brownson, R.C., Grabauskas, V., Shatchkute, A. and Stachenko, S. (2015) Prevention and control of noncommunicable diseases through evidence-based public health: implementing the NCD 2020 action plan. Global health promotion. 24. WHO (2013). Global action plan (GAP) for the prevention and control of noncommunicable diseases 2013-2020.WHO, Geneva. 25. Abegunde, D.O., Shengelia, B., Luyten, A., Cameron, A., Celletti, F., Nishtar, S., Pandurangi, V. and Mendis, S. (2007) Can non-physician health-care workers assess and manage cardiovascular risk in primary care? Bulletin of the world Health Organization, 85, 432-440. 26. Sadur, C., Moline, N., Costa, M., Michalik, D., Mendlowitz, D., Roller, S., Watson, R., Swain, B., Selby, J. and Javorski, W.C. (1999) Diabetes management in a health maintenance organization. Efficacy of care management using cluster visits. Diabetes care, 22, 2011- 2017. 27. Aubert, R.E., Herman, W.H., Waters, J., Moore, W., Sutton, D., Peterson, B.L., Bailey, C.M. and Koplan, J.P. (1998) Nurse case management to improve glycemic control in diabetic patients in a health maintenance organizationA randomized, controlled trial. Annals of internal medicine, 129, 605-612. 28. Carlfjord, S., Lindberg, M., Bendtsen, P., Nilsen, P. and Andersson, A. (2010) Key factors influencing adoption of an innovation in primary health care: a qualitative study based on implementation theory. BMC Family Practice, 11, 60. 29. Nichter, M.A. (1986) The primary health center as a social system: PHC, social status, and the issue of team-work in South Asia. Social science & medicine (1982), 23, 347-355. 30. Dockery, G. (1996) In De Koning K, M. M. (ed.), Participatory research in health: issues and experiences. Zed Books, London. 31. Bhutta, Z., Nundy, S. and Abbasi, K. (2004) Is there hope for South Asia?: Yes, if we can replicate the models of Kerala and Sri Lanka. BMJ, 328, 777. 32. Rannan-Eliya, R.P., Lankani Sikurajapathy. (2008) Good Practice in Expanding Health Care Coverage. Research Studies Series number 3. Institute for Health Policy Publications, Sri Lanka. 33. Engelgau, M.M., Okamoto, K., Navaratne, K.V. and Gopalan, S. (2009) Prevention and control of selected chronic NCDs in Sri Lanka: policy options and action. World Bank Publications, Washington. 34. WHO. (2012). Non Communicable Diseases, Country Situation: Sri Lanka. WHO publications, Colombo, Sri Lanka. 35. Ministry of Health (2012) Approach and Guidelines for strengthening health care at primary level. Policy analysis Unit, Colombo, Sri Lanka. 36. Katulanda, P., Jayawardana, R., Ranasinghe, P., Rezvi Sheriff, M. and Matthews, D.R. (2012) Physical activity patterns and correlates among adults from a developing country: the Sri Lanka Diabetes and Cardiovascular Study. Public health nutrition, 1, 1-9.

255 | P a g e

37. Jayawardena, R., Byrne, N.M., Soares, M.J., Katulanda, P., Yadav, B. and Hills, A.P. (2013) High dietary diversity is associated with obesity in Sri Lankan adults: an evaluation of three dietary scores. BMC Public Health, 13, 314. 38. Mendis, S., Al Bashir, I., Dissanayake, L., Varghese, C., Fadhil, I., Marhe, E., Sambo, B., Mehta, F., Elsayad, H. and Sow, I. (2012) Gaps in Capacity in Primary Care in Low-Resource Settings for Implementation of Essential Noncommunicable Disease Interventions. International journal of hypertension, 2012. 39. Gubhaju, B.B., Dutta, M. and Husain, Z. (2011) Demographic transition in Southern Asia: challenges and opportunities. Asia-Pacific Population Journal, 26, 3. 40. Premaratne, R., Amarasinghe, A. and Wickremasinghe, A.R. (2005) Hospitalisation trends due to selected non-communicable diseases in Sri Lanka, 2005-2010. The Ceylon medical journal, 50, 51-54. 41. Walley, J., Lawn, J.E., Tinker, A., De Francisco, A., Chopra, M., Rudan, I., Bhutta, Z.A. and Black, R.E. (2008). Primary health care: making Alma-Ata a reality. The Lancet 372, no. 9642 (2008): 1001-1007. 42. Rohde, J., Cousens, S., Chopra, M., Tangcharoensathien, V., Black, R., Bhutta, Z.A. and Lawn, J.E. (2008) 30 years after Alma-Ata: has primary health care worked in countries? The Lancet, 372, 950-961. 43. Gillam, S. (2008) Is the declaration of Alma Ata still relevant to primary health care? BMJ, 336, 536-538. 44. Bowen, S. and Zwi, A.B. (2005) Pathways to "evidence-informed" policy and practice: a framework for action. PLoS medicine, 2, e166. 45. Snape, D., Kirkham, J., Britten, N., Froggatt, K., Gradinger, F., Lobban, F., Popay, J., Wyatt, K. and Jacoby, A. (2014) Exploring perceived barriers, drivers, impacts and the need for evaluation of public involvement in health and social care research: a modified Delphi study. BMJ open, 4, e004943. 46. Adam, T. and de Savigny, D. (2012) Systems thinking for strengthening health systems in LMICs: need for a paradigm shift. Health policy and planning, 27 Suppl 4, iv1-3. 47. Alliance for health policy and systems research (2013) What is Health Policy and Systems Research?, What is Health Policy and Systems Research? WHO Publications, Geneva. 48. Cohen, L., Manion, L., & Morrison, K. (2000) Research Methods in Education (5th Edition) (Online Version). Routledge Falmer., London, UK. 49. Wickramage, K. and Nellapalli, P. (2008) Community participatory methods in disease surveillance and public health in war-affected camps, and its potential contribution to peace building. Global Journal of Health Education and Promotion, 11. 50. Wickramage, K. (2006) Building the capacity for health promotion in resource poor, tsunami devastated and conflict ravaged zones. Global Health Promotion Journal, 13, 208-210. 51. Rifkin, S.B. (2014) Examining the links between community participation and health outcomes: a review of the literature. Health policy and planning, 29 Suppl 2, ii98-106. 52. Mubyazi, G.M., & Hutton, G. . (2012) Rhetoric and reality of community participation in health planning, resource allocation and service delivery: a review of the reviews, primary publications and grey literature. . Rwanda Journal of Health Sciences, 1, 51-65. 53. Victora, C.G. (2013) Commentary: Participatory interventions reduce maternal and child mortality among the poorest, but how do they work? International journal of epidemiology, 42, 503-505.

256 | P a g e

54. Koon, A.D., Rao, K.D., Tran, N.T. and Ghaffar, A. (2013) Embedding health policy and systems research into decision-making processes in low- and middle-income countries. Journal of Health research policy and systems. 11, 30. 55. Anand, S. and Barnighausen, T. (2012) Health workers at the core of the health system: framework and research issues. Health policy, 105, 185-191. 56. Viergever, R.F., & Ghaffar, A. (2011) Reviews of WHO Documents on Health Research: Alliance for Health Policy and Systems Research. WHO, Geneva Publications. 57. Walley, J., Lawn, J.E., Tinker, A., De Francisco, A., Chopra, M., Rudan, I., Bhutta, Z.A. and Black, R.E. (2008) Primary health care: making Alma-Ata a reality. The Lancet, 372, 1001- 1007. 58. Green, G., Price, C., Lipp, A. and Priestley, R. (2009) Partnership structures in the WHO European Healthy Cities project. Health Promotion International, 24, i37-i44. 59. Green, G., Price, C., Lipp, A. and Priestley, R. (2009) Partnership structures in the WHO European Healthy Cities project. Health promotion international, 24 Suppl 1, i37-i44. 60. National Health Service (2008), Remote and Rural Steering Group. NHS, Edinburgh, UK. 61. Australian Government Department of Health and Ageing (2011) Improving Primary Health Care for All Australians. Commonwealth Government of Australia Publications, Canberra. 62. Cornwall, A. (2008) Unpacking ‘Participation’: models, meanings and practices. Community Development Journal, 43, 269-283. 63. Anderson, E., Shepherd, M. and Salisbury, C. (2006) 'Taking off the suit': engaging the community in primary health care decision‐making. Health Expectations, 9, 70-80. 64. Ritchie, D., Gnich, W., Parry, O. and Platt, S. (2008) 'People pull the rug from under your feet': barriers to successful public health programmes. BMC Public Health, 8, 173. 65. Chung, P., Grogan, C.M. and Mosley, J.E. (2012) Residents' perceptions of effective community representation in local health decision-making. Social Science & Medicine, 74, 1652-1659. 66. Charles, C. and DeMaio, S. (1993) Lay participation in health care decision making: a conceptual framework. Journal of Health Politics, Policy and Law, 18, 881-904. 67. OECD (2009) Focus on Citizens: Public Engagement for Better Policy and Services. OECD Publications, Paris, France. 68. Molyneux, S., Atela, M., Angwenyi, V. and Goodman, C. (2012) Community accountability at peripheral health facilities: a review of the empirical literature and development of a conceptual framework. Health policy and planning, 27, 541-554. 69. Bryson, L. and Mowbray, M. (1981) Community: the spray-on solution. Australian Journal of Social Issues, 16, 255-267. 70. White, D. (2000) In Gary L Albrecht, R. F., Susan C Scrimshaw (ed.), Handbook of social studies in health and medicine. SAGE Publications, London, UK, pp. 465-480. 71. Tritter, J.Q. (2009) Revolution or evolution: the challenges of conceptualizing patient and public involvement in a consumerist world. Health Expectations, 12, 275-287. 72. Attree, P., French, B., Milton, B., Povall, S., Whitehead, M. and Popay, J. (2011) The experience of community engagement for individuals: a rapid review of evidence. Health & social care in the community, 19, 250-260. 73. Morgan, L.M. (2001) Community participation in health: perpetual allure, persistent challenge. Health policy and planning, 16, 221-230. 74. Kahssay, H.M. and Oakley, P. (1999) Community involvement in health development: a review of the concept and practice. WHO Publications, Geneva.

257 | P a g e

75. Bracht, N. and Tsouros, A. (1990) Principles and strategies of effective community participation. Health Promotion International, 5, 199-208. 76. Smithies, J. and Webster, G. (1998) Community involvement in health: from passive recipients to active participants. Ashgate Aldershot Publishers, UK. 77. Wilcox, D. (1994) The guide to effective participation. Partnership Brighton Publishers, UK. 78. CDC (1997) Principles of community engagement (1st edition). CDC Publications, Atlanta. 79. Burke, B. (1998) Evaluating for a change: Reflections on participatory methodology. New Directions for Evaluation, 1998, 43-56. 80. Taylor, J., Wilkinson, D., & Cheers, B. (2006) Is it consumer or community participation? Examining the links between ‘community’and ‘participation’. Health Sociology Review, 15, 38- 47. 81. Jewkes, R. and Murcott, A. (1996) Meanings of community. Social science & medicine (1982), 43, 555-563. 82. Brown, I. (1994) Community and participation for general practice: perceptions of general practitioners and community nurses. Social science & medicine (1982), 39, 335-344. 83. Baum, F. (1996) Community health services and managerialism. Australian Journal of Primary Health 2.4, 31-41. 84. Abelson, J., Forest, P.G., Eyles, J., Smith, P., Martin, E. and Gauvin, F.P. (2003) Deliberations about deliberative methods: issues in the design and evaluation of public participation processes. Social science & medicine (1982), 57, 239-251. 85. McMichael, A.J. and Beaglehole, R. (2000) The changing global context of public health. The Lancet, 356, 495-499. 86. Sindall, C. (1999) Health reform: public success, private failure. Health promotion international, 14, 377-380. 87. Ryan, C.J., Santucci, M.A., Gattuso, M.C., Czurylo, K., O'Brien, J. and Stark, B. (2001) Perceptions about advance directives by nurses in a community hospital. Clin Nurse Spec, 15, 246-252. 88. Gideon, J. (2005) ‘Consultation’or co-option? A case study from the Chilean health sector. Progress in Development Studies, 5, 169-181. 89. Johnson, S. (1996) Management for primary health care. Australian Journal of Primary Health, 2, 98-106. 90. Taylor, J., Wilkinson, D., & Cheers, B. (2002) Understanding community participation in rural health service development. Rural Social Work 7, 36-46. 91. Atienza, A.A. and King, A.C. (2002) Community-based health intervention trials: an overview of methodological issues. Epidemiologic reviews, 24, 72-79. 92. Morgan, L.M. (2001) Community participation in health: perpetual allure, persistent challenge. Health policy and planning, 16, 221-230. 93. Merzel, C. and D'Afflitti, J. (2003) Reconsidering community-based health promotion: promise, performance, and potential. American journal of public health, 93, 557-574. 94. Donner, A. and Klar, N. (1996) Statistical considerations in the design and analysis of community intervention trials. Journal of clinical epidemiology, 49, 435-439. 95. Sorensen, G., Emmons, K., Hunt, M.K. and Johnston, D. (1998) Implications of the results of community intervention trials. Annual review of public health, 19, 379-416. 96. McLeroy, K.R., Norton, B.L., Kegler, M.C., Burdine, J.N. and Sumaya, C.V. (2003) Community- based interventions. American journal of public health, 93, 529-533. 97. Whitehead, T. (2002). Community based interventions, definitions and types. The cultural ecology of health and change. University of Maryland Publishing, Maryland, USA. 258 | P a g e

98. Chung, P., Grogan, C. M., & Mosley, J. E. (2012). Residents' perceptions of effective community representation in local health decision-making. Social science & medicine, 74(10), 1652-1659 99. Verspaandonk, R. (2001) Shaping relations between government and citizens: Future directions in public administration? Department of the Parliamentary Library of Australia. 100. Dryzek, J.S. (2002) Deliberative Democracy and Beyond [Electronic book]: Liberals, Critics, Contestations. Oxford University Press. 101. Organisation for Economic Cooperation and Development (2009) Focus on Citizens: Public Engagement for Better Policy and Services. OCED publications, Paris. 102. Charles, C. and DeMaio, S. (1993) Lay participation in health care decision making: a conceptual framework. Duke Univ Press, Vol. 18, pp. 881-904. 103. Holmes, B. (2011) Citizens' engagement in policymaking and the design of public services. Australian Government, Parliamentary Library. 104. Abels, G. (2007) Citizen Involvement in Public policy making: Does it improve democratic legitimacy and accountability. Interdiscriplinary Information Sciences, 13, 103-116. 105. Bridgman, P., & Davis, G. (2003). What use is a policy cycle? Plenty, if the aim is clear. Australian Journal of Public Administration, 62(3), 98-102. 106. Moran, T. (2010) The future of the Australian Public Service: challenges and opportunities (Conference Proceedings: Neil Walker Memorial Lecture). CPA, Australia. 107. Born, K. and Laupacis, A. (2012) Public engagement in Ontario's hospitals: opportunities and challenges. Healthcare quarterly, 15 Spec No, 16-20. 108. Irvin, R.A. and Stansbury, J. (2004) Citizen participation in decision making: is it worth the effort? Public Administration Review, 64, 55-65. 109. Saltman, R., Bankauskaite, V. and Vrangbaek, K. (2007) Decentralization in health care. McGraw-Hill International. 110. OECD (2001) Engaging Citizens in Policy-making: Information, Consultation and Public participation (Brief No. 10). OECD publications, Paris, France. 111. Michels, A., & De Graaf, L. (2010). Examining citizen participation: Local participatory policy making and democracy. Local Government Studies, 36(4), 477-491. 112. European Institute for Public Participation (2009) Public Participation in Europe: An international perspective. European Institute for Public Participation Publications. Bremen, Germany. 113. Obama, B. (2009) Transparency and Open Government (speech transcript) United States Government, Washington, DC. 114. Cameron, D. (2010) Big Society Speech (speech transcript) UK Government London, UK. 115. Rowe, G., & Frewer, L. J. (2000). Public participation methods: A framework for evaluation. Journal of Science, technology & human values, 25(1), 3-29. 116. Renn, O., Webler, T., Rakel, H., Dienel, P., & Johnson, B. (1993). Public participation in decision making: A three-step procedure. Policy Sciences, 26(3), 189-214. 117. Goodin, R. E., & Dryzek, J. S. (2006). Deliberative impacts: the macro-political uptake of mini-publics. Politics & society, 34(2), 219-244. 118. Fung, A. (2007). Democratic theory and political science: A pragmatic method of constructive engagement. American Political Science Review, 101(03), 443-458. 119. Taylor, M. (2007) Community participation in the real world: opportunities and pitfalls in new governance spaces. Urban studies, 44, 297-317.

259 | P a g e

120. Stith, S., Pruitt, I., Dees, J.E., Fronce, M., Green, N., Som, A. and Linkh, D. (2006) Implementing community-based prevention programming: a review of the literature. The journal of primary prevention, 27, 599-617. 121. Earth-Institute (2015) A History of Decentralization. Columbia University Press, USA, Vol. 2015. 122. Baran, P. (1964) On distributed communications networks. Journal of IEEE Transactions 12, 1- 9. 123. Crook, R (2003) Decentralization and poverty reduction in Africa: the politics of local-central relations. Public Admin Development, 17, 362-377. 124. Azfar, O., Kähkönen, S., Lanyi, A., Meagher, P., & Rutheford, D. (1999), Decentralization, governance and public services: the impact of institutional arrangements. Centre for Institutional Reform and the Informal Sector. University of Maryland, USA. 125. Church, J., Saunders, D., Wanke, M., Pong, R., Spooner, C. and Dorgan, M. (2002) Citizen participation in health decision-making: past experience and future prospects. Journal of Public Health Policy, 12-32. 126. Bossert, T. (1998). Analyzing the decentralization of health systems in developing countries: decision space, innovation and performance. Social science & medicine, 47(10), 1513-1527. 127. WHO (2002) Recommendations on management of decentralization of Health Care: The Fifty-fifth Session of the Regional Committee for South-East Asia. SEARO/WHO Publications, New Delhi 128. Loewenson, R. (1999) Public participation in health: making people matter. Institute of Development Studies, University of Sussex Publications, UK. 129. Korten, F.F. (1983) Community participation: a management perspective on obstacles and options. Bureaucracy and the poor: closing the gap, 181-200. 130. Bender, D.E. and Pitkin, K. (1987) Bridging the gap: the village health worker as the cornerstone of the primary health care model. Social Science & Medicine, 24, 515-528. 131. Shaikh, B.T., Kadir, M.M. and Pappas, G. (2007) Thirty years of Alma Ata pledges: Is devolution in Pakistan an opportunity for rekindling primary health care? Journal of the Pakistan Medical Association, 57, 259. 132. Macinko, J., Starfield, B. and Erinosho, T. (2009) The impact of primary healthcare on population health in low-and middle-income countries. The Journal of Ambulatory Care Management, 32, 150-171. 133. Helling, L., Serrano, R., & Warren, D. (2005) In Protection, W. B. S. (ed.), Discussion Paper. World Bank, Washington, DC, Vol. 535, pp. 1-79. 134. Goldfrank, B. (2007) The politics of deepening local democracy: decentralization, party institutionalization, and participation. Comparative Politics, 147-168. 135. Frumence, G., Nyamhanga, T., Mwangu, M. and Hurtig, A.K. (2012) Challenges to the implementation of health sector decentralization in Tanzania: experiences from Kongwa district council. Global health action, 6, 20983. 136. Work, R. (1999). The role of Participation and Partnership in Decentralised Governance: A brief). United Nations Development Program Publications, New York, USA. 137. Brownlea, A. (1987) Participation: myths, realities and prognosis. Social Science & Medicine, 25, 605-614. 138. Ramiro, L.S., Castillo, F.A., Tan-Torres, T., Torres, C.E., Tayag, J.G., Talampas, R.G. and Hawken, L. Community participation in local health boards in a decentralized setting: cases from the Philippines (2001). Oxford Univ Press, Vol. 16, pp. 61-69.

260 | P a g e

139. Campos-Outcalt, D., Kewa, K., & Thomason, J. (1995). Decentralization of health services in Western Highlands Province, Papua New Guinea: an attempt to administer health service at the subdistrict level. Social science & medicine, 40(8), 1091-1098. 140. Collins, C. D., Omar, M., & Tarin, E. (2002). Decentralization, health care and policy process in the Punjab, Pakistan in the 1990s. The International journal of health planning and management, 17(2), 123-146.. 141. Ahgren, B., and Runo Axelsson. . (2005) "Evaluating integrated health care: a model for measurement. International journal of integrated care 5. 142. Cabinet Office, Government of UK (2009) Power in Peoples’ Hands: Learning from the World’s Best Public Services. The Strategy Unit, Cabinet Office Publication. London, UK. 143. Mosca, I. (2007) Decentralization as a determinant of health care expenditure: empirical analysis for OECD countries. Applied Economics Letters,, 14, 511-515. 144. Magnussen, J., Vrangbaek, K. and Saltman, R. (2009) Nordic health care systems: recent reforms and current policy challenges. McGraw-Hill International. 145. Department of Health, UK (2008). A stronger local voice - A framework for creating a stronger local voice in the development of health and social care services. Department of Health Publications. UK, London. 146. Commonwealth Department of Health and Aged Care, Australian Government (2000) Improving health services through consumer participation: a resource guide for organization. Brown & Wilton Integrated Publishing, Canberra. 147. Government of Ireland (2009) Framework for Public & Service User Involvement in Health and Social Care Regulation in Ireland. Dublin, Ireland. 148. Irish Society for Quality and Safety in Healthcare (2009). “Now We’re Talking” A practical toolkit for public and patient involvement in healthcare. Dublin, Ireland. 149. Murray, C.J., Lauer, J. A., & Evans, D. B. (2001) Measuring overall health system performance for 191 countries. WHO Publications, Geneva. 150. Vuorenkosky, L., Mladovsky, P. and Mossialos, E. (2008). Finland: Health system review. World Health Organization and European Observatory on Health Systems and Policies. Vol. 10, pp. 1-168. 151. Rifkin, S.B. (1983) Primary health care in Southeast Asia: attitudes about community participation in community health programmes. Social Science & Medicine, 17, 1489-1496. 152. Rossi, P.H.F., H.E., & Lipsey, M.W. (2004) Evaluation: A Systematic Approach. 7th Ed ed. SAGE, Thousand Oaks CA, USA. 153. Pickin, C., Popay, J., Staley, K., Bruce, N., Jones, C. and Gowman, N. (2002) Developing a model to enhance the capacity of statutory organisations to engage with lay communities. Journal of Health Services Research & Policy, 7, 34-42. 154. Kelly, K., & Van Vlaenderen, H. (1995). Evaluating participation processes in community development. Evaluation and Program Planning, 18(4), 371-383. 155. Arah, O.A., Klazinga, N.S., Delnoij, D.M., ten Asbroek, A.H. and Custers, T. (2003) Conceptual frameworks for health systems performance: a quest for effectiveness, quality, and improvement. International journal for quality in health care, 15, 377-398. 156. Arnstein, S. R. (1969). A ladder of citizen participation. Journal of the American Institute of planners, 35(4), 216-224. 157. Cohen, J.M., & Uphoff, N. T. (1980) Participation's place in rural development: seeking clarity through specificity. World development, 8, 213-235. 158. Gaventa, J. (2005). Reflections on the uses of the ‘power cube’approach for analyzing the spaces, places and dynamics of civil society participation and engagement. Prepared for 261 | P a g e

Dutch CFA Evaluation ‘Assessing Civil Society Participation as Supported In-Country by Cordaid, Hivos, Novib and Plan Netherlands. 159. Evans, E. (2002) Measuring Success in Public Participation (conference presentation) International Association for Public Participation, Australian National Conference, Australia. 160. Molyneux, S., Atela, M., Angwenyi, V. and Goodman, C. (2012) Community accountability at peripheral health facilities: a review of the empirical literature and development of a conceptual framework. Health policy and planning, 27, 541-554. 161. Butterfoss, F.D. (2006) Process evaluation for community participation. Annual review of public health, 27, 323-340. 162. OECD (2001) Engaging Citizens in Policy-making: Information, Consultation and Public participation. OECD publications, Paris, France. 163. McCabe, A. (2010) Below the Radar in a Big Society? Reflections on community engagement, empowerment and social action in a changing policy context. Working Paper 87. University of Birmingham Publications, UK. 164. McInerney, C., & Adshead, M. (2010). The Challenge of Community Participation in the Delivery of Public Services. NESF: Dublin NESC (2009) Well-being Matters: A Social Report for Ireland. NESC: Dublin. 165. National Health Service, UK (1999). Patient and public involvement in the new National Health Service, UK. Department of Health, NHS. 166. Evans, D., Pilkington, P. and McEachran, M. (2010) Rhetoric or reality? A systematic review of the impact of participatory approaches by UK public health units on health and social outcomes. Journal of Public Health, 32, 418-426. 167. Simpson, E.L. and House, A.O. (2002) Involving users in the delivery and evaluation of mental health services: systematic review. BMJ, 325, 1265. 168. McCoy, D.C., Hall, J.A. and Ridge, M. (2012) A systematic review of the literature for evidence on health facility committees in low-and middle-income countries. Health policy and planning, 27, 449-466. 169. Repper, J. and Breeze, J. (2007) User and carer involvement in the training and education of health professionals: a review of the literature. International Journal of Nursing Studies, 44, 511-519. 170. Crawford, M.J., Rutter, D., Manley, C., Weaver, T., Bhui, K., Fulop, N. and Tyrer, P. (2002) Systematic review of involving patients in the planning and development of health care. BMJ (Clinical research ed.), 325, 1263. 171. Morgan, M.W., Zamora, N., & Hindmarsh, M. F. . (2007) An inconvenient truth: a sustainable healthcare system requires chronic disease prevention and management transformation. Healthcare Papers, 7, 6. 172. Pappaioanou, M., Malison, M., Wilkins, K., Otto, B., Goodman, R.A., Churchill, R.E., White, M. and Thacker, S.B. (2003) Strengthening capacity in developing countries for evidence-based public health: the data for decision-making project. Social science & medicine (1982), 57, 1925-1937. 173. Samb, B., Desai, N., Nishtar, S., Mendis, S., Bekedam, H., Wright, A., Hsu, J., Martiniuk, A., Celletti, F., Patel, K. et al. (2010) Prevention and management of chronic disease: a litmus test for health-systems strengthening in low-income and middle-income countries. The Lancet, 376, 1785-1797. 174. Litsios, S. (2002). The long and difficult road to Alma-Ata: a personal reflection. International journal of health services, 32(4), 709-732.

262 | P a g e

175. De Rivero, D.A.T. and David, A. (2003) Alma-Ata Revisited. Perspectives in Health Magazine: The Magazine of the Pan American Health Organization, 8, 1-6. 176. WHO (2004) Comprehensive Community and Home-Based Health Care Model. SEARO Regional Publication,New Delhi. 40, 16-1. 177. Porter, C. (2007) Ottawa to Bangkok: changing health promotion discourse. Health promotion international, 22, 72-79. 178. Gulland, A. (2012). World leaders agree to cut deaths from non-communicable diseases by a quarter by 2025. BMJ, 344. 179. WHO (2002) Integrating prevention into health care (Fact sheet), WHO Publications, Geneva. 180. Pruitt, S.D. and Epping-Jordan, J.E. (2005) Preparing the 21st century global healthcare workforce. BMJ, 330, 637-639. 181. World Health Assembly (2000) Resolution for Global strategy for the prevention and control of non-communicable diseases (WHA53.17) 182. World Health Assembly (2003) Resoultion on Framework Convention on Tobacco Control (WHA56.1). 183. World Health Assembly (2004) WHA Resolution on Global strategy on diet, physical activity and health (WHA57.17) 184. World Health Assembly (2007) Secretariat’s report on implementation of the global strategy on NCDs (WHA60.23) 185. World Health Assembly (2008) Action Plan for the Global Strategy for the Prevention and Control of Non-communicable Diseases (WHA61.14). 186. World Health Assembly (2010) Secretariat’s report on implementation of the global strategy on NCDs (WHA63.13). 187. World Health Assembly (2011) Resolution on role of WHO in UN General Assembly High level meeting (WHA64.11). 188. Narain, J.P., Garg, R. and Fric, A. (2011) Non-communicable diseases in the South-East Asia region: burden, strategies and opportunities. The National medical journal of India, 24, 280- 287. 189. Beaglehole, R., Bonita, R., Horton, R., Adams, C., Alleyne, G., Asaria, P., Baugh, V., Bekedam, H., Billo, N. and Casswell, S. (2011) Priority actions for the non-communicable disease crisis. The Lancet, 377, 1438-1447. 190. Riley, B.L., Taylor, S.M. and Elliott, S.J. (2001) Determinants of implementing heart health: promotion activities in Ontario public health units: a social ecological perspective. Health education research, 16, 425-441. 191. Putland, C., Baum, F. and Macdougall, C. (1997) How can health bureaucracies consult effectively about their policies and practices?: some lessons from an Australian study. Health Promotion International, 12, 299-309. 192. Bartram, M., Abell, A. and Murphy, R. (1999). National Consumer Council, Great Britain. 193. Kent, H. and Read, J. Measuring consumer participation in mental health services: are attitudes related to professional orientation? International Journal of Social Psychiatry, 44, 295-310. 194. Bhatti-Sinclair, K. (1999) Evaluating social work and medical practice with black and ethnic minority groups using the clinical audit model. British Journal of Social Work, 29, 303-320. 195. Abbott-Chapman, J., Denholm, C. and Wyld, C. (2007) Pre-service professionals' constructs of adolescent risk-taking and approaches to risk management. Journal of sociology, 43, 241-261. 263 | P a g e

196. Abbott-Chapman, J. and Easthope, G. (1998) Participation and inclusion-a discourse of welfare or a discourse of rights? Australian Journal of Social Issues, 33, 99-118. 197. Yang, K. and Callahan, K. (2007) Citizen involvement efforts and bureaucratic responsiveness: Participatory values, stakeholder pressures, and administrative practicality. Public Administration Review, 67, 249-264. 198. Pruitt, S.D. and Epping-Jordan, J.E. (2005) Preparing the 21st century global healthcare workforce. BMJ, 330, 637-639. 199. Coleman, K., Austin, B.T., Brach, C. and Wagner, E.H. (2009) Evidence on the Chronic Care Model in the new millennium. Health affairs, 28, 75-85. 200. Bowl, R. (1996) Involving service users in mental health services: social services departments and the National Health Service and Community Care Act 1990. . Journal of Mental Health, 5, 287-304. 201. Freyens, P., Mbakuliyemo, N. and Martin, M. (1993) How do health workers see community participation? World Health Forum, 14, 253-257. 202. Hogg, R. and Hanley, J. (2008) Community development in primary care: opportunities and challenges. Community Pract, 81, 22-25. 203. McCann, T.V., Baird, J., Clark, E. and Lu, S. (2008) Mental health professionals' attitudes towards consumer participation in inpatient units. J Psychiatr Ment Health Nurs, 15, 10-16. 204. Bryant, J., Saxton, M., Madden, A., Bath, N. and Robinson, S. (2008) Consumers' and providers' perspectives about consumer participation in drug treatment services: is there support to do more? What are the obstacles? Drug Alcohol Rev, 27, 138-144. 205. Kidd, S., Kenny, A. and Endacott, R. (2007) Consumer advocate and clinician perceptions of consumer participation in two rural mental health services. Int J Ment Health Nurs, 16, 214- 222. 206. Happell, B., & Roper, C. . (2002) Promoting consumer participation through the implementation of a consumer academic position. Nurse Education in Practice, 2, 73-79. 207. Brown, I. (2001) Organizational values in general practice and public involvement: case studies in an urban district. Health Soc Care Community, 9, 159-167. 208. Brandstetter, S., McCool, M., Wise, M. and Loss, J. (2014) Australian health promotion practitioners' perceptions on evaluation of empowerment and participation. Health promotion international, 29, 70-80. 209. Nathan, S., Braithwaite, J. and Stephenson, N. (2013) Facilitating the action of community representatives in a health service: the role of a community participation coordinator. BMC health services research, 13, 154. 210. Haigh, F.A. and Scott-Samuel, A. (2008) Engaging communities to tackle anti-social behaviour: a health impact assessment of a citizens' jury. Public health, 122, 1191-1198. 211. Fridinger, F., Goodwin, G. and Chng, C.L. (1992) Physician and consumer attitudes and behaviors regarding self-help health support groups as an adjunct to traditional medical care. J Health Soc Policy, 3, 19-36. 212. Rifkin, S.B. (1983) Primary health care in Southeast Asia: attitudes about community participation in community health programmes. Social science & medicine (1982), 17, 1489- 1496. 213. Simpson, W.H. (1981) Reaction of a harassed administrator. Community Ment Health J, 17, 98-104. 214. Sheikh, K., Gilson, L., Agyepong, I.A., Hanson, K., Ssengooba, F. and Bennett, S. (2011) Building the field of health policy and systems research: framing the questions. PLoS medicine, 8, e1001073. 264 | P a g e

215. Ghaffar, A., Tran, N.T., Reddy, K.S., Kasonde, J., Bajwa, T., Ammar, W., Ren, M., Rottingen, J.A. and Mills, A. (2013) Changing mindsets in health policy and systems research. Lancet, 381, 436-437. 216. Viergever, R.F., Olifson, S., Ghaffar, A. and Terry, R.F. (2010) A checklist for health research priority setting: nine common themes of good practice. Health research policy and systems / BioMed Central, 8, 36. 217. Gross, K., Pfeiffer, C. and Obrist, B. (2012) "Workhood"-a useful concept for the analysis of health workers' resources? An evaluation from Tanzania. BMC health services research, 12, 55. 218. Perera, M. (2006) Intersectoral action for health in Sri Lanka. Institute for Health Policy Publications, Sri Lanka, Colombo. 219. Hornby, P. and Perera, H. (2002) A development framework for promoting evidence‐ based policy action: drawing on experiences in Sri Lanka. The International journal of health planning and management, 17, 165-183. 220. WHO. (2002) Marcoeconomics and health initiatives for Sri Lanka: investing for Health. WHO Publications, Sri Lanka. Colombo. 221. WHO. (2001) National Commission on Macroeconomics and Health. The case of Sri Lanka: First lessons and framework for comparing progress between countries. WHO publications, Geneva. 222. Hsiao, W.C. and Li, K. (2000) A preliminary assessment of Sri Lanka's health sector and steps forward. Institute of Policy Studies Publication, Sri Lanka. 223. Ministry of Health, Sri Lanka (2007) Health Master Plan Sri Lanka: Healthy & Shining Island in the 21st Century. Colombo, Sri Lanka. 224. Ministry of Health, Sri Lanka (2008). Annual Health Bullitin. MoH Publication, Colombo, Sri Lanka. 225. Mills, A., Vaughan, J.P., Smith, D.L. and Tabibzadeh, I. (1990) Health system decentralization: concepts, issues and country experience. WHO Publications, Geneva. 226. Lawn, J.E., Rohde, J., Rifkin, S., Were, M., Paul, V.K. and Chopra, M. (2008) Alma-Ata 30 years on: revolutionary, relevant, and time to revitalise. The Lancet, 372, 917-927. 227. Rosato, M., Laverack, G., Grabman, L.H., Tripathy, P., Nair, N., Mwansambo, C., Azad, K., Morrison, J., Bhutta, Z., Perry, H. et al. (2008) Community participation: lessons for maternal, newborn, and child health. The Lancet, 372, 962-971. 228. Pearson, M., Anthony, Z.B. and Buckley, N.A. (2010) Prospective policy analysis: how an epistemic community informed policymaking on intentional self poisoning in Sri Lanka. Health research policy and systems, 8, 19. 229. Withanachchi, N. and Uchida, Y. (2006) Healthcare rationing: a guide to policy directions in Sri Lanka. Health policy, 78, 17-25. 230. Yasuoka, J., Mangione, T.W., Spielman, A. and Levins, R. (2006) Impact of education on knowledge, agricultural practices, and community actions for mosquito control and mosquito- borne disease prevention in rice ecosystems in Sri Lanka. American Journal of Tropical Medicine & Hygiene, 74, 1034-1042. 231. Ranasinghe, P., Mendis, J. and Hanwella, R. (2011) Community psychiatry service in Sri Lanka: a successful model. Sri Lanka Journal of Psychiatry, 2, 3-5. 232. Jayawardene, W.P., Lohrmann, D.K., YoussefAgha, A.H. and Nilwala, D.C. (2011) Prevention of Dengue Fever: An Exploratory School‐Community Intervention Involving Students Empowered as Change Agents. Journal of School Health, 81, 566-573.

265 | P a g e

233. Konradsen, F., Pieris, R., Weerasinghe, M., van der Hoek, W., Eddleston, M. and Dawson, A.H. (2007) Community uptake of safe storage boxes to reduce self-poisoning from pesticides in rural Sri Lanka. BMC public health, 7, 13. 234. Wickramage, K., and Nellapalli, P. (2008) Community Participatory Methods in Disease Surveillance and Public Health in War-Affected Camps, and Its Potential Contribution to Peace Building. International Journal of Health Education, 11, 95-108. 235. Devendra, T. (1984) Sri Lanka: an innovative approach to the implementation of integrated parasite control, nutrition and family planning projects. Asian-Pacific population programme news, 13, 28-31. 236. Holmes, W.R. and Joseph, J. (2011) Social participation and healthy ageing: a neglected, significant protective factor for chronic non communicable conditions. Globalization and health, 7, 43. 237. Gammanpila , U.P. (1997) Effect of community participation on primary health care. (MSc Thesis Reference: 0560.D 561), University of Colombo. 238. Jayasekera, C. (2000) Assessment of male participation and the effectiveness of male- focused, occupational-based intervention in promoting male participation in selected areas of reproductive health, in the Kalutara educational zone. (MD thesis Reference: D1619. 0808), University of Colombo. 239. Perera, M. (1985) Social dynamics of voluntary health workers in primary health care. (MD thesis Reference: D 19. 1381), Colombo University. 240. Pathmanathan, I. and Liljestrand, J. (2003) Investing in maternal health: learning from Malaysia and Sri Lanka. World Bank Publications, Washington DC. 241. Zaidi, A.K., Awasthi, S. and Janaka deSilva, H. (2004) Burden of infectious diseases in South Asia. British Medical Journal, 328, 811-815. 242. Wickramage, K. and Galappaththy, G. (2013) Malaria burden in irregular migrants returning to Sri Lanka from human smuggling operations in West Africa and implications for a country reaching malaria elimination. Transactions of the Royal Society of Tropical Medicine and Hygiene, 107, 337-340. 243. Hevia, C. and Loayza, N. (2013) Saving and growth in Sri Lanka. World Bank Policy Research Working Paper. World Bank Publications, Washington DC. 244. Chandola, T. (2012) Spatial and social determinants of urban health in low-, middle-and high-income countries. Public health, 126, 259-261. 245. Vogli, R.D., Kouvonen, A., Elovainio, M. and Marmot, M. (2013) Economic globalization, inequality and body mass index: a cross-national analysis of 127 countries. Critical Public Health, 1-15. 246. De Silva, W.I. (2013) The Age Structure Transition and the Demographic Dividend: An Opportunity for Rapid Economic Take-off in Sri Lanka. Sri Lanka Journal of Advanced Social Studies, 2, 3-46. 247. Østbye, T., Chan, A., Malhotra, R. and Kothalawala, J. (2010) Adult children caring for their elderly parents: Results from a national survey of ageing in Sri Lanka. Asian Population Studies, 6, 83-97. 248. Jones, M. and de Silva, A. (2012) The Sri Lankan experience. Health Transitions and the Double Disease Burden in Asia and the Pacific: Histories of Responses to Non-Communicable and Communicable Diseases, 126. 249. Mendis, S. and Ekanayake, E. (1994) Prevalence of coronary heart disease and cardiovascular risk factors in middle aged males in a defined population in central Sri Lanka. International journal of cardiology, 46, 135-142. 266 | P a g e

250. Jayawardena, R., Byrne, N.M., Soares, M.J., Katulanda, P. and Hills, A.P. (2012) The obesity epidemic in Sri Lanka revisited. Asia-Pacific Journal of Public Health. 251. Katulanda, P., Jayawardena, M., Sheriff, M., Constantine, G. and Matthews, D. (2010) Prevalence of overweight and obesity in Sri Lankan adults. Obesity reviews, 11, 751-756. 252. Katulanda, P., Rathnapala, D., Sheriff, R. and Matthews, D. (2012) Province and ethnic specific prevalence of diabetes among Sri Lankan adults. Sri Lanka Journal of Diabetes Endocrinology and Metabolism, 1, 2-7. 253. Rannan-Eliya, R.P., & Sikurajapathy, L. (2008), Good practices in health financing: lessons from reforms in low-and middle-income countries. World Bank, Washington, DC, pp. 311-354. 254. Wickramage, K. (2013) Health Work for Shared Societies. Public Policies in Shared Societies: A Comparative Approach, 96. 255. Ministry of Health, Sri Lanka (2003). Situation Analysis, Healthy and Shining Island in the 21st Century. Ministry of Health Publications, Colombo, Sri Lanka. 256. Beaglehole, R., Epping-Jordan, J., Patel, V., Chopra, M., Ebrahim, S., Kidd, M., & Haines, A. (2008). Improving the prevention and management of chronic disease in low-income and middle-income countries: a priority for primary health care. The Lancet, 372(9642), 940- 949. 257. Badara, S, Alwan, A. and Etienne, C. (2011) Commonwealth Health Ministers' Health Update 2011. Commonwealth Secretariat. 258. Ministry of Health , Sri Lanka (2012) Approach and Guidelines for strengthening health care at primary level. Ministry of Health Publications, Colombo, Sri Lanka. 259. Ministry of Health , Sri Lanka (2008) Report on the consultative workshop on Primary care level public health staff categories and their work responsibilities. Policy Analysis and Planning Unit, Ministry of Health Publications, Colombo, Sri Lanka. 260. WHO (2014). The WHO Health Systems Framework. WHO-Western Pacific Regional Office Publications. 261. De Savigny, D., & Adam, T. (2009) Systems thinking for health systems strengthening. World Health Organization Publications, Geneva. 262. Van Olmen, J., Criel, B., Bhojani, U., Marchal, B., Van Belle, S., Can Dormael, M., Hoeree, T., Pirard, M and Kegels, G. (2012) Analysing health system dynamics: a framework. Studies in Health Services Organisation & Policy, 28, 117. 263. Van Olmen, J., Criel, B., Bhojani, U., Marchal, B., Van Belle, S., Chenge, M. F., Hoerée, T.,Pirard, M., Van Damme, W & Kegels, G. . (2012) The Health System Dynamics Framework: The introduction of an analytical model for health system analysis and its application to two case-studies. . Health, Culture and Society, 2, 1-21. 264. Atun, R.A., Lennox-Chhugani, N., Drobniewski, F., Samyshkin, Y.A. and Coker, R.J. (2004) A framework and toolkit for capturing the communicable disease programmes within health systems: tuberculosis control as an illustrative example. European journal of public health, 14, 267-273. 265. Hoffman, S.J., Røttingen, J. A., Bennett, S., Lavis, J. N., Edge, J. S., & Frenk, J. (2012) Background paper on conceptual issues related to health systems research to inform a WHO global strategy on health systems research. WHO Publications, Geneva. 266. Kutzin, J. (2001) A descriptive framework for country-level analysis of health care financing arrangements. Health policy, 56, 171-204. 267. Marshall M, R.M., Campbell S, Kirk S, ReevesD, Brook R, McGlynn E. (2003). Measuring general practice: a demonstration project to develop and test a set of primary care clinical quality indicators. The Nuffield Trust Publications, London, UK. 267 | P a g e

268. Mounier-Jack, S., Griffiths, U.K., Closser, S., Burchett, H. and Marchal, B. (2014) Measuring the health systems impact of disease control programmes: a critical reflection on the WHO building blocks framework. BMC public health, 14, 278. 269. McLaren, L. and Hawe, P. (2005) Ecological perspectives in health research. Journal of epidemiology and community health, 59, 6-14. 270. Kickbusch, I. (1989) Approaches to an ecological base for public health. Health promotion international, 4, 265-268. 271. Lang, T. and Rayner, G. (2012) Ecological public health: the 21st century's big idea? An essay by Tim Lang and Geof Rayner. BMJ (Clinical research ed.), 345, e5466. 272. Stokols, D. (1996) Translating social ecological theory into guidelines for community health promotion. American journal of health promotion : AJHP, 10, 282-298. 273. Green, L.W., Richard, L. and Potvin, L. (1996) Ecological foundations of health promotion. American journal of health promotion : AJHP, 10, 270-281. 274. Scott, A., Ejikeme, C.S., Clottey, E.N. and Thomas, J.G. (2013) Obesity in sub-Saharan Africa: development of an ecological theoretical framework. Health promotion international, 28, 4- 16. 275. Lang, T. and Rayner, G. (2007) Overcoming policy cacophony on obesity: an ecological public health framework for policymakers. Obesity reviews : an official journal of the International Association for the Study of Obesity, 8 Suppl 1, 165-181. 276. Parcel, G.S., Perry, C. L. and Taylor, W. C. (1990) In Bracht, N. (ed.), Health Promotion at the Community Level. SAGE, Newbury Park, CA, pp. pp. 229–251. 277. Orlandi, M.A. (1996) Health promotion technology transfer: organizational perspectives. Canadian journal of public health, 87 Suppl 2, S28-33. 278. Nutbeam, D. (1998). Theory in a Nutshell: A Practitioner's Guide to Commonly Used Theories and Models in Health Promotion. University of Sydney Publication, Australia. 279. Stokols, D. (1992) Establishing and maintaining healthy environments. Toward a social ecology of health promotion. The American psychologist, 47, 6-22. 280. Shiell, A., Hawe, P. and Gold, L. (2008) Complex interventions or complex systems? Implications for health economic evaluation. BMJ, 336, 1281-1283. 281. Begun JW, Z.B., Dooley KJ. . (2003) In Mick SS, W. M. (ed.), Advances in health care organization theory 8. Jossey-Bass, San Francisco, USA, pp. 253-258. 282. Atun, R., de Jongh, T., Secci, F., Ohiri, K. and Adeyi, O. (2010) Integration of targeted health interventions into health systems: a conceptual framework for analysis. Health policy and planning, 25, 104-111. 283. Medical Research Council, UK (2000). A framework for the development and evaluation of randomised controlled trials for complex interventions to improve health. MRC, London, UK. 284. Bar-Yam, Y. (2006) Improving the effectiveness of health care and public health: a multiscale complex systems analysis. American journal of public health, 96, 459-466. 285. Frenk, J. (2010) The global health system: strengthening national health systems as the next step for global progress. PLoS medicine, 7, e1000089. 286. Campbell, N.C., Murray, E., Darbyshire, J., Emery, J., Farmer, A., Griffiths, F. (2007) Designing and evaluating complex interventions to improve health care. . British Medical Journal, 334, 455. 287. Atun, R.A., Menabde, N., Saluvere, K., Jesse, M. and Habicht, J. (2006) Introducing a complex health innovation--primary health care reforms in Estonia (multimethods evaluation). Health policy, 79, 79-91.

268 | P a g e

288. Marchal, B., Van Belle, S., De Brouwere, V. and Witter, S. (2013) Studying complex interventions: reflections from the FEMHealth project on evaluating fee exemption policies in West Africa and Morocco. BMC health services research, 13, 469. 289. Silverman, D. (2013) Doing qualitative research: A practical handbook. SAGE publications. 290. Creswell, J. W. (2003). Research Design: Quantitative and Qualitative Approaches. (2nd ed) SAGE publications, Thousand Oaks, California. 291. Huberman, A. M., & Miles, M. B. (1994). Data management and analysis methods. pp. 428- 444. SAGE publications, Thousand Oaks, California. 292. Creswell, J.W., Fetters, M.D. and Ivankova, N.V. (2004) Designing a mixed methods study in primary care. Annals of family medicine, 2, 7-12. 293. Mackenzie, N., & Knipe, S. (2006) Research dilemmas: Paradigms, methods and methodology. Issues in educational research, 16, 193-205. 294. Greene, J.C., Caracelli, V. J., & Graham, W. F. (1989) Toward a conceptual framework for mixed-method evaluation designs. . Educational evaluation and policy analysis, 11, 255-274. 295. Barbour, R.S. (1999) The case for combining qualitative and quantitative approaches in health services research. J Health Serv Res Policy, 4, 39-43. 296. Baxter, P., & Jack, S. (2008) Qualitative Case Study Methodology: Study Design and Implementation for Novice Researchers. The Qualitative Report, 13, 544-559. 297. Hyett, N., Kenny, A., & Virginia Dickson-Swift, D. (2014) Methodology or method? A critical review of qualitative case study reports.International journal of qualitative studies on health and well-being, 9. 298. Miles, M. and Huberman, A.M. (1994) Qualitative Data Analysis: An Expanded Sourcebook. 2nd ed. ed. SAGE, Thousand Oaks, CA, USA. 299. Coyne, I.T. (1997) Sampling in qualitative research. Purposeful and theoretical sampling; merging or clear boundaries? Journal of advanced nursing, 26, 623-630. 300. Tanner, M. (2005) Strengthening district health systems. Bulletin of the World Health Organization, 83, 403-403. 301. Pope, C., van Royen, P. and Baker, R. (2002) Qualitative methods in research on healthcare quality. Quality & safety in health care, 11, 148-152. 302. Small, M.L. (2008) Interdisciplinary standards for systematic qualitative research. National Science Foundation Publication. Washington, DC, USA. 303. Sandelowski, M. (1995) Sample size in qualitative research. Research in nursing & health, 18, 179-183. 304. Ministry of Health (2008) Sri Lanka Evaluation on District Plans implemented under Health Sector Development Project. Management Development and Planning Unit, MoH publications. Colombo, Sri Lanka. 305. Carlsen, B. and Glenton, C. (2011) What about N? A methodological study of sample-size reporting in focus group studies. BMC medical research methodology, 11, 26. 306. Guba, E.G., and Y.S. Lincoln. 1994. Competing paradigms in qualitative research. Pp. 163– 194 in Handbook of qualitative research, vol. 2. London: Sage 307. Neuman, W.L. (2003) Social research methods: Qualitative and quantitative approaches (5th ed.). Allyn & Bacon, Boston, USA 308. DiCicco‐Bloom, B., & Crabtree, B. F. (2006). The qualitative research interview. Medical education, 40(4), 314-321. 309. MacLean, L.M., Meyer, M. and Estable, A. (2004) Improving accuracy of transcripts in qualitative research. Qualitative health research, 14, 113-123.

269 | P a g e

310. Twohig, P.L. and Putnam, W. (2002) Group interviews in primary care research: advancing the state of the art or ritualized research? Family practice, 19, 278-284. 311. Lake, A.J. and Staiger, P.K. (2010) Seeking the views of health professionals on translating chronic disease self-management models into practice. Patient education and counseling, 79, 62-68. 312. Duggleby, W. (2005) What about focus group interaction data? Qualitative health research, 15, 832-840. 313. Milewa, T., Dowswell, G. and Harrison, S. (2002) Partnerships, power and the “new” politics of community participation in British health care. Social Policy & Administration, 36, 796-809. 314. Kelly, K. and Van Vlaenderen, H. (1995) Evaluating participation processes in community development. Evaluation and program planning, 18, 371-383. 315. Marmot, M. (2005) Social determinants of health inequalities. The Lancet, 365, 1099-1104. 316. Roper JM, S.J. (2000) Ethnography in Nursing Research. SAGE, Thousand Oaks CA, USA. 317. MH., A. (1997) Ethnography: an overview. Substance use & misuse, 32, 1155-1173. 318. Ellsberg, M.C., & Heise, L. (2005) Researching violence against women. WHO Publications. 319. Higginbottom, G., Pillay, J. J., & Boadu, N. Y. . (2013) Guidance on performing focused ethnographies with an emphasis on healthcare research. The Qualitative Report, 18, 1-6. 320. Knoblauch, H. (2005) Focused Ethnography. Qualitative Social Research, 6. 321. Cruz, E.V., & Higginbottom, G. (2013) The use of focused ethnography in nursing research. Nurse researcher, 20, 36-43. 322. Kilian, C., Salmoni, A., Ward-Griffin, C. and Kloseck, M. (2008) Perceiving falls within a family context: a focused ethnographic approach. Canadian journal on aging = La revue canadienne du vieillissement, 27, 331-345. 323. Savage, J. (2006) Ethnographic evidence the value of applied ethnography in healthcare. Journal of Research in Nursing, 11, 383-393. 324. Muecke, M. (1994) In (Ed), M. J. (ed.), Critical Issues in Qualitative Research Methods. SAGE., Thousand Oaks CA, USA. 325. Spiers, J.A. and Wood, A. (2010) Building a therapeutic alliance in brief therapy: the experience of community mental health nurses. Archives of psychiatric nursing, 24, 373-386. 326. Croughan-Minihane, M.S., Thom, D.H. and Petitti, D.B. (1999) Research interests of physicians in two practice-based primary care research networks. West J Med, 170, 19-24. 327. Hart, E. (2004) Applications of anthropology in the National Health Service. . Anthropology in Action 11, 21–23. 328. Carthey, J. (2003) The role of structured observational research in health care. Quality and safety in health care, 12, ii13-ii16. 329. Patton, M.Q. (1990) Qualitative evaluation and research methods. SAGE Publications, USA. 330. Bowen, G.A. (2009) Document analysis as a qualitative research method. Qualitative research journal, 9, 27-40. 331. Colebatch, H.K. (2006) The work of policy: An international survey. Lexington Books, UK. 332. Sri Lankan Ministry of Justice (2013) Sri Lankan Ministry of Justice (SLMOJ) 'Law Net'. Colombo, Sri Lanka. URL: http://www.lawnet.lk/index.php 333. Commonwealth Legal Information Institute Sri Lanka (2013) Commonwealth and Common Law database. URL: http://www.commonlii.org/resources/264.html 334. National Consumer and Carer Forum of Australia (2005) Best Practice Principles for Inclusion in a Participation Policy. Canberra, Australia.

270 | P a g e

335. Victorian Government and Banyule-Nillumbik Primary Care Alliance (2000) Consumer Participation Resource & Training Kit for Service Providers. Victorian Government Publication, Victoria, Australia. 336. Monson, K. and Thurley, M. (2011) Consumer participation in a youth mental health service. Early intervention in psychiatry, 5, 381-388. 337. Alliston, C., Kluge, C. and Fudge, E. (2009) Commentary on consumer and carer participation in a national mental health promotion initiative. Advances in Mental Health, 8, 305-314. 338. Van Schaik, L. and Egenhofer, C. (2006) Policy coherence for development in the eu Council: strategies for the way forward. CEPS. 339. Mirza, Z., Krattiger, A., Taubman, A., Bartels, H.G., Beyer, P., Kampf, R. and Watal, J. (2013) Policy coherence for improved medical innovation and access. Bulletin of the world Health Organization, 91, 315-315. 340. Coulter, A., Parsons, S., & Askham, J. (2008) Where are the patients in decision-making about their own care? WHO-EURO Publications, Brussels, Belgium. 341. DeRoeck, D. (2004) The Importance of Engaging Policy-makers at the Outset to Guide Research on and Introduction of Vaccines: The Use of Policy-maker Surveys. Journal of Health Popultiona and Nurition, 22, 322-330. 342. Avey, P.C., and Desch, M. C. (2014) What Do Policymakers Want From Us? Results of a Survey of Current and Former Senior National Security Decision Makers. International Studies Quarterly, 58, 227-246. 343. Mitton, C. and Donaldson, C. (2002) Setting priorities in Canadian regional health authorities: a survey of key decision makers. Health policy, 60, 39-58. 344. Government of Sri Lanka (2012) The Gazette extraordinary of the Democratic Socialist Republic of Sri Lanka. Allocation of Powers, duties, functions of Departments, Statutory Institutions and Public Corporations. No. 1615/20. 30th April 2010. Presidential Secretariat, Colombo. Sri Lanka. 345. Hornby, P., & Perera, H. S. R. (2002) A development framework for promoting evidence‐ based policy action: drawing on experiences in Sri Lanka. The International journal of health planning and management, 17, 165-183. 346. Wright, K.B. (2005) Researching Internet based populations: Advantages and disadvantages of online survey research, online questionnaire authoring software packages, and web survey services. Journal of Computer Mediated Communication, 10. 347. Teng, F., Mitton, C., & MacKenzie, J. (2007). Priority setting in the provincial health services authority: survey of key decision makers. BMC health services research, 7(1), 84. 348. Vogel, J. P., Oxman, A. D., Glenton, C., Rosenbaum, S., Lewin, S., Gulmezoglu, A. M., & Souza, J. P. (2013). Policymakers’ and other stakeholders’ perceptions of key considerations for health system decisions and the presentation of evidence to inform those considerations: an international survey. Health Res Policy Syst, 11, 19. 349. Mitton, C., & Donaldson, C. (2002). Setting priorities in Canadian regional health authorities: a survey of key decision makers. Health Policy, 60(1), 39-58. 350. Sheehan, K. B. (2001). E‐mail survey response rates: A review. Journal of Computer Mediated Communication, 6(2), 0-0. 351. De Leeuw, E.D., Hox, J.J. and Dillman, D.A. (2008) Mixed-mode surveys: When and why. In (Ed). de Leeuw, J. J. Hox, & D. A. Dillman (Eds.), International handbook ofsurvey methodology (pp. 299–316).

271 | P a g e

352. Kaplowitz, M. D., Hadlock, T. D., & Levine, R. (2004). A comparison of web and mail survey response rates. Public opinion quarterly, 68(1), 94-101. 353. Kelley, K., Clark, B., Brown, V. and Sitzia, J. (2003) Good practice in the conduct and reporting of survey research. International journal for quality in health care : journal of the International Society for Quality in Health Care / ISQua, 15, 261-266. 354. Holloway, I., & Todres, L. . (2003) The status of method: flexibility, consistency and coherence. Qualitative research, 3, 345-357. 355. Dixon-Woods, M. and Critchley, S. (2000) Medical and lay views of irritable bowel syndrome. Family practice, 17, 108-113. 356. Strauss, A.C., J.M. (1990) Basics of Qualitative Research. SAGE, Newbury Park, California, USA. 357. Braun, V. (2006) Using thematic analysis in psychology. Qualitative Research in Psychology, 3, 77-101. 358. Ulin, P. R., Robinson, E. T., & Tolley, E. E. (2012). Qualitative methods in public health: a field guide for applied research. John Wiley & Sons. 359. Gilson, L., Hanson, K., Sheikh, K., Agyepong, I.A., Ssengooba, F. and Bennett, S. (2011) Building the field of health policy and systems research: social science matters. PLoS medicine, 8, e1001079. 360. Carter, S.M., Ritchie, J.E. and Sainsbury, P. (2009) Doing good qualitative research in public health: not as easy as it looks. New South Wales public health bulletin, 20, 105-111. 361. Malterud, K. (2001) Qualitative research: standards, challenges, and guidelines. The Lancet, 358, 483-488. 362. Glaser, B.G. (1992) Emergence vs forcing: Basics of grounded theory analysis. Sociology Press, Chicago, USA. 363. Provalis Research and Voyant Tools (2013). QDA miner software, version 4.0 (Text Analysis software). Provalis Research Sinclair, Canada. 364. Kvale, S. (1996) InterViews: An introduction to qualitative research interviewing. SAGE, Thousand Oaks, CA, USA. 365. Coggle (2014) Coggle - Mind Mapping software. England, URL: https://coggle.it. 366. Barbour, R.S. (2001) Checklists for improving rigour in qualitative research: a case of the tail wagging the dog? BMJ, 322, 1115-1117. 367. Lincoln, Y. S., & Guba, E. G. (1985) Naturalistic Inquiry. SAGE, Newbury Park, CA, USA. 368. Altheide, D.J., JM (1994). (1994) In Lincoln, N. D. Y. (ed.), Handbook of qualitative research. SAGE, Thousand Oaks, CA, pp. 485-499. 369. Popay, J., Rogers, A. and Williams, G. (1998) Rationale and standards for the systematic review of qualitative literature in health services research. Qualitative health research, 8, 341-351. 370. Hoddinott, P. and Pill, R. (1997) A review of recently published qualitative research in general practice. More methodological questions than answers? Family practice, 14, 313- 319. 371. Seale C, S.D. (1997) Ensuring rigour in qualitative research. European journal of public health, 7, 379-384. 372. Mays, N., Pope, C., & Popay, J. (2005). Systematically reviewing qualitative and quantitative evidence to inform management and policy-making in the health field. Journal of health services research & policy, 10(suppl 1), 6-20. 373. Spano, R. (2005) Potential Sources of Observer Bias in Police Observational Data. Social Science Research, 34, 591–617. 272 | P a g e

374. Monahan, T., & Fisher, J. A. (2010) Benefits of ‘observer effects’: lessons from the field. . Qualitative Research, 10, 357-376. 375. Lincoln, Y.S., & Guba, E. G. (1986) But is it rigorous? Trustworthiness and authenticity in naturalistic evaluation. New directions for program evaluation, 30, 73-84. 376. Reeves, S., Kuper, A. and Hodges, B.D. (2008) Qualitative research methodologies: ethnography. BMJ, 337, a1020. 377. Creswell, J.W., & Miller, D. L. (2000) Determining validity in qualitative inquiry. Theory into practice, 39, 124-130. 378. Morse, J.M., Barrett, M., Mayan, M., Olson, K., & Spiers, J. (2008) Verification strategies for establishing reliability and validity in qualitative research. International journal of qualitative methods 1, 13-22. 379. Appleton, J.V. (1995) Analysing qualitative interview data: addressing issues of validity and reliability. Journal of advanced nursing, 22, 993-997. 380. Douglas, S.P., & Craig, C. S. . (2007) Collaborative and iterative translation: an alternative approach to back translation. Journal of International Marketing, 15, 30-43. 381. Halcomb, E.J., & Davidson, P. M. . (2006) Is verbatim transcription of interview data always necessary?. . Applied Nursing Research, 19, 38-42. 382. Larkin, P.J., Dierckx de Casterle, B. and Schotsmans, P. (2007) Multilingual translation issues in qualitative research: reflections on a metaphorical process. Qualitative health research, 17, 468-476. 383. Mays, N. and Pope, C. (1995) Rigour and qualitative research. BMJ (Clinical research ed.), 311, 109-112. 384. Greenhalgh, T., Robert, G., Macfarlane, F., Bate, P., & Kyriakidou, O. . (2004) Diffusion of innovations in service organizations: systematic review and recommendations. Milbank Quarterly, 82, 581-629. 385. Agampodi, S., Wickramage, K. (2014) Maternal mortality revisited: the application of the new ICD-MM classification system in reference to maternal deaths in Sri Lanka. . BMC Reproductive health, 11, 17. 386. Government of Sri Lanka (2006) Public Administration Circular No.06/2006. Ministry of Public Administration and Home Affairs. Government of Sri Lanka, Colombo. 387. Rifkin, S. B., Hewitt, G., & Draper, A. (2007) Community participation in nutrition programmes for child survival and anemia. Centre for Public Health Nutrition Publication, University of Westminster, UK. 388. Ministry of Health (2010) National Guidelines for Improvement of Quality and Safety of Health Care Institutions. Government of Sri Lanka Colombo. 389. Colebatch, H.K. (2002) Policy (2nd Ed.). Open University Press, Buckingham, UK. 390. Tan, L.H., Fukasaku, K. and Plummer, M.G. (1995) OECD and ASEAN Economies: The Challenge of Policy Coherence. OECD Publication, Paris. 391. Fernando, D., Gunawardena, N. and Weerasinghe, C. (2006) Report on Rapid Assessment of Essential Public Health Functions-Sri Lanka.. University of Colombo, Department of Community Medicine, Faculty of Medicine, Sri Lanka., Colombo, Sri Lanka. 392. Ministry of Health (2003) Health Sector Master Plan – Sri Lanka: healthy and shining island. Government of Sri Lanka, Colombo. 393. Meier, B.M., Pardue, C. and London, L. (2012) Implementing community participation through legislative reform: a study of the policy framework for community participation in the Western Cape province of South Africa. BMC international health and human rights, 12, 15.

273 | P a g e

394. WHO (2011) The right to health in the constitutions of the member states of the WHO South East Asian Region. WHO/SEARO Publications, Delhi, India. 395. Government of Sri Lanka (2000) The Constitution of the Democratic Socialist Republic of Sri Lanka. URL: http://www.priu.gov.lk/Cons/1978Constitution/Index.html 396. Ranjith, J.S. (2007) Governance and Community Participation: A Collective Approach to Regional (Urban) Development in Sri Lanka. The Peradeniya Journal of Economics. 1, (1 & 2) (2007): 28-58. 397. WHO (1986) National Health Development Networks in support of Primary Health Care.WHO Publications, Genvea, Switzeland. 398. Goodhand, J. (2013) Sri Lanka in 2012: Securing the State, Enforcing the “Peace”. Asian Survey, 53, 64-72. 399. Athukorala, P.-C. and Jayasuriya, S. (2013) Economic Policy Shifts in Sri Lanka: The Post- Conflict Development Challenge. Asian Economic Papers, 12, 1-28. 400. Hensman, R. (2013). Independent Judiciary and Rule of Law: Demolished in Sri Lanka. Economic and Political Weekly, 48(09). 401. Bandara, S., Jayaratne, N. and Madurawala, S. (2012) Intersectoral Action for the Health in Addressing Social Determinants of Health through Public Policies in Sri Lanka: Health in All Policies. Institute of Policy Studies Publication, Colombo. 402. Franco, L.M., Bennett, S., & Kanfer, R. (2002) Health sector reform and public sector health worker motivation: a conceptual framework. Social science & medicine, 54, 1255-1266. 403. Hudson, B. (2002). Interprofessionality in health and social care: the Achilles' heel of partnership?. Journal of interprofessional care, 16(1), 7-17. 404. Franco, L. M., Bennett, S., & Kanfer, R. (2002). Health sector reform and public sector health worker motivation: a conceptual framework. Social science & medicine, 54(8), 1255-1266. 405. Mathauer, I., & Imhoff, I. (2006). Health worker motivation in Africa: the role of non-financial incentives and human resource management tools. Human resources for health, 4(1), 24. 406. De Silva, W.I. (1998) Puerperal morbidity: A neglected area of maternal health in Sri Lanka. . Bio demography and Social Biology, 45, 223-245. 407. Colombo Page (2014) Sri Lanka's midwives and nurses strike over controversy in midwifery training.[news article] [news article, Issue Date: Wed, Apr 2, 2014], URL: http://www.colombopage.com/archive_14A/Apr02_1396418081JR.php 408. McCann, T.V., Baird, J., Clark, E. and Lu, S. (2008) Mental health professionals' attitudes towards consumer participation in inpatient units. Journal of Psychiatric & Mental Health Nursing, 15, 10-16. 409. Brown I. (1994) Community and participation for general practice: perceptions of general practitioners and community nurses. Social Science & Medicine 39, 335-344. 410. Freyens, P., Mbakuliyemo, N. and Martin, M. (1993) How do health workers see community participation? World Health Forum, 14, 253-257. 411. Ryan, N.P., Wade, J.C., Nice, A., Shenefelt, H. and Shepard, K. (1996) Physical therapists' perceptions of family involvement in the rehabilitation process. Physiother Res Int, 1, 159- 179. 412. Bradley, E.H., Curry, L. A., & Devers, K. J. . (2007) Qualitative data analysis for health services research: developing taxonomy, themes, and theory. Health services research, 42, 1758-1772. 413. Veillard, J.H.M., Brown, A. D., Barış, E., Permanand, G., & Klazinga, N. S. (2011) Health system stewardship of national health ministries in the WHO european region: Concepts, functions and assessment framework. . Health policy, 103, 191-199. 274 | P a g e

414. Murray, C.J., & Frenk, J. . (2000) A framework for assessing the performance of health systems. Bulletin of the world Health Organization, 78, 717-731. 415. Allotey, P., Davey, T. and Reidpath, D.D. (2014) NCDs in low and middle-income countries - assessing the capacity of health systems to respond to population needs. BMC public health, 14 Suppl 2, S1. 416. WHO (2008) The world health report 2008: primary health care: now more than ever. World Health Organization Publications, Geneva. 417. Al-Mazroa, Y. and Al-Shammari, S. (1991) Community participation and attitudes of decision-makers towards community involvement in health development in Saudi Arabia. Bulletin of the world Health Organization, 69, 43. 418. Hooper, K., Thomas, Y. and Clarke, M. (2007) Health professional partnerships and their impact on Aboriginal health: An occupational therapist's and Aboriginal health worker's perspective. Australian Journal of Rural Health, 15, 46-51. 419. Simpson, W.H. (1981) Reaction of a harassed administrator. Community Mental Health Journal, 17, 98-104. 420. Goodwin, V. and Happell, B. (2008) Psychiatric nurses' attitudes toward consumer and carer participation in care: barriers to participation. Policy Polit Nurs Pract, 9, 249-256. 421. Speers, J. (2008) Service user involvement in the assessment of a practice competency in mental health nursing - stakeholders' views and recommendations. Nurse Educ Pract, 8, 112- 119. 422. Cook, J.A., Jonikas, J.A. and Razzano, L. (1995) A randomized evaluation of consumer versus nonconsumer training of state mental health service providers. Community Ment Health J, 31, 229-238. 423. Nathan, S., Harris, E., Kemp, L. and Harris-Roxas, B. Health service staff attitudes to community representatives on committees. Journal of Health Organization & Management, 20, 551-559. 424. Milewa, T. User participation in service planning. A qualitative approach to gauging the impact of managerial attitudes. Journal of Management in Medicine, 11, 238-245. 425. Hogg, R. and Hanley, J. (2008) Community development in primary care: opportunities and challenges. Community Practitioner, 81, 22-25. 426. Rifkin, S.B. (1996) Paradigms lost: toward a new understanding of community participation in health programmes. Acta tropica, 61, 79-92. 427. MacQueen, K.M., et al. (2001) "What is community? An evidence-based definition for participatory public health." American journal of public health, 91, 1929-1938. 428. Jewkes, R. and Murcott, A. (1996) Meanings of community. Social Science & Medicine, 43, 555-563. 429. Parker, E., Margolis, L.H., Eng, E. and Henríquez-Roldán, C. (2003) Assessing the capacity of health departments to engage in community-based participatory public health. American journal of public health, 93, 472-476. 430. Summers, A. and McKeown, K. (1996) Evolving a realistic strategy on public consultation. Public health, 110, 145-150. 431. Brown, I. (2001) Organizational values in general practice and public involvement: case studies in an urban district. Health & social care in the community, 9, 159-167. 432. Goodwin, V. and Happell, B. Psychiatric nurses' attitudes toward consumer and carer participation in care: part 2--barriers to participation. Policy, Politics, & Nursing Practice, 9, 249-256.

275 | P a g e

433. Kidd, S., Kenny, A. and Endacott, R. Consumer advocate and clinician perceptions of consumer participation in two rural mental health services. International Journal of Mental Health Nursing, 16, 214-222. 434. Rhodes, C.A. and Nyawata, I.D. (2011) Service user and carer involvement in student nurse selection: Key stakeholder perspectives. Nurse Education Today, 31, 439-443. 435. Happell, B., Moxham, L. and Platania-Phung, C. The impact of mental health nursing education on undergraduate nursing students' attitudes to consumer participation. Issues in Mental Health Nursing, 32, 108-113. 436. Speers, J. (2008) Service user involvement in the assessment of a practice competency in mental health nursing–Stakeholders’ views and recommendations. Nurse Education in Practice, 8, 112-119. 437. Owen, C. and Reay, R.E. (2004) Consumers as tutors–legitimate teachers? BMC medical education, 4, 14. 438. Monahan, A. and Stewart, D. (2002) The role of lay panelists on grant review panels. Chronic diseases in Canada, 24, 70-74. 439. Weiss, S.J. The influence of discourse on collaboration among nurses, physicians, and consumers. Research in nursing & health, 8, 49-59. 440. Gabbay, J., & le May, A. (2004). Evidence based guidelines or collectively constructed “mindlines?” Ethnographic study of knowledge management in primary care. BMJ, 329(7473), 1013. 441. Concato, J., Shah, N. and Horwitz, R.I. (2000) Randomized, controlled trials, observational studies, and the hierarchy of research designs. New England Journal of Medicine, 342, 1887- 1892. 442. Petticrew, M., Tugwell, P., Welch, V., Ueffing, E., Kristjansson, E., Armstrong, R., Doyle, J. and Waters, E. (2009) Better evidence about wicked issues in tackling health inequities. Journal of Public Health, 31, 453-456. 443. Agar, M.H. (1996) The professional stranger: An informal introduction to ethnography. 444. Cook, J.A., Jonikas, M.J.A. and Razzano, L. (1995) A randomized evaluation of consumer versus nonconsumer training of state mental health service providers. Community Mental Health Journal, 31, 229-238. 445. Kweit, M.G. and Kweit, R.W. (1981) Implementing citizen participation in a bureaucratic society: A contingency approach. Praeger New York. 446. Daykin, N., Evans, D., Petsoulas, C. and Sayers, A. (2007) Evaluating the impact of patient and public involvement initiatives on UK health services: a systematic review. Evidence & Policy: A Journal of Research, Debate and Practice, 3, 47-65. 447. Daykin, N., Orme, J., Evans, D., Salmon, D., McEachran, M. and Brain, S. (2008) The Impact of Participation in Performing Arts on Adolescent Health and Behaviour A Systematic Review of the Literature. Journal of Health Psychology, 13, 251-264. 448. Public Health Resource Unit (2006) Public Health Resource Unit: 10 Questions to Help You Make Sense of Qualitative Research. Critical Appraisal Skills Programme, Oxford, UK. 449. Smith, K.E., Bambra, C., Joyce, K.E., Perkins, N., Hunter, D.J. and Blenkinsopp, E.A. (2009). Faculty Public Health, Vol. 31, pp. 210-221. 450. Woelk, G.B. (1992) Cultural and structural influences in the creation of and participation in community health programmes. Social Science & Medicine, 35, 419-424. 451. Zakus, J.D.L. and Lysack, C.L. (1998) Revisiting community participation. Health policy and planning, 13, 1-12.

276 | P a g e

452. Jackson, N. and Waters, E. (2004) The challenges of systematically reviewing public health interventions. Journal of Public Health, 26, 303-307. 453. Nancholas, S. (1998) A logical framework. Health policy and planning, 13, 189-193. 454. Habicht, J.-P., Victora, C. and Vaughan, J.P. (1999) Evaluation designs for adequacy, plausibility and probability of public health programme performance and impact. International journal of epidemiology, 28, 10-18. 455. Parfitt, T. (2004) The ambiguity of participation: a qualified defence of participatory development. Third World Quarterly, 25, 537-555. 456. Haas, P.M. (1992) Knowledge, power, and international policy coordination. University of South Carolina Press. 457. Dwan, K., Altman, D.G., Arnaiz, J.A., Bloom, J., Chan, A.W., Cronin, E., Decullier, E., Easterbrook, P.J., Von Elm, E. and Gamble, C. (2008) Systematic review of the empirical evidence of study publication bias and outcome reporting bias. PLoS one, 3, e3081. 458. Katulanda, P., Ranasinghe, C., Rathnapala, A., Karunaratne, N., Sheriff, R., & Matthews, D. (2014). Prevalence, patterns and correlates of alcohol consumption and its' association with tobacco smoking among Sri Lankan adults: a cross-sectional study. BMC public health, 14(1), 612.

277 | P a g e

Annexures

Image 7. A community based psychosocial intervention targeting young children living in a displaced camp in Northern Sri Lanka

278 | P a g e

Annexure 1. Systematic review of literature: perceptions of health workers toward CP in health care If CP is to be translated from rhetoric into reality, then understanding the perspectives and attitudes of health professionals, who are the key catalysts for community engagement, need to be assessed. In undertaking the literature review, I discovered that the perspectives of the health care workers on CP and its utilization have not been systematically analysed. In this review I sought to synthesize such evidence.

Scope and Review Methods

The key question addressed by this review study was: what are the perceptions, attitudes and behaviours of health care workers toward community participatory approaches in health care? I also aimed at exploring the differences (if any) in attitudes, and behaviours towards CP between the various categories of health care workers, and between workers working at different levels of a health system (primary, secondary and tertiary care levels). The review did not aim to examine the literature in respect to the effectiveness of CP in health programming. Rather the purpose was focused on characterizing the attitudes, beliefs and behaviours of health care workers on the value and implementation practice of CP approaches.

Search strategy

The search strategy to address this question involved searching PubMed, OVID/EMBASE and Google Scholar data bases. The Medical Subject Headings (MeSH) search strings were grouped into two primary search domains: ‘community participation’, and ‘attitude of health personnel’. The word groups were combined with ‘AND’ between word groups and ‘OR’ within domains Box (a). Search limitations are also presented. Titles and abstracts of initially identified papers were assessed to identify those studies likely to meet the inclusion criteria. Once full texts were retrieved, the exclusion criteria were applied and a final list agreed of included studies for quality appraisal. The reference lists of included studies were also scrutinised for possible further relevant studies but none were identified.

Box (b). listed the inclusion and exclusion criteria for the study. Qualitative studies were also included in the study, and a narrative analytic approach was taken for the overall categorization of health care worker attitudes towards community participation. This involved an in-depth analysis of narrative findings in each paper. A qualitative assessment based on content analysis was performed across two dimensions. First, the overall measure of satisfaction of health professionals towards the consumer involvement in the health intervention/program were explored. Second, the overall measure of commitment towards sustaining or promoting CP intervention processors were examined. The overall attitudes towards community involvement in health were given the categorical variable: ‘Non-Supportive’, ‘Neutral’ or ‘Supportive’. For instance, if the narrative analysis of the majority of HCW surveyed/interviewed in a particular setting favoured

279 | P a g e

consumer or patient involvement, then a ‘Supportive’ label was presented. If the study indicated that the HCWs actively promoted CP practices/interventions within their setting a “committed” label was provided. In all studies, the conclusions and synthesis of the authors in defining the perception of the HCWs were analysed. Wherever this was not clear or ambiguous, the study was excluded from final selection. Health workers were also defined into the following categories. (A) Central level administrator, policy maker, planner; (B) Provincial and/or District head of health; (C) primary level health care worker.

Box (a). Search strategy

Mesh terms/translations community "consumer participation"[MeSH Terms] OR ("consumer"[All Fields] AND participation "participation"[All Fields]) OR "consumer participation"[All Fields] OR ("community"[All Fields] AND "participation"[All Fields]) OR "community participation"[All Fields] attitude of "attitude of health personnel"[MeSH Terms] OR ("attitude"[All Fields] AND health "health"[All Fields] AND "personnel"[All Fields]) OR "attitude of health personnel personnel"[All Fields] Databases: PubMed, Ovid(MEDLINE), Google Scholar Other search restrictions: Journal Articles, Abstracts available, Humans studies and English language articles, period 1980 to 2012.

Box (b). Inclusion and Exclusion criteria for studies

Population: All populations, communities and groups Interventions: Any empirical study of an intervention which has sought to involve, engage or work in partnership with a community or population. Setting: Any level of health care setting (primary care to tertiary care setting) Outcomes: Two broad measures of perceptions/attitudes of health care workers and health administrators involved in intervention/program were assessed; the overall satisfaction of the intervention involving CP approaches, and, the measure of the overall commitment to sustaining or promoting CP intervention processors. The measures need not be quantifiable through for instance Likert-scales. Qualitative paper did however need to articulate health worker perceptions on CP. If the study did not contain substantive data describing such perceptions, attitudes and behaviours of the intervention or program involving communities/health consumers/patients, the study was excluded from the analysis. Study design: Any study design including case studies and qualitative methods. Excluded were commentary and opinion pieces. Time limits: 1980–2012 Language: English language only (apply as limit) Journal quality: Peer-reviewed publications only.

280 | P a g e

Results

Figure (a). Literature search strategy

Table (a). Characteristics of selected studies

Study Characteristic n % Country/Region* Australia 14 36% Africa 3 8% Middle-East 1 3% UK 11 28% USA 6 15% Asia 4 10% Research Method Qualitative (including 3 case reports) 21 54% Quantitative 13 33% Mixed method 1 3% Experimental (RCT) 1 3% Ethnographic 1 3%

281 | P a g e

Health worker category PHC workers 16 41% Clinical staff (Mental health) 10 26% Nursing students 4 10% Mental health nurses, students, clinicians 3 8% Health Administrators 3 8% Clinical staff (Venereology) 1 3% Medical students 1 3% health promotion experts 1 3%

Institutional setting of HCW A - policy makers/senior administrator 4 9% level B - Tertiary and Secondary level institution 14 30% C - PHC level 23 49% D - trainees/students 6 13%

Disease/health intervention domain Generic 23 59% Mental Health 10 26% Cancer 2 5% Other** 4 10%

*Study by Rifkin (1983) was a multi-country study (Philippines, Hong Kong and Indonesia). **Occupational therapy (n=1), Substance abuse (n=1), sexually transmitted infection (n=1)

Discussion

Characteristics of selected studies

A total of 131 papers were identified after eligibility criteria was applied and duplicates removed from a total of 1798 retrieved from electronic and hand searches. Only 39 papers met screening criteria and were included in final analysis (Figure a). Of these papers identified, 19 (54%) utilized qualitative methods including 2 case-studies and 13(37%) quantitative survey methods. A single study used a mixed method approach. Of importance was a multi-site ethnographic study, and a randomized control study. In the later study, HCWs in the 'intervention arm' were assigned to receive training from health care consumers, and their experience/attitude with such an approach examined.

The majority of studies emerged from Australia (n=14, 38%), followed by the UK (27%) and North America (16%). There were only a small number of studies from Asia (11%), Middle East (3%) and from the African continent (8%). Therefore, studies from OCED countries dominated analysis with only 8 from Developing country perspectives.

282 | P a g e

A characteristic of most studies were that they were institution based, and only one attempt to capture a national sample of health administrators from each region (417). Most were also tailored to ascertain perspective in terms of a specific service type or condition e.g. mental health service.

Qualitative research methods were the most commonly used method to elucidate perspectives of HCWs, followed by quantitative methods (mainly surveys), case-reports, ethnographic studies and a randomized control study. Twenty-one studies utilized qualitative research methods (54%), quantitative data contributed to 33% (n=13) of the total.

Papers described the perceptions of a range of health care workers in diverse work settings within health system and in training/teaching institutions. The different effects of involving patients and the nature of the evidence to support these findings are summarized in Table c.

PHC workers were the most common category of health care worker to be surveyed/interviewed in determining the perceptions of health workforce towards CP (37%). When grading the category of worker by their position within the health system (i.e. administrator/policy maker grade, tertiary/secondary care institution, primary care institution to trainee/student grade), it was the primary care level, which elicited the most responses (49%), followed by those staff working in tertiary and secondary level institutions (31%). Policy makers/senior administrator level grades had limited involvement (7%).

The examination of other categories of workers revealed that clinical staff/trainees involved in mental health care contributed to 38% (n=13) of all HCWs. The importance of CP to mental health service delivery is a finding that was clearly highlighted. Supporting this finding is the fact that 26% of all studies included in the review specifically articulated mental health services as the health care issue where the platform of participation was being analysed. Cancer treatment and care formed the other disease category (6%) for which CP was explored in.

The number of health care workers participating showed a large variance. The largest number was from a multi-country study undertaken by Rifkin covering 113 HCWs(151). The smallest sample were form a study by Hooper (2007) involving the perspective of two HCWs (an occupational therapists and Aboriginal health worker) (418), and one study which described a narrative experience of a single health care administrator on his involvement in CP within a primary care setting (1981) (419).

CP in health care is increasingly becoming an expectation in the mental health area. The rationale for such studies have cited the need for ‘greater consumer involvement’ in the education of mental health professionals, and has explored CP since it has been ‘presented as a strategy to encourage the development of a greater acceptance of consumer involvement by health professionals’. For instance, consumer participation in all

283 | P a g e

aspects of mental health service delivery, including the education of mental health professionals, is now a policy expectation in Australia (420).

Papers often described changes to services that were attributed to involving patients, including attempts to make services more accessible and producing information leaflets for patients. Changes in the attitudes of organisations to involving patients and positive responses from patients who took part in initiatives were also reported.

2.9.4 Quantitative analysis

An in-depth review of each paper were undertaken to explore how authors had described perceptions and attitudes of HCWs towards CP. Perception on whether health professionals supported the role CP played in their health intervention/setting were summarized into a range of responses from “Non-Supportive”, “Neutral” or “Supportive”, and the commitment to the pursuit of CP within their service area/setting categorized as “committed” or “non-committed”. For instance, if the majority of HCW surveyed or interviewed in a particular setting favoured patient involvement in design of health care program, then a ‘Supportive’ label is registered. If the same study indicated that the majority of HCWs were actively promoting CP practices within their work setting, then a “committed” label was also provided.

Findings from review indicated that the majority of HCWs (46%) were supportive with the idea of CP, and were committed to its implementation in routine practice (Table b). However, there were a small proportion of studies (18%), where the majority of workers supported the principles of CP but showed unwillingness to pursue its goals in practice. A number of studies also iterated HCW expressions of non-supportiveness and non- commitment to CP in the health settings/practices (21%).

Table b. Health care worker attitudes and commitment to community participation n % Supportive, Committed 17 44% Supportive, non-committal 7 18% Mixed reaction, non-committal 4 10% Mixed reaction, Committed 3 8% Non-Supportive, non-committal 8 21%

When studies highlighting a positive and committed response from HCWs were analysed (n=16), no clear patterns emerged which characterized any specific level of worker. Whilst some indicated that nurses appeared to have a greater receptiveness in creating a work culture that enabled CP (420-422), there were other studies which showed mixed or opposite effects involving nurses (203,205,206,213).

Table c. Summary of studies showing health care worker attitudes and commitment towards CP

284 | P a g e

HCW attitudes Health worker category (A,B,C,D) and Study Study methods and setting type* commitment to CP*

Rhodes Quantitative: questionnaire; Supportive, D: Nursing students , in a tertiary institution in UK (2011) Qualitative: FGDs committed

Happell D: Nursing students (N = 68) in an Australian Supportive, Quantitative: survey (2011) University committed

Owen Quantitative: survey (pre Supportive, D: medical students, Australian University (2004) and post) committed

Goodwin B: 30 psychiatric nurses. Hospital settings, Victoria, Supportive, Qualitative: FGD (2008) Australia. committed

Non- McCann B: 47 clinicians in the psychiatric units of a large Quantitative: cross-sectional Supportive, (2008) Australian hospital, Australia survey design non-committal

C, D: mental health nurses, nursing students, service Speers neutral, users, mental health lecturers, and mentors in a UK Qualitative: KI interviews (2008) committed PHC trust.

Non- Hogg (2008) C: PHC workers at PHC trust, NHS, Scotland. Qualitative: FGDs Supportive, non-committal

Bryant C: 64 providers and 179 consumers of drug Quantitative: survey; Supportive non- (2008) treatment services, Australia. Qualitative: KI interviews committal

B: survey of heads in social services departments Non- Bowl (1996) responsible for mental health services; including Qualitative: KI interviews Supportive, service users, PHC trust, NHS, U.K. non-committal

B: mental health professionals from two adult in- McCann neutral, patient psychiatric units situated in a large Quantitative: survey (2008) committed Australian public general hospital

B,C: 10 Mental health professionals in two mental Supportive, Kidd (2007) Qualitative; KI interviews health services at PHC level, Australia. non-committal

Nathan Quantitative; self- Supportive, C: HCWs in one Australian hospital in urban setting. (2006) administered survey committed

Happell D: postgraduate psychiatric/mental health nursing Supportive, Quantitative: survey (2002) students(n = 25), Australia non-committal

Quantitative study: pre- validated instrument (Mental Health Consumer Happell Supportive, D: Overall, the 150 nursing students, Australia Participation Questionnaire) (2010) committal to explore attitude levels and perceptions on psychometric grounds

Mixed reaction Qualitative: FGDs and KI Brown C: HCWs in 4 general practice/PHC organisations in to satisfaction interviews and case study (2001) urban district in England and methods commitment

C: 23 general practitioners in practice settings in Brown Supportive, inner city Sheffield, U.K.; 20 practice nurses; Qualitative: KI interviews (1994) committal? 11 health visitors and 7 district nurses.

285 | P a g e

Randomized control study (RCT): HCWs were randomly assigned to one of two conditions; one in which Highly A,B,C: Fifty-seven state mental health professionals Cook (1995) they received the second Supportive, in USA. day of training from a committal consumer and the other involving training by a nonconsumer.

Bastani B,C: 96 health workers (mostly nurses), rural health Supportive, Quantitative: Survey (1994) settings, USA. committal

Non- Freyens C: 29 health workers in PHC institutions in Rwanda Quantitative: Survey Supportive, (1993) non-committal

Non- Uzochukwu C: 20 health workers in PHC health centres in Quantitative: Survey Supportive, (2004) Nigeria in the Bamako Initiative programme non-committal

Jayanna C: Physicians working in sexually transmitted disease Qualitative study: FGDs Supportive, (2010) clinics at PHC level, in Karnataka, South India. and KI interviews non-committal

C: Occupational therapists and Aboriginal health- Qualitative study: In-depth, Hooper Supportive, workers in rural and remote communities of North semi structured telephone (2007) Committed Queensland, Australia. interviews.

B,C: The sample population comprised of 12 health Mixed reaction, Anokbonggo workers from the two hospitals, 11 district health difficult to Qualitative study: FGDs (2004) administrators, and 67 Local Council Leaders in remain Uganda committal

B,C: Community participants (N=19) and professional stakeholders (N=13) of a community Bandesha Qualitative: KI interviews DisSupportive, health project for a South Asian population in the (2005) and FGD non-committal Greater Manchester Primary Health Care Trust in England.

A: 17 experienced health practitioners in community- Non- Brandstetter Qualitative: Semi-structured based health promotion in New South Wales, Supportive, (2012) interviews Australia. non-committal

A,B,C: 31 HCW from a range of service types in the largest metropolitan health care service provider in the state of Victoria, Australia. Health professionals Qualitative: Semi-structured Supportive, Lake (2009) were medical specialists, general practitioners, interviews Committed nurses and clinical psychologists in a range of primary care settings, hospital outpatient care and community health services.

C: HCWs, non-health related organisation workers, Qualitative: 8 Focus groups Supportive, Litva (2002) health consumers, UK, NHS. and in-depth KI interviews Committed

B: 17 HCW staff were surveyed on their opinions Quantitative: survey; regarding their participation in projects that Qualitative: KI Interviews Supportive, Carr (2001) involved consumers as part of the team at the Royal using structured Committed Women’s Hospital in Brisbane, Australia. questionnaires

Quantitative: self- administered questionnaire C: 47 HCW surveyed in 2001 and 43 in 2002. Nathan as baseline measure and 12 Supportive, Study site is located on the edge of Sydney, (2006) months after community Committed Australia in a semi-rural area with a population representatives had been appointed.

Multi-method, multi-site Nathan C: 10 Community Representatives, 19 HCWs and 7 Supportive, ethnographic study: (2013) ‘Community participation Coordinators” employed in Committed observational study of an area health service in an urban setting in interactions between HCWs 286 | P a g e

Australia. and community members; in- depth interviews

Haigh B,C: 30 cancer patients and health professionals Supportive, Qualitative: KI interviews (2008) from two teaching hospitals, UK Committed

Qualitative: FGDs and KI’s with HCWs and patients affected by lung cancer, prior to and following an Forbat, L Supportive, B: lung cancer patients and health professionals, UK intervention (where lung (2009) Committed cancer teams were supported to engage with patients and family members)

B, C: physicians (N = 120) and group members (N = Fridinger F. Supportive, 73) located in the Dallas Worth Metropolitan area, Qualitative: KI interviews (1992) non-committal USA.

Mixed method approach of Weiss C: 72 HCWs (nurses, physicians), and consumers in Qualitative (observational Supportive, (1985) USA. study ) and Quantitative non-committal (survey) methods were used.

C: medical professionals (18), community development (15) workers and the health volunteers Supportive, Rifkin (1983) Quantitative: survey (80) in 3 countries: Hong-Kong, Philippines, non-committal Indonesia.

B: National Institute of Mental Health (NIMH)’s Simpson Qualitative: Case-study Mixed reaction, Community Mental Health Center (CMHC) Program, (1981) analysis non-committal USA.

Non- Steckler A: Health Administrators in Health Agency in Qualitative: Case-study Supportive, (1978) southeastern USA analysis, KI interviews non-committal

Al-Mazroal A: sample of health administrators (n= 29) in Saudi Mixed reaction, Qualitative: KI (1991) Arabian health system. non-committal

Summers C: Mental health unit and Maternal Health Service Qualitative: Case-study Mixed reaction, (1996) unit, at PHC level in UK. analysis, KI interviews committal

B: Clinical staff in a major metropolitan hospital, Supportive, Carr (2001) Quantitative: questionnaire Australia Committed

Mixed method: Quantitative Monahan A: Administrators at National Cancer Institute of Supportive, survey ; Qualitative KI (2003) Canada (n=129). Committed interviews.

Qualitative analysis

There was great diversity in the studies that were captured in the review (Table c). A number of studies set out to explicitly explore the perceptions of HCWs community, consumer or lay participation in specific organisational contexts and processes (417) (151,313,423-425); whilst others sought set to explore CP as part of an intervention process. Some involved assessing CP in HCWs in a disease specific domain, whilst others examined CP amongst trainees. Other studies explored how HCW perceptions on CP practice changed over time. Examples of studies across each of these study frames are presented below:

287 | P a g e

Studies that sought to explicitly survey perceptions of HCWs

A study by Al-Mazroal (1991) who used structured interview methods with 29 health administrators and health care decision-makers across the kingdom of Saudi Arabia showed that most respondents adequately recognised and interpreted the importance of CP, however less than half indicated the practice of CP in planning and evaluation and implementation of health services in their respective regions (417). Such studies formed expels of direct assessment of attitudes. Other studies explored the HCW perspectives on CP after or during a process of an intervention that sought to implicitly or explicitly involve CP practice. Similarly, another study by Freyens (1993) surveying 29 HCWs in PHC institutions in Rwanda showed that 83% were reluctant to accept the involvement of public in the promotion and implementation of PHC programmes, and responded HCWs should not take the initiative in health promotion activities with communities. There was however confusion about the notion of community, with the health workers tending to underestimate people's potential for action and reaction, and to insist upon the need for a hierarchical structure. Multiple definitions were provided for understanding of community participation where authors concluded there was a tendency to 'underestimate' CP practices. A common challenge plaguing previous reviews of literature, and indeed any effort to explore the phenomena of participation is in how each study defined what was meant by ‘Community participation’ (73,426), and what was meant by ‘community’ (427,428) (426) (429).

Examples of studies that sought to obtain perceptions of HCWs through case-study approach

Three studies employed an intensive case-study approach to document the experiences of HCWs a study by Summers (1996) explored the utility of CP by the health system in affecting change (430). The study used findings of an evaluation and case studies of two health projects carried out in one on adult mental health unit and maternity service in a UK NHS setting concluded that public consultation will not necessarily make health system decision making easier, more cost-effective or even more acceptable to the whole population. The study found community capacity, staff-skills, the dominant professional and overall organisational ethos and culture all contribute to the effectiveness of organisational-community partnerships. The choice of strategy to involve public and communities will be driven as much by implicit value judgments of HCWs and administrators as by explicit technical considerations (430). An in-depth case-study analysis by Simpson (1981) explored the extensive narrative of an executive director of mental health clinics at primary care level in the USA (167). The study found these clinics did not have the capacity to ‘deal successfully with the complex issues of citizen participation in governance, service delivery and evaluation’. A primary obstacle cited by health administrators and HCW were the ‘continuous financial crises of the state community mental health network (CMHC). In another intensive case-study analysis, Brown and colleagues (2001) documented HCW perceptions to CP in 4 general practice/PHC organisations in an urban district in England (431). They provide evidence of how organisational values can differ markedly in general practice in relation to ideas

288 | P a g e

of public involvement, with consequences for the quantity and quality of activities for involving local people and service users. The differences manifest themselves in the beliefs and attitudes of service providers about the purpose of the organisation and the types of relationships that are appropriate with service users and local people. Power is a critical issue in this setting with evidence in the study of the dominance of the medical practitioners in establishing organisational values and the nature of public involvement activities.

Studies that sought to explore perceptions of HCWs within specialist domains

A cross-sectional survey of 47 clinicians at a psychiatric unit in a large Australian hospital showed that attitudes about patient engagement in care provision were influenced by gender and length of time the staff worked in the psychiatric unit (432). Less experienced staff showed greater support than more experienced staff on mental health consumer involvement in treatment-related matters and consumer consultants in units. Female staff were also more likely than their male counterparts to support patient engagement. New staff members were more likely to register agreement-to-uncertainty regarding consumer involvement in treatment-related issues, whereas established staff members were more likely to record uncertainty about this issue.

A study by Kidd (2007) of 10 mental health professionals working in two mental health services at primary care level in Australia showed that despite philosophical commitment to the idea of patient participation, there has is little translation into practice (433). Kidd’s study showed that all health workers uniformly supported patient participation as being important for the provision of mental health services. However these were not enacted in actual practices at their respective institutional setting. Authors suggested that a systemic change to the way CP is incorporated into mental health services is required. The incorporation of consumers as stakeholders within mental health services requires commitment to the idea, the establishment of clear purposes and roles, and commensurate funding to enable consumers to be stakeholders with similar resources to other stakeholder groups.

Studies that sought to explore perceptions of HCWs within training contexts

Studies with trainee HCWs reveal that the principles of CP were largely viewed positively by both nursing students (434-436) and medical students (437), although some advocated for limited involvement of health consumers and community carers in health care services. Similarly, a study investigating the attitudes of scientists and lay panel members in the National Institute of Canada grant review panels found that most scientists viewed the role of lay members on the panels as legitimate and valuable in decision-making, at statistically significant levels (438).

Studies that sought to examine changes in HCW perceptions on CP over time

The way in which attitudes of health professionals change over time as a direct result of the experience of working alongside community members were explored in three studies

289 | P a g e

(423,439) (437). Nathan explored the attitudes of 47 hospital staff in a large hospital setting that participated in health service committees towards their participation (423). Their attitudes towards CP changed positively over time. The paper finds that significantly, more staff at the follow-up survey reported that they became clear about the role of community representatives and how to work with them on committees. Although there was no significant increase in the percentage of staff who felt that CP influenced real decision making, there were strong trends in a positive direction over time. The paper implies that the development of ongoing and constructive relationships between health services and communities clearly takes time and, at least in part, results from direct experience working alongside community members.

A study by Weiss (1985) on 72 HCWs (nurses and physicians) in USA examined the extent of collaboration and consultation amongst health consumers/patients and clinical staff in delivery of health care interventions (439). The intervention involved small group discussions with clinicians and health consumers/patients in a hospital setting undertaking group discussions on topics on role and participation in health, how each actor can influence health outcomes. These changes were measured by a standardized instrument (the Health Locus of Control Scale) using survey methods. The study found there was no positive change in value and support for consumer involvement in health by clinicians. This was measured to statistical significant through a pre and post-test score. Consumers and physicians demonstrated significant pre-test differences in the degree to which responsibilities should be shared in health care. These differences existed in these beliefs after 2 years of discussion. The post-test alignment between physicians and consumers seemed a function of the consumers’ decline in their orientation toward shared responsibility and a greater leaning toward increased responsibility for physicians. Interaction between HCWs and community actually served to enhanced traditional, status quo beliefs regarding the authority and power of the physician rather than fostering collaborative values. A study by Owen (437) on medical students at an Australian University over a 4 year period found that medical students showed positive attitudes towards mental health consumers as tutors and CP in their training/education programs. There was also general trend towards improved attitude of students on consumer involvement across all measures. The study concluded that consumer tutors along with professional tutors have a place in the education of medical students, are an ‘untapped resource and deliver largely positive outcomes for students and themselves’. Both Ethnographic studies(440) and randomized controlled studies (441) provide powerful research methods in exploring phenomena of interest in public health research (442). The finding of two such studies which emerged from the literature review are presented here.

Ethnography is a social science research method that relies heavily on up-close, personal experience and possible participation by researcher for an intensive period of time (443). Ethnographic methods can give shape to new constructs or paradigms, and new variables, for further empirical testing in the field (440). An ethnographic study by Nathan (2006) over a two years period revealed significant changes in the attitudes of organisations to involving consumers/patients on health committees over a time in a large hospital setting (423). The results are encouraging as they highlight the potential for

290 | P a g e

HCWs to develop constructive relationships with community members on decision-making committees over time. However, the study also revealed a lack of significant change in HCW attitudes around the influence and independence of community representatives, which signalled a need to go beyond changing the organisational structures and processes in health services to facilitate community participation”. A related study on the same study setting revealed the importance of having adequate resources and specific technical support to ensure an enriching and meaningful CP process. Findings called attention to the need for health services to employ a CP ‘facilitator’ who can support, engage, navigate and advocate for community representative’s participation and influence in health service policy and practice(423).

The study by Cook and colleagues (444), involved fifty-seven state mental health professionals who participated in a two-day training designed to acquaint trainees with the attitudes and knowledge necessary for delivering assertive case management services to improve clinical outcomes. Participants were randomly assigned to one of two conditions: one in which they received the second day of training from a health care consumer and the other involving training by a non-consumer. Analyses revealed that post-training attitudes were significantly more positive for those participants trained by the consumer. The study formed a landmark in its empirical investigations on the use of lay-persons to train mental health professionals. Subjective evaluations undertaken through qualitative methods also reflected positive reactions to the use of consumers as trainers. Both studies highlighted the importance and need to clearly define the role of consumer and community members in any CP venture. Exploring within a collaborative space, the methods of community engagement, how they can meaningfully contribute to decision- making and with dedicated resource support were seen as essential steps to ensure sustainable commitment to CP.

2.9.6 Factors influencing community participation in healthcare identified by Health care workers

This in-depth review of literature aimed at exploring the various obstacles to CP described in the narratives of HCWs. During the content review of each selected paper, factors which emerged which were important to enabling CP or inhibiting its progress were identified, and then listed. These were categorized to broader themes that were constructed to be reflective of the findings. These larger domains or environments (e.g. Policy and regulatory environment) are terms that are linked to the multi-domain model of the Ottawa charter for health promotion and the social determinants of health framework.

The summarized results are presented in Table d. Systematic barriers, deficits in technical skills and competencies, cultural barriers and resource barriers were identified influencing HCW perceptions on CP. If the goals of CP is to be translated from rhetoric into meaningful reality within health systems or health care delivery settings, then understanding such determinants and constraints to forging effective and meaningful partnerships between statutory organisations and communities are crucial.

291 | P a g e

Through and analysis of the literature review, I was able to capture a number of factors that may influencing the culture and practice of CP within health care settings. These were further classified under the broad domains of: administrative environment of the health unit, facility or institution; the attitudinal environment of health workers; the technical competency and knowledge environment of health worker in relation to CP approach; resource environment such as time, material and other monetary resources conducive for CP practice; and finally a policy and regulatory environment. These domains may independently or synergistically work to influence the intervention and practice environment in health care settings. I drew upon these findings to formulate the overall theoretical framework for my study.

Table d. Factors influencing community participation Author Findings Factors (inhibitory and (year) enabling) classified under broader theme Bowl  Confusion about the meaning and purpose of  Administrative (CP (1996) CP in health approaches not  lack of consensus on how health stipulated by consumers/service users can best be management, part of represented on committees job duty; Confusion  Unwillingness to listen to community views over role in CP)  Limited resources to enable participation from  Attitudinal (value management placed on CP poor,  CP does not directly relate to their spheres of not wanting responsibility and thus forms a push factor engagement from against its implementation non-health stakeholders)  Technical competency (methods on effective CP practice unknown)  Resources (time and resource constraints not allowing for CP practices) Freyens  Health workers tending to underestimate  Attitudinal (little (1993) people's potential for action and reaction, and value placed on CP) to insist upon the need for a hierarchical  Technical structure. competency  Confusion about the notion of community and (methods on effective multiple definitions were provided for CP practice unknown) understanding of community participation.  Resources (time and  Resource limitations where even basic services resource constraints are unable to be provided not allowing for CP  Lack of financial resources practices)  Non-conducive management practices  Administrative (CP  Ignorance about the value of participation approaches not  Taboos, customs, and traditions producing stipulated by resistance to change management, part of job duty; Confusion  Confusion by community on their roles in CP over role in CP)  Lack of education by community on their roles in CP Hogg  An increasingly medicalised view of health  Attitudinal (value

292 | P a g e

(2008) promotion pushes out CP in routine practice placed on CP poor in  The perceived lack of leadership and interest medicalised from some communities prevents CP from being approach to service pursued delivery)  Attitudinal (perception that community does not want CP) McCann  policies that promote participation have little  Policy (2008) impact to promote culture of CP implementation  Younger staff are more willing to engage in (Policy rhetoric not new methods than older ones translating to practice) Bryant  believe in the inadequacy of consumers' skills,  Attitudinal (2008)  perceive consumers' lack of interest (perception that  Do not value consumer involvement in health community does not decision-making. possess capacity to engage)  Attitudinal (value placed on CP poor in medicalised approach to service delivery) Kidd  A systemic change to the way consumer  Administrative (2007) participation is funded and incorporated into  Resources mental health services is required.  the establishment of clear purposes, policies and roles, and commensurate funding to enable consumers to be stakeholders with similar resources to other stakeholder groups Happell  ambivalence regarding the value and necessity  Attitudinal (value (2002) of CP placed on CP limited)

Brown  Dominance of the medical practitioners in  Attitudinal (value (2001) establishing organisational values during public placed on CP poor in involvement activities. medicalised  Local political tensions in primary care approach to service influences CP delivery) Brandstetter  The importance of having community member’s  Attitudinal (value (2012) involvement is important but should only be placed on CP limited included in the evaluation process. to only evaluation of programs) Nathan  The need for health services to employ a  Resources (2013) facilitator who can support, engage, navigate  Technical and advocate for community representative’s competency participation and influence in health service policy and practice. In the absence of a CP facilitator, it becomes difficult to build skills and confidence of HCWs for CP and engage them in agendas for action Haigh  lack of insight on appropriate methods of  Technical (2008) engaging communities competency  Health professionals prefer consultative, rather (methods on effective than decision-making role for consumes. CP practice unknown)  Variable patient interest, health professional  Attitudinal attitudes 293 | P a g e

Fridinger F.  Limited time available due to busy schedule for  Resources (time (1992) communication between health care constraints not professionals to engage in CP. allowing for CP practices)

Rifkin  HCWs in Asia have a ‘medicalised’ view of  Attitudinal (value (1983) health and want responsibility to remain in the placed on CP poor in hands of the medical professional. medicalised approach to service delivery) Simpson  The primary obstacle is funding CP strategies  Administrative and (1981) and lack of political commitment to policy frameworks decentralization (lack of decentralized approach of health governance non- conducive for CP)

Review Limitations

Caution should always be made to any analysis which seeks to ascribe categorical labels for HCW attitudes and perceptions towards CP. Whilst a strict process of selection and analysis accompanied the review, I was limited by the narratives within these peer- reviewed papers. Some papers elaborated an in-depth account of HCW perceptions, while others provided only superficial constructs of the narrative analysis. Whilst every effort was made to capture a larger range of papers, I excluded those studies which fit the inclusion criteria yet failed to provide an in-depth narrative on worker perceptions.

As discussed in first section of this literature review, the definitions of CP in health is highly contested, and have divergence of meanings. Unfortunately most authors did not seek to define the meaning of CP as perceived by the workers. Few provided information on how HCWs defined CP practice.

Conclusion

The review is the first to attempt to ascertain perspectives of HCWs toward CP in various health care settings and institutions. The main finding of this review is that while most HCWs supported the idea of CP in health care, there was mixed evidence on the commitment needed for its meaningful implementation in health programs and routine practices. A small number of studies iterated dissatisfaction and non-commitment to CP by workers.

It is evident from this review that HCW staff attitudes play an important role in determining the acceptability of CP within an organisation and its likely influence on decision-making. Studies of attitudes of health workers are useful in increasing our understanding of how their current attitudes may support or limit the input of consumer or community representatives in decision-making.

Scholars within the field of governance and political science suggests that in government bureaucracies, the decision to involve communities as part of program interventions 294 | P a g e

reflects ‘administrative responsiveness’ (197) (445). Yang (2007) also iterates that prevailing attitudes and cultures of staff associated with organisations are crucial determinants of the benefits derived from citizen participation (197). As highlighted by studies in this review supportive HCW attitudes towards CP have been identified as a key contributor to effective partnerships between communities and health systems in decision-making.

Further studies are needed to explore perceptions of CP within the different categories of workers within a health system to identify if there are variations according to level of health care for which the HCW is located in (primary, secondary, tertiary care services), role in health care (educators, practitioners/clinicians, prevention/promotion sector) and professional category of worker (doctors, nurses, allied health professionals).

295 | P a g e

Annexure 2. Review of literature examining CP in Sri Lanka’s health system

Introduction and Rationale

A comprehensive review of literature was undertaken to document evidence of CP practice in health care programming and decision making in Sri Lanka. The review was focused on mapping of CP practice rather than examining the effectiveness or outcomes of the use of CP to achieve health program objectives.

Review method

The key research question addressed by the in-depth review was: What is known about the nature, form and extent of community participation in health care programming in Sri Lanka, and of the impact such approaches on health outcomes?

Search strategy

The search strategy to address this question involved searching 9 databases (Box a.), with five word groups (Box b.), for the period 1974 to October 2012, limited the English language. Inclusion and exclusion criteria were identified (Box c.). First, the Medical Subject Headings (MeSH) search strings were grouped into five primary search domains: typology of community, intervention, organisational setting, health outcome and country. These were combined with ‘AND’ between the word groups, and ‘OR’ within word groups. Published journals articles within the archives of Sri Lankan journals online a dedicated portal for national scientific publications were also searched. Finally, hand searches from two of the largest institutional libraries on medicine, health and social science in Sri Lanka were conducted: the Post Graduate Institute of Medicine (PGIM) and the University of Colombo library, to examine the repository of peer-reviewed publications, masters and doctoral theses.

Data analysis

All references were downloaded into Endnote database for processing. Duplicates were removed and papers which described different outcomes of the same study were considered as a singular study. Titles and abstracts of initially identified references were assessed to identify those studies likely to meet the inclusion criteria (Box c). Once full texts were retrieved, these were processed using a data extraction form developed for previous systematic reviews (446) (447) (166). Inclusion and exclusion criteria were again applied and a final list agreed of included studies for quality appraisal and content analysis.

Page | 296

When reading articles for inclusion in the review, the nature of the evidence contained in the paper was analysed. This included study type, nature of data (quantitative/qualitative), the context of involvement (geographical setting, nature of the service, and patient/consumer group), the research method used, and whether intervention outcomes were reported. A critical appraisal of both qualitative and quantitative data as described from the Critical Appraisal Skills Programme criteria (CASP) (448) and other sources (449) (166) were utilized (Box d). Critical appraisal is mandatory to ensure methodological rigour, credibility and relevance in the screening of papers selected for systematic review. The following nodes of information were collected:  Typology of Study design/Nature of evidence. Categorizes papers according to Quantitative research (observational or experimental studies), Qualitative research or Case studies.  Typology of community/group. Describes the definition and organisation of ‘community’ within the study paper. Details of size, nature and geographical location were also recorded.  Disease type. Classifies target diseases into two broad epidemiological categories; Chronic Non-Communicable Diseases (NCD) and Communicable Diseases. Wherever possible, specific disease conditions are also reported.  Health system involvement. Characterizes the role and involvement of the formal health system within the intervention, and their interactions within the ‘intervention space’.  Assessing form of community participation/community participatory approaches in health care. Difficulties with designing and implementing community participation programmes have in part been attributed to a lack of consensus on what constitutes 'community' and 'participation' (450). Community participation in health is therefore complex and challenging to assess (451). Participation may be seen as an end in itself, as well as a means towards an end. Conceptualized as an end in itself, participation helps people to acquire the skills, knowledge and experience to take more responsibility over their own development, ultimately being empowered to transform their lives and environment. Conceptualized as a mean to an end, participation is a process whereby local people cooperate and collaborate in an externally introduced project. Conducting systematic reviews of complex health promotion interventions is methodologically challenging (452), yet vital to persuade policy makers, planners and/or health care workers on the effectiveness of CP in practice (166,170) (449).

Reflections on the level and extent of community participation

To analyse the dynamics of CP in health programs described in papers that emerged from the systematic review I utilized the earlier described Community Participatory Score Card. The CP interventions described in each selected paper were also analysed according to its use in the ‘project cycle phase’. The ‘project cycle’ is based on the classic model of project development which describes the stages of conceptualization, design, development, implementation, evaluation and monitoring of any given health intervention/program

Page | 297

(453,454). Assessing the ‘phase’ at which participatory action manifested in specific health interventions may provide a better understanding of its utility in the health system. A typology of steps (Step 1 to Step 6) were assigned to fit the dynamics of the project cycle (Figure a).

Box a. List of databases searched

A. Electronic databases  Medline. Produced by the U.S. National Library of Medicine, is widely recognised as the premier source for bibliographic and abstract coverage of biomedical literature from 1966. Medline was accessed via PubMed (a free database accessing primarily the MEDLINE database and extended papers on life sciences and biomedical topics).  EMBASE. EMBASE, the Excerpt Medical database, produced by Elsevier Science, is a major biomedical and pharmaceutical database indexing over 3,500 international journals.  Global Health. Brings together resources of 2 international databases: Public Health & Tropical Medicine (PHTM) database, previously produced by the Bureau of Hygiene & Tropical Diseases (BHTD), and the human health and diseases information extracted from CAB ABSTRACTS.  Scopus. Scopus covers the broadest available coverage of scientific, technical, medical and social science literature including: 15,000 peer-reviewed journals from more than 4,000 international publishers.  Sri Lanka Journals Online. Medical, Social Science and allied health journals.  Google Scholar. Search of scholarly literature across many disciplines and sources, including theses, books, abstracts and articles. B. Institutional Databases (hand searched)  The Post Graduate Institute of Medicine (PGIM) collection of Post-graduate thesis (PhD, MD, DSc, MSc)  University of Colombo (Ouch) Repository of Research Publications.

Box b. Medical Subject Headings (MeSH) subject headings and free text searches were performed combining 5 search string domains: population, intervention, organisational setting, outcome and country.

Word group 1: Word group 2: Word group 3: Word group 4: Word group Community organisational intervention health 5: Typology setting outcome Country Community Community Health Health Campaigns Quality of Sri Lanka Planning Health Care Neighbourhood Community Health Promotion of Health Outcome Ceylon Systems Assessment (Health Care) Resident Delivery of Health Promotional Items Process Care Assessment (Health Care) Residential Health Care Wellness Programs Health Care Selection Systems Review Consumer Healthcare Systems Community Health Professional Participation Education Review

Page | 298

Organisations Consumer Delivery of Health Education, Community Program Involvement Care, Integrated Health Evaluation Community Integrated Health Education, Health Quality Participation Care Systems Assurance, Health Care Public Health Information Health Education, Quality Participation Systems Community Improvement Patient Health Systems Health Planning Quality Participation Agencies Indicators, Health Care Social Health Systems Health Care Utilization Participation Plans Financing Review Administration Community Healthcare Common Health Facility Health Care Action Procedure Coding Administration Quality, System Access, and Evaluation Patients Healthcare Health Services Health Care Administration Evaluation Mechanisms Clients Healthcare Delivery Health Resources Organisational Administration Case Studies Inpatients Community Health Public Health Outcome and Services Administration Process Assessment (Health Care) Outpatients Comprehensive Root Cause Health Care Analysis Population Primary health care Treatment Outcome Consumer Primary Care Advocacy Public Community Health Advocacy Workers Village Health Workers

Box c. Inclusion and exclusion criteria

Journal quality: Peer-reviewed publications and published Doctoral and Master’s thesis that have undergone peer-review at a University or Post-Graduate Institute.

Population: All populations, communities and groups in Sri Lanka.

Studies/Interventions:  Any study of an intervention which has sought to involve, engage or work in partnership with a community group or population setting in Sri Lanka, and/or;  Any paper, that described community participatory approaches and described the involvement of community, consumers, patients in health services delivery or policy development were included.  Excluded papers were those which described CP in community based research projects, or Page | 299

studies which only described or measured ‘satisfaction’ of participatory approaches by communities/consumers/patients without describing how CP practices developed/impacted health care services or policies.  Studies analysing participation dynamics of the community without reference to a particular intervention were also included

Study design: Any study design (quantitative, qualitative, experimental, quasi-experimental), including case studies. Excluded are commentary and opinion pieces. Outcomes: Any health or social outcome, including measures of ‘empowerment’ and/or increased involvement in health-related decision making were considered. Setting: Sri Lanka. Time limits and Language: Searches were restricted to papers published in English between January 1966 and September 2012

Box d. Critical appraisal criteria (adapted from Public Health Resource Unit (2006)(448), Smith (2009) (449) and Evans (2010) (166)).

Definitions 1. Was the use of term ‘community participation’ defined by authors? 2. Was the use of term ‘community’ defined by authors?

Qualitative studies 1. Is there a clear statement of the research question and aims? 2. Was the methodology appropriate for addressing the stated aims of the study? 3. Was the recruitment strategy appropriate and was an adequate sample obtained to support the claims being made? 4. Were the data collected in a way that addressed the research issue? 5. Are the methods of data analysis appropriate to the subject matter? 6. Is the description of the findings provided in enough detail and depth to allow interpretation of the meanings and context of what is being studied? (Are data presented to support interpretations, etc.?) 7. Are the conclusions/theoretical developments justified by the results? 8. Have the limitations of the study and their impact on the findings been considered? 9. Is the study reflexive? (Do authors consider the relationship between research and participants adequately and are ethical issues considered?) 10. Do researchers discuss whether or how the findings can be transferred to other contexts or consider other ways in which the results may be used?

Quantitative studies 1. Is the study prospective? 2. Is there a representative sample? 3. Is there an appropriate control group? 4. Is the baseline response greater than 60%? 5. Is the follow-up greater than 80% in a cohort study or greater than 60% in a repeat cross- sectional study? 6. Have the authors adjusted for non-response and drop-out? 7. Are the author’s conclusions substantiated by the data presented? 8. Is there adjustment for confounders? 9. Were the entire intervention group exposed to the intervention? 10. Was there any contamination between the intervention and control groups? 11. Were appropriate statistical tests used? Overall grading of the quality of responses in critical appraisal criteria Page | 300

Overall grading of the quality of responses in critical appraisal criteria Excellent Good Average Poor Very poor Non-existent +++++ ++++ +++ ++ + -

Figure a. Identifying points of participatory engagement of any given health intervention within steps of the project cycle. The basic steps involved in the classic health intervention project cycle involves: conceptualization, design, implementation, monitoring evaluation and resolution of programs.

To summarize, the CP dynamics and participatory approaches described in extracted papers were analysed qualitatively across two dimensions: by the ‘degree of community participation’ (from a scale between 0 to 4), and by the ‘project cycle phase’ (Step 1 to Step 6). Interpreting the inherent complexity of the rubric of participation though the narrative analysis described in a scientific paper may be limiting. However the values ascribed through use of such scales described are not intended to be an empirical measurement of ‘success’ or outcomes, rather a reflection of the extent of CP within each health system intervention.

Page | 301

Results Figure b. Selection process from database search to final analysis.

Results from 9 electronic databases, including hand searches from Ministry of Health, Post- graduate Institute of Medicine and University databases (n= 785)

Application of eligibility criteria: 73 titles/abstracts met search criteria after removing duplicates

Application of inclusion/exclusion criteria: 20 papers met criteria

Application of Critical appraisal criteria: 11 papers met criteria and included for final analysis

Study selection

Of the 785 papers that were captured through database and hand searchers, 73 met the search criteria. These were reduced to 20 once the eligibility criteria were applied and duplicate references were removed. Following the full-text review and application of quality assurance criteria, 11 papers were included in final analysis (Figure a). The results of the search and the nature of the evidence to support these findings are summarized in Table 1.2. The quality of the 11 studies was generally satisfactory (rating greater than or equal to score of ‘+++’) when tested against the quality appraisal criteria.

Methodological characteristics of the studies

Methodological heterogeneity was found across the 11 studies. Six of the included studies employed qualitative methodologies (46%). Three utilized a mixed-method approach of

Page | 302

utilizing both qualitative and quantitative methods (23% of all study types). Four studies (31%) were purely quantitative in nature, with the use of descriptive cross-sectional survey methods. One study by Jayasekera (2000) was labelled as an interventionist ‘quasi- experimental’ study by design. However, since there were no clear ‘control’ arms as per the critical appraisal criteria applied, the paper was classified as a quantitative study which used “pre and post-survey” analysis with statistical measures to determine intervention effects.

Defining meaning and rationale for the community participatory approach

As explored in literature review, defining what CP means for a particular program has implications on its mode and utility for practice (455). Whilst many studies did not define CP its rationale for use was articulated within the introduction and conclusions of the study. For instance, the study by Ranasinghe (2011) articulated that establishment of community psychiatry services with the active involvement of community members in planning and implementations phases were shown to lower defaulter rates, ensure better treatment follow-up, and improved referral and case-management practice. Chapter 2 explored how CP as an approach to health care may be viewed from two perspectives − as a demand from communities or as a strategy for professional practice (73). In assessing the interventions described, it is clear that the community participatory approach was seen as a ‘technocratic process’ that served to fulfil a health program’s goal/objective, rather than as a ‘horizontal process’ which sought CP from a beneficiary ‘empowerment’ lens.

Only one study was found that utilized a horizontal ‘empowerment’ perspective. Wickramage et al. (2008) described a disease surveillance intervention within conflict displaced camp settings which involved communities actively leading surveillance, prevention and mitigation efforts (234). Intervention planners conceptualized CP as means of not only enhancing communicable disease surveillance but also a way of building community capacity, communal harmony and social cohesion amongst the ethnically diverse communities within the post-conflict setting. The authors suggested that participation of displaced communities during humanitarian ‘usually tends to concentrate on ‘consulting’ beneficiaries about their needs, rather than entrusting beneficiaries with control over the programme’. However, within this program, the goals were to not only for diseases surveillance but ensure intra-community dialogue and mechanisms to support these interactions.

Defining ‘community’

A diversity of community settings were described in the selected studies. These ranged from school settings, rural community settings, municipal council/urban setting, a network of senior citizens clubs, Internally Displaced Persons living in camp settings to epistemic communities for policy making. The dominant paradigms across all studies were in framing ‘community’ as health consumers: as a demarcation of a patient group, health consumer group, or a geographically defined population within a health service area. For instance, in the paper Page | 303

by Ranasinghe (2011) the definition adopted by health workers for a project in community psychiatry defined the ‘intervention space’ to be the provision of mental health services outside a hospital institution and encompass a well-defined catchment area demarcated geographically and administratively by the health system.

Of all studies reviewed, the meaning of ‘community’ was explicitly defined in only two studies. Pearson and colleagues (2010) who articulated community to be an ‘epistemic’ one - a network of professionals with recognised expertise and competence in a particular domain and an authoritative claim to policy-relevant knowledge in that domain or issue area (456). The development and consideration of policy options within this epistemic community enabled a richer understanding of the many important aspects of evidence, knowledge and linkages for health policymaking in suicide prevention, especially due to ingestion of pesticide. Wickramage (2008) defined community as “those displaced persons within the confines of a IDP camp”.

Disease characteristics

Most studies (46%) described interventions that involved CP approaches to address chronic non-communicable diseases (NCDs) such as mental health, suicide prevention and reproductive health care services (Table a). Nearly a third of all studies described community interventions that sought to address communicable diseases (CDs) such as dengue, malaria and community based infectious diseases surveillance systems in internally displaced camp settings. Two studies described interventions that were relevant to both NCDs and CDs.

Where did CP emerge during the intervention?

I was also interested in analysing at what point during the course of the health program/intervention did community participation emerge or was sought by those undertaking them. To do this I assessed each intervention against the major steps involved in the classic health intervention project cycle involves: conceptualization/design phase, appraisal and debate, implementation phase, monitoring and evaluation phase, and finally resolution of program/s. Figure c. shows the result of plotting the community participatory processors within steps of the ‘project cycle’. CP ‘manifests’ mainly during program implementation and program design phases. This result is was not surprising given the ‘technocratic approach’ in utilizing participation as a means of achieving health intervention goals.

Page | 304

Figure c. Where did CP emerge during the intervention? Assessing the manifestation of community participatory processors within steps of the project cycle.

Project cycle phase Paper reference Step 1 Step 2 Step 3 Step 4 Step 5 1 Jayawardene, x x 2011 2 Ranasinghe, x x x 2011 3 Konradsen, 2000 x x x x x 4 Yasuoka, 2006 x 5 Wickramage, x x x x 2008 6 Devendra, 1984 x x x x 7 Holmes, 2011 x x 8 Pearson, 2010 x x x 9 Gammanpila, x x x x 1997 10 Jayasekera, x x 2000 11 Perera, 1985 x x x 3 6 9 10 5

Page | 305

Effects on health outcomes and service provision

The following intervention effects were described in the studies reviewed.

Two studies reported that use of CP approaches led to enhanced service coverage. The study by Gammanpila (1997) reported community partnerships “boosted program coverage for malaria prevention and control”. Wickramage (2008) described how community based volunteer networks within IDP camps were able to enhance disease surveillance and outbreak response mechanisms that were “overstretched due to human resource for health limitations”.

Two studies described how efforts to involve community representatives in designing and planning interventions improved service accessibility. The study by Konradsen (2000) found the establishment of a ‘community co-operative health center’ reduced the delay of malaria treatment for adults by at least 2 days, and “significantly reduced the stress and discomfort” experienced by the elderly and handicapped segment of the community. Several studies described the formation of new services manifesting as a result of community involvement. The study by Holmes (2011), reported the formation of senior citizen clubs with retired tea estate workers after a period of sustained advocacy from a range of community stakeholders.

Other studies described how participatory approaches facilitated improvements in community health knowledge, attitudes and protective practices. The study by Jayasekera (2000) highlighted the improvements in child rearing and reproductive health behaviours as a result of the intervention.

Studies involving vector born disease control iterated the importance of cost-effectiveness in CP program design: dengue prevention and control activities through school-children and local community groups (Jayawardene,2011), the establishment of a community led malaria initiative through the establishment of a village health centre (Konradsen, 2000) (233), and a 20-week health promotion program to improve community knowledge on vector control within agricultural communities (Yasuoka, 2006) (230). In the study by Yasuoka and colleagues, the intervention program not only enhanced knowledge for behavioural change in relation to vector borne diseases but the low cost of materials used in intervention were also shown to reduce pest damage to rice crops, further enhancing motivation to sustain participatory action.

Limitations

A common problem in the analysis and interpretation of the results of intervention studies is of publication bias, where publication of reports from initiatives that were only judged to be successful would be presented for publication (457). The lack of information presented

Page | 306

within the confined word-limits of a publication formed a limitation. Being more explicit about program features, including the nature of the relationships between lay participant/s and the health system, the nature of community and health service, may have assisted in understanding the underlying mechanisms of participatory practices, potential differences between interventions, and changes over time.

Discussion

A review of more than 785 papers found only 11 studies that described CP in the planning, implementation and/or development of health care programs in Sri Lanka. This dearth of literature is remarkable given the historical support of PHC in Sri Lanka since 1978, for which CP is enshrined as a core pillar. No study sought to explicitly neither assess CP nor describe the nature/dynamics of participation and perspectives of health workers. Descriptions on CP were secondary/incidental to the core purpose of the research paper. All studies described were interventions that were not conducted as part of ‘routine’ health system interventions, but rather catalysed as part of a pilot project, interventional research process or a time-limited donor funded program (Table a). No studies reported an overall ‘negative result’ or adverse intervention affects. Most concluded with a recommendation to further develop intervention to routine system’ or ‘scale-up’.

Overall, the review of selected papers provide small body of evidence that suggests involving communities may have led to some enhancements in service coverage of specific interventions, improved health knowledge and led to better understanding of health needs and health determinants. Four studies described the effects of such participation on enhancing accessibility and acceptability of services.

None of reviewed papers analysed the perceptions of health workers who participated in the initiatives. As highlighted in the previous systematic literature review, perceptions of PHC workers on CP forms a powerful determinant of participatory practice in health care, and are therefore an important aspect of understanding CP practice.

In summary, despite Sri Lanka’s longstanding health policy commitment to CP, the literature revealed a scarcity of research undertaken in Sri Lanka that aimed at documenting interventions and approaches using CP in health. The literature reviewed suggests the ‘potential’ for community involvement in health service delivery and policy making in Sri Lanka. There is however a limited amount of empirical research evidence on the true effects of involving consumers and beneficiaries in studies reviewed.

Page | 307

Table a. Summary of Papers in Systematic Review Paper Study Aim Study design, Typology of Disease Health system Reported outcome/s reference and research method ‘community’ and type involvement and conclusions (Quality Intervention location appraisal (CD or score) NCD)*

Jayawarde Vector born Mixed method. Seventh to ninth- CD - PHC workers, Importance of the ne, 2011 disease control Quantitative (Pre grade students in Dengue. District Education community of school program and Post two schools in officers. Existing children in vector borne +++++ involving school entomological Ampara District, disease surveillance disease control students survey, analysis of Eastern Province system was highlighted. Surveillance routine public health of Sri Lanka. One augmented to data showed a 73% surveillance school was in monitor progress. reduction in case load systems) and rural setting, (pilot project) of vector born disease Quantitative other in urban for the urban area and (observational setting. a 61% reduction in the study, content rural area during the analysis of student year following journals). intervention.

Ranasinghe Development Qualitative data 11 volunteer NCD - Partnership Establishment of , 2011 of a (Case-study). community mental Mental between the community psychiatry Community health support Health. National Institute of services in this zone. ++ based workers across Mental Health and Lower defaulter rates, psychiatry 15 Colombo city PHC level of Increased community service municipal Council Colombo Municipal awareness of mental zones. Health System illness by the community, (pilot project) reduced stigmatization, better treatment follow- up, and improved referral and case- management practices.

Konradsen, Establishment Qualitative and Community in CD - Researchers in The village treatment of a community Quantitative collaboration with center it quickly took Page | 308

2000 led malaria methods. rural setting, Malaria. communities, over the role of main initiative provincial health provider for diagnosis ++++ including Anuradhapura authorities and the and treatment of establishment Anti-Malaria malaria. The treatment of a village District, North (intervention center did not improve health centre Central Province. research project) the response time in to reduce seeking treatment for morbidity, Campaign (AMC). young children, but the improve Community leaders delay for adults was diagnosis and forged a reduced by 1±2 days. enhance partnership where The treatment center treatment- the community co- significantly reduced the seeking invested in the stress and discomfort behaviour of a centre experienced by the rural establishment and elderly and community. maintenance. handicapped segment of the community.

Yasuoka, A 20-week Qualitative and Rural community CDs. Agriculture Ministry, CP interventions were 2006 education Quantitative Mosquit Development effective in increasing program to methods. Farmers in two o-borne (Mahaweli) understanding and ++++ improve villages in Sri diseases Authority, and active involvement in community Lanka were , Ministry of Health's mosquito control and knowledge selected for including regional officers of disease prevention in and mosquito intervention in malaria, Anti-Malaria rice ecosystems control with Anuradhapura dengue, Campaign (pilot regardless of socio- community district, North Japanes project) demographic participatory Central province e characteristics. approaches of Sri Lanka. Encepha litis, and filarisis.

Wickramage Explore the Quantitative and Internally CDs of International An early warning , 2008 impact of a CP Qualitative. Using a Displaced public humanitarian health system (EWARN) was

Page | 309

+++++ intervention combination of Persons living in health agency established in strategy during observational IDP camps importa collaboration with partnership with trained the studies, analysis of Batticaloa nce to Ministry of Health community health humanitarian operational data District. (390 IDPs and IDP community volunteers led to crisis in Sri records, case-study community based such as leaders improvements in early Lanka’s East narratives, and pre IDP health diarrhea detection and outbreak that sought to and post volunteers across l (one-off intervention response measures, engage war- intervention survey 27 camp disease. due to conflict) enhanced health affected methods. settings). promotion and communities to environmental health enhancing the management in routine comparison to IDP communicable settings with only the disease routine disease surveillance surveillance systems. system, and undertake health promotion activities in displaced camp settings.

Devendra, The Mixed method Two rural villages CDs A local community It is possible to involve 1984 intervention approach using in Hambantota (Helmint steering committee the community in health sought to qualitative methods district in Sri hic was assigned by decision making and ++++ integrate (case study), and Lanka’s Southern Parasite the local health managerial process of community quantitative province and one s), NCDs authority (MOH) on even complex/multi- based needs methods (a baseline a semi-urban (nutrition initiation of the level health projects at assessments survey). village in & family project to PHC level. A significant and health Panadura, Planning undertake a reduction in parasite priority setting, Colombo district ). sociological profile borne infections in all 3 parasite in Sri Lanka’s of all community project areas were (helminthic) Western based health observed. The control, child province. volunteers, a achievement in the nutrition, and baseline survey of fields of nutrition and

Page | 310

family Planning parasitic child health are not through use of infestations and readily quantifiable, CP knowledge on but future nutritional approaches. family planning surveys are expected to Other activities. objectively indicate the objectives Interventions were improvement which was include developed in subjectively observed. studying the partnership with The project success was mechanism of community to attributed to a number community address health care of enabling factors: high participation in gaps.(pilot project) level interagency 3 distinct representation and community close coordination with settings so that community based these would steering committees; provide viable field level coordination models for of government officials future were supported through expansion. resources, including intervention activities; project leadership by project Manager, Medical Officer; and considerable autonomy at project level for both PHC workers and community groups.

Holmes, Improving Qualitative (Case Fifty-five elderly NCDs. Unclear as to exact Forming Elders’ Clubs 2011 wellbeing of study) citizen clubs (total coordination with retired tea estate elderly tea of 3,913 persons) Aged mechanism with the workers in Sri Lanka +++ estate workers in Nuwara Eliya care. health system but suggests many benefits, over the age district, Central involved including social support of 60 years Province. coordination with and participation, who has little Ministry of Social psychological or no income, Services and wellbeing, inter-

Page | 311

living in poor Government generational contact, a and over- Administrative units. collective voice to crowded living (pilot project) influence politicians, conditions with government services limited access and estate to services. management, and facilitated access to health promotion activities, and to health care and social welfare services. These were suggested through qualitative data.

Pearson, Development Qualitative data Epistemic NCD - Participation of Use of epistemic 2010 of policy community Pesticide health care workers community actors to options to (Delphi method). defined as those poisonin in expert group identified factors that ++++ reduce access ‘outside’ g and based on an can facilitate policy to pesticides Government suicide epistemic change. and prevent policy making preventi community model. poisoning and process, but on on. (Conducted as part suicides. clinical, academic of research project) and agricultural domains.

Gammanpil Study aimed at Mixed method: Medawachiya Both The MOH of the The routine public health a, 1997 evaluating the Descriptive cross- MOH area (a NCDs area and the PHC registries were effectiveness sectional study primary health (matern workers (PHM, analysed to compare +++++ of CP utilizing care unit) of al and PHIs) worked with health indicators such as approaches Quantitative Vavuniya District, child VHCs and HVs coverage of bed-nets utilized in a (survey) and Northern health, attached to these for malaria control, MOH area Qualitative methods Province. nutrition, committees to indoor residual over a 4 year (FGDs). reprodu support the vertical spraying, contraceptive period. The ctive programs that use, maternal and child effect of CP in health) emerged from the nutritional indicators the form of and CDs PHC units. There etc., over the 4 year.

Page | 312

village health (malaria also appeared Data showed a committees control). opportunities for significant improvement (VHCs) and the VHCs and HVs in maternal and child health to provide input to health care outcomes, volunteers PHC workers on environmental health (HVs) on addressing certain and control of primary health public health issues. communicable disease care service (intervention indicators showing the provision of the research project) programme’s positive area, as well impact on the health as the extent care status of the of CP gained population concerned. through the Causality could not be programme determined since this intervention was not a experimental were analysed. study and since no comparative analysis was carried out. The effects of the CP intervention cannot be definitely linked to the improvements in MHC, CD, RH outcomes described.

Page | 313

Jayasekera, Study was Quasi-experimental The study was NCDs. Intervention was The IEC package 2000 carried out to study design conducted in the Reprodu coordinated by the ‘delivered’ through a assess the level (randomization was Kaluthata ctive Provincial Director PHC worker directly to +++++ of male not performed). A education zones, health of Health (PDHS) a group in the local participation in stratified cluster Southern and and delivered by community (in this case areas of sampling technique Province, among child- the PHI to selected married males), were contraception, was used to select 440 married men develop community groups effective in improving child rearing participants. The in 47 workplaces. ment. after training. knowledge, attitudes, and ‘intervention’ was Program monitoring child rearing and undertaking given to males in was undertaken by reproductive health house-hold the study group at the MOH with practices (contraceptive activities for their workplaces by PDHS office at the methods). The study wellbeing of PHI while the PHC level. showed that the positive mother and control group did (intervention effects were child. An not receive any research project) significantly felt when educational intervention. The the intervention was intervention study and control delivered by the PHI was carried groups were after an intensive out by Public geographically training period with Health separated to close follow up Inspectors (PHI) minimize monitoring by the health for selected contamination of program managers. male the intervention. households providing information on contraception, child rearing and household interventions using lectures, IEC material and participant discussions.

Page | 314

Perera, The study was Qualitative study The study was Both The VHU committee Both Induced and 1985 undertaken to (FGDs) and analysis conducted in the NCDs consists of village indirect CP in primary identify of routine public Southern (matern leaders which health care is feasible. +++ demographic health reports. Province. al and nominates the HVs A VHC and a network characteristics child to work with the of HVs can boost PHC and activities health, PHC units in the worker capacities of a group of nutrition, MOH area. through assisting with voluntary environ (intervention vertical programs such health workers mental research project) as environmental health. (HVs) and health) Potential power assess the and CDs differentials between impact of (vector those elected community activities of a born representatives in VHCs primary health disease and HVs were not care control). examined. Effects of the development CP intervention cannot project. be casually linked to the improvements in health outcomes.

*Non-communicable Disease (NCD), Communicable disease (CD) type.

Page | 315

Implications of findings of literature review on research objectives

A review of more than 785 papers found only 11 studies that described CP in the planning, implementation and/or development of health care programs in Sri Lanka. This dearth of literature is remarkable given the historical support of PHC in Sri Lanka since 1978, for which CP is enshrined as a core pillar. No study sought to explicitly neither assess CP nor describe the nature of participation and perspectives of health workers. Descriptions on CP were incidental to the core purpose of the research paper. All studies described were interventions that were not conducted as part of ‘routine’ health system interventions, but rather catalysed as part of a pilot project, intervention research process or a time-limited donor funded program (Table a.). No studies reported an overall ‘negative result’ or adverse intervention affects. Most concluded with a recommendation to further develop intervention to routine system’ or ‘scale-up’.

Page | 316

Annexure 3. Ethics Approval

HUMAN RESEARCH ETHICS 12 S e p tember 2011 COMMITTEE (HREC)

Professor Anthony Zwi School of Public Health and Community Medicine

Dear Professor Zwi

“How participatory are we?”- Role of Primary Health Care units in addressing the growing burden of Chronic Non-Communicable Diseases through community participatory approaches in Sri Lanka HREC 11240

Thank you for the email and attachments from Kolitha Wickramage to Mrs Annamarie D'Souza dated 27 September 2011.

The Executive of the Human Research Ethics Committee considered the above protocol at its meeting held on 27 September 2011 and is pleased to advise it is satisfied that this protocol meets the requirements as set out in the National Statement on Ethical Conduct in Human Research*.

Having taken into account the advice of the Committee, the Deputy Vice-Chancellor (Research) has approved the project to proceed.

Would you please note -:

approval is valid for five years (from the date of the executive meeting i.e. 27 September 2011);

you will be required to provide annual reports on the study’s progress to the HREC, as recommended by the National Statement;

you are required to immediately report to the Ethics Secretariat anything which might warrant review of ethical approval of the protocol (National Statement 3.3.22, 5.5.7) including:

a) serious or unexpected outcomes experienced by research participants (using the Serious Adverse Event proforma on the University website at http://www.gmo.unsw.edu.au/Ethics/HumanEthics/InformationForApplicants/Proforma sTemplates/C13_SAE%20Proforma.rtf) ; b) proposed changes in the protocol; and c) unforeseen events or new information (eg from other studies) that might affect continued ethical acceptability of the project or may indicate the need for amendments to the protocol;

any modifications to the project must have prior written approval and be ratified by any other relevant Human Research Ethics Committee, as appropriate; ..1..

UNSW SYDNEY NSW 2052 A U S T R A L I A Telephone: +61 (2) 9385 4234 Facsimile: +61 (2) 9385 6648 Email: [email protected] Location: Rupert Myers Building Grants Management Office/ Ethics Gate 14 Barker Street Kensington ABN 57 195 873 179

Page | 317

HUMAN RESEARCH ETHICS COMMITTEE (HREC)

(HREC 11240 cont’d)

..2..

if there are implantable devices, the researcher must establish a system for tracking the participants with implantable devices for the lifetime of the device (with consent) and report any device incidents to the TGA;

if the research project is discontinued before the expected date of completion, the researcher is required to inform the HREC and other relevant institutions (and where possible, research participants), giving reasons. For multi-site research, or where there has been multiple ethical review, the researcher must advise how this will be communicated before the research begins (National Statement 3.3.23 and 5.5.6);

consent forms are to be retained within the archives of the School and made available to the Committee upon request.

Yours sincerely,

Professor Michael Grimm Presiding Member HREC

*http://www.nhmrc.gov.au

Annexure 4. Official endorsement and letter of support from Ministry of Health

Page | 319

Page | 320

Annexure 5. Semi-structured interview guide (prompts in brackets)

1. What does ‘community’ mean for you? What does ‘community participation’ mean for you?

What rights and duties does it imply (for both the community and the workers themselves)? What do you view as the benefits (if any) or disadvantages (if any) of community participation?

2. How do you as a health professional engage with your community?

What does working in a ‘participatory’ approach mean for you? What value and importance do you give to community participatory approaches in health services? Can you describe the nature/form of community participation in the setting in which you work? Are community participatory approaches to health care formally organised within the existing district health care system? Do you have community volunteers that work within your organisation? Are there mechanisms to enable health volunteerism at your institution? In your experience, what activities and roles enhance participation? What do you think working ‘effectively with communities’ mean? What are the skills and competencies workers need (if any) to work effectively with communities in a participatory manner? What are the resources that are required to enable and/or sustain meaningful participation? What are the factors that would influence or motivate you to engage in participatory approaches? What do you feel are the barriers for you as a worker for enabling community engagement in your programs?

3. What are your views and experiences with Chronic Non-Communicable Diseases (NCDs) in your community?

What are the types of issues and conditions you see in your daily work (if any)? What do you feel are the risk factors for NCDs relevant to the community you work in? What value do you give to NCDs in relation to their other diseases/health priorities you face? How do you address NCDs through your work (if at all)? Do you feel primary health care workers have a role to in the prevention and control of NCDs at Community level?

Page | 321

4. What types of strategies/interventions do you perceive as being effective in addressing NCDs at the community level?

Do they aim at addressing a range of risk factors or formulate specific strategies for selected risk factors/diseases? Do you feel equipped and able to address these strategies? In your view what are the ‘resources’ needed to address NCD prevention and promotion at community level? What type of knowledge, training and skills do you require in carrying out the described activities and duties? What type of training /exposure on NCD’s (if any) have you received/gained? Are there any examples of programs, approaches or strategies currently in place at your work setting in address NCDs? If yes, describe these (goal of the intervention, its evolution etc.). Who were the stakeholders involved in catalysing it, sustaining it? What led to its sustainability/decline or devolution? What lessons (if any) have been learnt or are being learnt from such interventions? What were the resources needed/allocated, and costs associated with the intervention? How were these organised and managed? How were communities engaged in the intervention? In your view, did the intervention achieve the any established/intended health outcomes? What were the challenges and opportunities and constraints? How was the impact evaluated and measured?

5. Application of the Participatory Score Card

In your view, what is the level/degree of community participation that currently exists in the health care setting you work in? Categorize the level of participation (gradient ranging from no community involvement to communities as equal shareholders/partners in driving health programs).

Page | 322

Annexure 6. Checklist for exploring CP within health policy documents

Page | 323

Annexure 7. Policies, strategic plans and domestic legal frameworks pertaining to health care in Sri Lanka. Policies, strategic plans and domestic legal frameworks pertaining to health care in Sri Lanka. “A” denotes if CP or community participatory approaches have been explicitly articulated within the main body of the policy document as a means of achieving policy goals or objectives. “B” denotes if the accompanying documents and guides pertaining to the police (e.g. policy action plan, policy implementation framework or policy guidance note) explicitly express CP or community participatory approaches. An unclear result is indicated with a dash (-).

Health policies, strategic planning frameworks and domestic legal documents ‘A’ ‘B’ pertaining to health care in Sri Lanka (A) Policies 1. National Health Policy of Sri Lanka (2002) Yes - 2. National Mental Health Policy (2005) Yes Yes 3. National Maternal and Child Health policy (2005) Yes Yes 4. Sri Lanka National Health Promotion Policy (2010) (draft) Yes No 5. National HIV/AIDS Policy, Sri Lanka (2011) Yes Yes 6. National policy and strategic framework on injury prevention and Yes No management in Sri Lanka (2009) 7. National Policy for Prevention and Control of Chronic NCDs (2009) Yes Yes 8. National Medicinal Drug Policy for Sri Lanka (2005) No No 9. National Nutrition Policy of Sri Lanka (2008)* Yes Yes 10. National Policy on Health of the young (2011) Yes Yes 11. National Oral Health Policy for Sri Lanka (2012) (draft) No No 12. National Migration Health Policy (2013) No Yes 13. National Policy on disability and health (2012 - draft) Yes Yes 14. Sri Lanka code for the promotion, protection and support of breast feeding Yes Yes and marketing of designed products (2002) 15. Draft National e-Health Policy (2012) Yes - 16. National Strategy for Research Priorities in Health (2013) Yes - 17. National Policy on Tobacco and Alcohol (2007) Yes Yes (B) Health Sector Master Plan and other strategic planning documents 1. The Charter for Health Development (1980) Yes Yes 2. Health Sector Master Plan (2006-2016) Yes Yes 3. Annual Action Plan 2010 (Programmes & Directorates) No Yes 4. Annual Action Plan 2010 (Line Ministry Institutions) No Yes 5. Guidelines for Development of Annual Action Plan & Progress Report For No Yes Programmes and Directorates (October 2009) 6. Guidelines for Development of Medium-Term Strategic Plans For Programmes No Yes and Directorates (March 2010) 7. Guidelines for Development of Medium-Term Plans, Annual Plans & Progress No Yes Reports For Line Ministry Institutions (December 2009) 8. Guidelines for strengthening health care at primary level. Policy analysis Unit Yes Yes and Primary Care Services Unit (2012) (C) Government’s Political Manifesto “Mahinda Chintana” - Sri Lanka’s Ten Year Horizon Development Framework Yes - 2006-2016 forming the platform for current government policy Page | 324

(D) Domestic Legal frameworks pertaining to health** Constitutional law 1. Constitution of Sri Lanka Health Related laws 1. National Authority on Tobacco and Alcohol Bill (2007) 2. National Blood Transfusion Services Bill 3. Drug Dependant Persons (Treatment & Rehabilitation) Act 2007 4. Food Act (No. 26 of 1980) 5. Food (Amendment ) Act No. 29 OF 2011 6. Health Services Act (Acts. nos. 12 of 1952, 10 of 1956, 13 of 1962, Law No. 3 of 1977) 7. Sri Lanka Code for Promotion, Protection and Support of Breast feeding and. Marketing of Designated Products (2004) 8. National Authority on Tobacco and Alcohol Act 2006 9. National Kidney Foundation of Sri Lanka (Incorporation) Act 2006 10. Nurses Act 2005 11. Drug Dependant Persons (Treatment & Rehabilitation) Act 2007 12. Poisons Opium And Dangerous Drugs Act 1984 (Act no. 13) 13. The Penal Code (Ordinance no. 2 of 1983 as subsequently amended), Chapter 14 on public health and safety relating to illicit drugs. 14. The Cosmetics, Devices and Drugs Act (Act no.27 of 1980, as amended by Act no 38 of 1984) 15. The Customs Ordinance (Ordinance no. 17 of 1869, as subsequently amended) schedule B on medicinal substances 16. The Ayurveda Act (Act no. 31 of 1961 as amended by Act no 5 of 1962). 17. National Health Development Fund Act No13. of 1981 18. Private Medical Institutions (Registration) Act 2006 19. Protection of the Rights of Persons with Disabilities Act 2003 20. Sri Lanka Federation of the Visually Handicapped Act 2007 21. Tobacco Tax (Amendment) Act 2004 22. Vijaya Kumaratunga Memorial Hospital Board Act 1999 23. The Sri Lanka Animal Diseases Act of 1992 24. Quarantine Act. No. 12 of 1952. (Chap. 553) 25. Butchers (amendment) act, No. 13 OF 2008 26. Prevention of mosquito breeding act, No. 11 OF 2007 27. National authority on tobacco and alcohol, No. 27 OF 2006 28. Sri Lanka disaster management Act, No. 13 OF 2005(relevant for health sector) 29. Increase of fines Act, No. 12 OF 2005(relevant for Public health inspectors) 30. Competent authority (powers and functions) Act, No. 3 OF 2002 (relevant for Public health inspectors) 31. Factories (amendment) Act, No. 19 OF 2002 (provisions for occupational health and safety) *Please note the Food and Nutrition Policy of Sri Lanka developed in 2003, was revitalized and updated as the National Nutrition Policy of Sri Lanka in 2008. **Legislation (or statutory law) is law which has been enacted by a legislature (parliament). Once approved in Parliament (i.e. after enactment), under the Westminster system (for which Sri Lanka ascribes to), a legislation is known as an Act of Parliament.

Page | 325

Annexure 8. Hypothesizing the potential linkages between domains

The CP in PHC Analytical framework presented in Figure 36 depicts the conceptual framework that indicated 4 broad domains/environments that influence the meaning, realization, manifestation and diffusion of community participatory practices to address NCDs at primary health care level in Sri Lanka. This section provides a description of the possible links between these different domains. It must be noted however that not all such associations described have been elicited through the empirical work. Only those with definitive causal links have been rendered within the analysis of the main manuscript. This analysis should therefore be considered an explanatory frame that requires further research for establishing definitive links/associations between the domains.

Administrative and Resource Environment Policy and Regulatory Environment  Despite an enabling health policy framework that advocates CP in health, health administrators and program directors of various programs within the Ministry of Health has not embraced CP methods and in particular within PHC practice and for NCD prevention and control. This may be a result of health stewardship failure by senior health executives and policy makers to meaningfully engage and sensitize appropriate program directorates of the Ministry of Health and to mobilize technical and financial resources to translate policy to practice. The links between policy and practice need to be strengthened through collaborative engagement of stakeholders within the health system (as elaborated in section 6.3.1).  The lacuna between stated policy objectives and organizational and programmatic practices requires critical evaluation. The advancement of policy rhetoric to programs need sustained advocacy well beyond the phase of policy formulation and endorsement. The establishment of a monitoring framework (with program indicators) on CP and more broadly a policy accountability framework (in the form of a review

Page | 326

board or some form of audit) has not been included in the majority of national health policy documents and accompanying action plans. Resource allocations for such activities are not monitored (see Annexure 7). Such measures are anchored to good health sector governance and the stewardship function of the Ministry of health. Administrative and Resource Environment Attitudinal Environment  Little or no participatory engagement in the system may serve to demotivate PHC workers and health administrators in engaging with each other through consultation for PHC program design, and/or demotivate PHC workers to work collaborative with communities as partners in developing community health programs to tackle NCDs. Administrative and Resource Environment Technical competency and knowledge environment  Whilst it may be intuitive to understand why CP is important for NCD prevention and control, tailoring strategies to affect CP within the health system (i.e. the ‘how’ of manifesting CP in health) is less intuitive for program directors and planners within the Ministry of Health. Administrators may more readily engage with disease specific programs such as malaria control or tuberculosis than in health promotive ones.  Whilst the study clearly indicated the commitment of senior experts, front line health workers, middle level managers and directors at Ministry of Health to promote CP in health, the in-depth knowledge and technical competencies of those interviewed in design and implementation of CP interventions were not systematically assessed.  There is a critical need for technical support and capacity building for assessing the various models of CP, evaluation of existing programs and planning for evidence- informed approaches that are sensitive to cultural contexts and resource realties. Policy and Regulatory Environment Technical competency and knowledge environment  None of the policy documents indicated any specific program indicators targeting CP in health within monitoring frameworks, and only a few promoted the need for general policy accountability frameworks. For instance, the National NCD policy which articulated periodic reviews and establishment of result based evaluation systems, even though specific measures on CP were not discussed. The National Policy on Tobacco and Alcohol (2007), emphasized how the policy will be evaluated and financed and what measures would be taken to monitor its implementation. More critically, the need to identify specific technical competencies and resource allocations to catalyse CP for the achievement of health policy goals have not been explicitly indicated in the 17 policies analysed.  The limited emphasis on ensuring and boosting technical competency within policy and administrative environments is reflected in the responses by PHC workers (see sections 5.2.1 and 5.2.2). PHC workers noted a lack of formal training and expertise on the processes and strategies to enable CP. Their current role was articulated by a PHM as to “simply encourage people to go get tested”. Beyond clinical knowledge, the ability to effectively communicate messages on NCDs to general public and patients was also highlighted by senior administrators as a key gap. Research findings from Section 5.7 suggest that even when innovations do emerge from PHC workers, these community-based health initiatives are not readily absorbed or

Page | 327

supported by health administrators, thereby inhibiting their further development. The diffusion of such innovative CP approaches within the health system is also inhibited. Policy and Regulatory Environment Attitudinal Environment  Existing PHC work plans, worker performance monitoring and reporting requirements do not prioritize CP approaches, nor do they stipulate community-based NCD prevention and health promotion interventions. Even locally driven NCD prevention initiatives that were found to utilize highly successfully community participatory mechanisms with indigenously sourced resources (such as those documented in Section 5.7) were not readily harnessed by district-level health systems. Community-based health promotion initiatives have rarely been integrated into routine PHC practice. This showed the clear disjoint between underlying policy and practice.  PHC workers indicated that intensive engagement with communities provides better opportunities to understand current community health needs, practices and health behaviours. These findings may not only assist PHC workers and health administrators to tailor health interventions according to local capacities, but also promote health system accountability. Greater investments in resources and training for PHC workers to utilize CP strategies may lead to better population coverage and effective uptake of existing PHC interventions, and have the concomitant effect of building greater interest and support for CP among the PHC workforce. Since CP enables the formation of active community groups or citizen coalitions, these may be harnessed by the health system to initiate other health sector interventions. Technical competency and knowledge Attitudinal Environment environment  As indicated in sections 5.1.7, 5.2.2 and 5.2.3, there is a lack of knowledge among PHC workers and HAs of what constitutes effective CP approaches, and how these are to be implemented. Confusion in the coordination roles of members within each PHC team was also noted, especially with regards to chronic disease prevention. The confusion in coordination roles among members of some PHC units in undertaking community health initiatives extended to fuelling anxieties among some MOHs, many of whom were engaged in private sector general practice clinics in the local area. They resisted any empowerment of allied health professionals within PHC teams and opposed efforts to enhance even basic clinical skills and knowledge. Suggestions that front line PHC workers such as PHMs and PHIs be upskilled to offer basic clinical examinations such as blood pressure assessments at household level, and to provide knowledge on commonly used chronic disease medications such as statin drugs, were not supported.  Training curricula, at both pre-service and in-service levels, place little emphasis on empowering PHC workers with the required skills in community development and health promotion (section 6.3). Workers also expressed the need to enhance their capacities in community development and health advocacy, which they saw as crucial if underlying determinants of NCDs were to be addressed (section 6.3.4). Boosting capacities for inter-sectoral efforts were also highlighted as a priority by district health administrators.

Page | 328

Annexure 9. Some reflections on study challenges/limitations a. Challenges in analysing the ‘thick’ layers of research evidence distilled from an in-depth, multi-method research approach

Health system scholars have suggested that adopting a multiple-method approach results in “expanding the breadth and range of research inquiry” and can play a valuable role in health services research. Documenting the “CP landscape” in Sri Lanka’s health system requires first the unpacking of the meaning, perceptions and experiences of CP through interviews and narratives of health workers and authorities, and then searching for observations and manifestations of CP practice across the PHC settings. As reflected in the section on study design (Figure 15), the applied ethnographic approach was divided into four parts (Figure 18) that focused on ethnographic and observational study methods that enabled me to comprehensively study the phenomena of CP within the routine interventions of PHC workers.

Several kinds of triangulation were formally employed in this study: the first was across multiple sources of data (health professionals at central, district, village and PHC administrative levels, and those involved in policy formulation as well as direct service delivery); and the second was across a multiplicity of methods (from surveys, interviews, focused ethnography, descriptive analysis of work plans, policy documents). Such triangulation techniques across these research methods served to ‘strengthens credibility of research’ (291), despite being extremely time-intensive to perform15.

The strength of this multi-method approach was that it allowed for a more robust and rigorous approach in documenting the extent of PHC practice. It also generated voluminous layers of qualitative and quantitative data, for instance on the routine work practices of PHC workers and their work settings.

However, the key strength of this multi-dimensional approach underpinning rich layers of evidence also provided its greatest challenge. Layers upon layers of data (‘thick descriptions’) were generated during the multi-method research process. In-depth analysis of dense layers of data, that also emerged ‘sequentially’, proved a major challenge given time and resource constraints. The results of the 7 over-arching research methods emerged sequentially since different research components were initiated and undertaken in a ‘staggered way’ over the course of 4.5 years of field research. In- country research realities, time and resource constraints meant the diverse array of research methods needed to “fit” in the staggered formation in order to ensure completion. In particular, I underestimated the extensive time and resource intensiveness in undertaking the data synthesis and analysis of the rich folds of qualitative and quantitative data. As a result, there were components of the study that were not presented in the final thesis due to time constraints. Section 4.1 only presented 5 distinct research methods used in the study, where In fact, I had undertaken additional

15 As described in Figure 22, I harnessed the capacity of two bi-lingual research assistants to assist me undertake data transcription and translation. This collaborative method strengthened the rigor of the language-based inquiry, and as indicated by Larkin (2007), offered greater possibilities for construction of nuance and meaning. These assistants were not however involved in data analysis and theory development. Page | 329

components. For instance a Delphi study involved an expert panel of health administrators and senior health policy makers on reforming PHC and role of PHC workers. Through the iterative process catalysed through the Delphi-method, participants highlighted barriers for the PHC worker revitalization of process: a) insufficient human resources at PHC level to carry out screening, b) Follow up of high risk individuals, c) Lack of training on health guidance and health promotion and meaningful participation with communities, d) a lack of efficient referral system for continuity of care, e) Inadequate participation between curative and preventive sector for screening, community based partnerships for health promotion. Consensus was reached that there should be better linkages between the primary level curative institutions such as rural hospitals and the community health services delivered through MOH units. These linkages were proposed to improve health seeking behaviour, greater individual and community responsibilities towards health and for improving follow-up care from those discharged with chronic diseases. The inclusion of the above analysis though not included in the final thesis, served to further validate the findings of the other components of the study, notably the qualitative interviews and group discussions.

The challenge in analysing the ‘thick’ layers of research evidence emerging from a comprehensive multi-method approach, with limited time and resource support is highlighted. Better planning and ‘pruning’ of some methods (such as applied ethnography) would have also allowed for a more manageable (and less stressful!) data-analysis process. Nevertheless, the multiple-method approach adapted resulted in providing a more rigorous analysis that served to expand the breadth and range of research inquiry on CP. b. Limited exploration of the “healthy life-style centres” initiative

During the 2013/4 period, the Ministry of Health embarked on an expansion of the “healthy life-style centres” program that are driven by the goal of PHC units systematically assessing those in local communities between 35 and 65 years of age on a physical risk factor assessment. The context of participation in the model was for PHC workers to encourage and recruit community members to undertake the screening during a clinic that was to be held once a month. A “Package of Essential Non-communicable Disease Interventions for primary care” (PEN) was also aligned to these centres and supported by the WHO country office in Sri Lanka. The PEN program focused primarily on providing medical equipment such as blood glucometers, peak flow meters and essential drugs to PHC settings. Whilst such measures are vital for empowering PHC units with the capacity to treat conditions like hypertension, the health promotion interventions with engagement from local communities for combatting NCDs has received little attention and resourcing. Although the “healthy life-style centres” were identified and mapped as an ongoing initiative, it was excluded from case-study analysis since it met the established exclusion criteria of being a “pilot intervention” (Table 13). In the design and formative planning of the study, there was clear reluctance by WHO country office focal point managing the PEN project and some health administrators in providing access and clearance for undertaking an assessment of CP within the pilot project. However, the inclusion of this initiative would have been useful in revealing the potential mechanisms of CP action (or inaction) within the PHC space. This rigorous application of investigative Page | 330

tools utilized in this thesis to explore CP (as applied to two other NCD prevention interventions (presented in Part D, Section 5.7) would have been useful. However, since the pilot project was programmatically sensitive for the interests of WHO, official approval from relevant authorities such research engagement was not possible. As highlighted in the policy recommendations section, establishment of an NCD research consortium under the auspices of the Ministry of Health and in partnership with academia, research institutes and health development partners may contribute to systematically assessing all NCD prevention initiatives and providing evidence-informed guidance for future programs and policies. c. Attitudinal barriers to investing in CP by senior policy makers require further elucidation.

Interviews with health policy makers as to why financing and investment of CP initiatives within PCH has not catalysed, despite the highly positive policy rhetoric towards CP, requires further investigation. The results section presented a number of reasons for the lack of commitment to resourcing CP in health. These ranged from the divergent understanding by policy makers and planners into resourcing costs, design and the management dynamics/requirements of such CP interventions for NCD prevention. Though majority of health administrators supported CP approaches in PHC, a few responders highlighted the lack of data/evidence as to the effectiveness of such CP interventions and the difficulty in harnessing this evidence of effectiveness. Some also favoured primary prevention efforts such as screening for NCD risk factors rather than health promotive initiatives that sought community engagement. Further in-depth exploration of policy maker perceptions are therefore needed.

Page | 331

Definitions of some key concepts used in the thesis

Stewardship

Stewardship, sometimes more narrowly defined as governance, refers to the wide range of functions carried out by governments as they seek to achieve national health policy objectives. In addition to improving overall levels of population health, objectives are likely to be framed in terms of equity, coverage, access, quality, and patients' rights. National policy may also define the relative roles and responsibilities of the public, private and voluntary sectors - as well as civil society - in the provision and financing of health care.

Stewardship is a political process that involves balancing competing influences and demands. It will include: maintaining the strategic direction of policy development and implementation; detecting and correcting undesirable trends and distortions; articulating the case for health in national development; regulating the behaviour of a wide range of actors - from health care financiers to health care providers; and establishing effective accountability mechanisms. Beyond the formal health system stewardship means ensuring that other areas of government policy and legislation promote - or at least do not undermine - peoples' health. In countries that receive significant amounts of development assistance, stewardship will be concerned with managing these resources in ways that promote national leadership, contribute to the achievement of agreed policy goals, and strengthen national management systems. While the scope for exercising stewardship functions is greatest at the national level, the concept can also cover the steering role of regional and local authorities.

A key concern in many countries is to build the capacity needed to carry out stewardship functions effectively. This, in turn, requires a better understanding of what constitutes best practice when it comes to stewardship and how national leadership can be developed. It is increasingly recognized that the provision of development assistance needs to be geared to fulfilling these objectives.

Reference: Saltman, R. B., & Ferroussier-Davis, O. (2000). The concept of stewardship in health policy. Bulletin of the World Health Organization, 78(6), 732-739. Pedagogical approach Pedagogy concerns the study and practice of “how best to teach”. Its aims range from the general education of an individual to the narrower specifics of vocational education such as imparting and acquisition of specific skills. CP in health is seen by most PHC workers and administrators from a “technocratic perspective” in which CP is an intervention tool to be “wielded” by health program designers to help fulfil health program goals - where communities can assist PHC workers in program implementation to achieve public health outcomes. This concept of CP is quite distinct from a pedagogical approach in which CP aims to empower communities and facilitate agency of community members in taking forward their own health development. The pedagogical approach views CP as a process for ensuring equitable reshaping of power and facilitating the empowerment of community members to actively participate in their own development. Reference:

References:

Page | 332

 Issel, L. M. (2004). Health program planning and evaluation: A practical, systematic approach for community health. Jones & Bartlett Learning.USA.  Wallerstein, N., & Bernstein, E. (1988). Empowerment education: Freire's ideas adapted to health education. Health Education & Behavior, 15(4), 379-394.

A Human Rights-Based Approach

The right to the highest attainable standard of health (referred to in short as “the right to health”) was first reflected in WHO's Constitution and has been firmly endorsed in a wide range of international and regional human rights instruments. The most authoritative interpretation of the right to health is outlined in Article 12 of the International Covenant on Economic, Social and Cultural Rights, which has been ratified by approximately 150 countries. In May 2000 the UN Committee on Economic, Social and Cultural Rights, which monitors the Covenant, adopted a General Comment on the right to health that further clarifies the nature, scope and content of the right to health.

A human rights-based approach to development cooperation and development programming may include:

 All programmes of development cooperation, policies and technical assistance should further the realization of human rights as laid down in the Universal Declaration of Human Rights and other international human rights instruments.  Human rights standards contained in, and principles derived from, the Universal Declaration of Human Rights and other international human rights instruments guide all development cooperation and programming in all sectors and in all phases of the programming process.  Development cooperation contributes to the development of the capacities of “duty- bearers” to meet their obligations and/or of “rights-holders” to claim their rights.

In relation to health, a rights-based approach means integrating human rights norms and principles in the design, implementation, monitoring, and evaluation of health-related policies and programmes. These include human dignity, attention to the needs and rights of vulnerable groups, and an emphasis on ensuring that health systems are made accessible to all. The principle of equality and freedom from discrimination is central, including discrimination on the basis of sex and gender roles. Integrating human rights into development also means empowering poor people, ensuring their participation in decision-making processes which concern them and incorporating accountability mechanisms which they can access.

References:  De Vos, P., De Ceukelaire, W., Malaise, G., Pérez, D., Lefèvre, P., & Van der Stuyft, P. (2009). Health through people's empowerment: a rights-based approach to participation. health and human rights, 23-35.  London, L. (2008). What is a human-rights based approach to health and does it matter?. Health and human rights, 65-80.  WHO (2015) Human Rights-Based Approach to Health

Page | 333

Expert Panels on Primary Health Care (PHC) Reform including Role of PHC Workers

This expert feedback method was applied to a series of workshops involving a range of experts from the Ministry of Health, academia and health sector partners such as United Nations agencies involved in formulating a strategy for ‘Primary level health care staff work responsibilities’ (MOH, 2012)16.

During my PhD study period, a series of three technical workshops were coordinated by the Directorate of Policy and development of the MOH as part of the development of a national guidance document on revitalizing Sri Lanka’s health system. The final report was entitled: “Approach and Guidelines for strengthening health care at primary level” system (MOH, 2008)17, and emphasized the need towards re-orienting primary health care and the role of PHC workers (MOH, 2012). I was invited by the MOH to co- convenor of one of the workshops, and participated in all three.

The experts were identified by the Ministry of Health, and were invited to attend to a series of three consecutive workshops led by the Directorate of Policy and Planning. Three workshops held were: (1) ‘Primary health care level staff categories and their work responsibilities’ [16th September 2008]; (2) “Primary Curative Care Model” [18th to 19th March 2010], and (3) “National Scientific and Policy Forum for Strengthening Primary level healthcare” [3rd September 2010]. The series of three workshops involved experts from a diverse range of backgrounds ranging from senior policy makers and government administrators, leading clinicians in chronic disease management, health system specialists, academics, independent health policy analysis and health economists.

All technical discussions involved reflections and deconstructions of PHC worker duty lists in relation to the debate surrounding revitalization of Sri Lanka’s PHC system. The first two workshops used an iterative method to build expert consensus: First, presentations by health system analysts were made on current health system model and their perspective on such models (for reform or augmentation) were shared18. This was followed by group discussion through the format for ‘group work’. The groups were organized according to the participants’ expertise and place of work to ensure relative heterogeneity. After the group work, presentations were made by a nominated group leader and final consensus built on emergent themes. The third and final workshop involved consolidation of the emergent themes from the proceedings of previous workshops, and final synthesis of consensus statements. Consensus was achieved through sharing the ‘working document’ synthesizing discussed content to all expert participants, who in turn provided feedback to the moderator to further consolidate their views in subsequent workshops (2 and 3).

I assessed the statements and actions plans pertaining to PHC workers that emerged after multiple rounds of expert consensus in results.

16 Ministry of Health (2012) Approach and Guidelines for strengthening health care at primary level. Policy analysis Unit and Primary Care Services Unit.MOH, Publications. 17 Ministry of Health (2008) Report on The Consultative workshop on Primary care level public health staff categories and their work responsibilities. 16th September 2008. Policy analysis Unit, MoH 18 Pre-prepared posters highlighting existing duty list and human resource issues relevant to public health and NCD programs were also displaced for the benefit of the participants as helpful references. Page | 334

Through iterative discussions and debate, goal of the series was through expert analysis construct a list of potential roles for PHC workers and a National Strategy for reforming PHC at District level health systems. Although no attempt was made to deliberately construct these workshops according to a Delphi method19 [43], the methodology used yielded to steps of a Delphi method (Figure A) [45]. This model was not a ‘true’ Delphi method since the investigators did not have control over the development of the organizational process. In the classic Delphi method, the same experts will provide feedback to the research panel on the resulting discussions and final consensus reached. However, since these workshops were not formally organized to illicit and track individual feedback, the method did not hold true of a Delphi method.

Figure A. Organisation of Expert Panels and Technical Workshops on exploring role of PHC workers

Participants

The aim of the series of workshops organized by the Directorate of Policy of the Ministry of Health was to involve in-country experts to build consensus on how best to revise the roles of PHC workers in the context of major shifts in epidemiological disease burden such as rising NCDs. Held through three iterative cycles of group work, the ‘delphi method like’ approach aimed at identifying strategies and reforms at PHC level to instill such changes.

Experts were all Sri Lankan based comprised mainly of clinicians, health administrators and academics identified by the Ministry of Health. The domain knowledge of experts ranged from those with expertise in community medicine, clinical medicine (e.g. Endocrinologists), medical administrators, and health economists. In each ‘round’ (technical workshop), an average of 40 experts participated in the technical discussion. The full lists of participants are provided in the Annexure.

19 An ‘expert survey’ performed in sequence of rounds, whereby the results of the previous round are given as feedback to experts to gain consensus Page | 335

Summary of findings

Rationale for revitalization. A number of clear conclusions emerged from the expert meetings on the need for revitalization of Sri Lanka’s existing PHC. Firstly, the agreement that current job duties, referral pathways and institutional capacities were not fully conditioned to effectively address the growing burden of chronic disease and Sri Lanka’s ageing population. Secondly, the importance of meaningful involvement of the health system with local communities. Thus the need for PHC system to better engage with communities and utilize community participatory approaches were endorsed. Third, models of inter-professional cooperation within PHC teams and creating multi- disciplinary team within a ‘family medicine’ approach was espoused.

Barriers for revitalization. Through the iterative process, participants highlighted barriers for the PHC worker revitalization of process: a) insufficient human resources at PHC level to carry out screening, b) Follow up of high risk individuals, c) Lack of training on health guidance and health promotion and meaningful participation with communities, d) a lack of efficient referral system for continuity of care, e) Inadequate participation between curative and preventive sector for screening, community based partnerships for health promotion.

Proposed strategies formulated through expert group consensus to enhance PHC worker roles/job duties.

Consensus was reached that there should be better linkages between the primary level curative institutions such as rural hospitals and the community health services delivered through MOH units. These linkages were proposed to improve health seeking behavior, greater individual and community responsibilities towards health and for improving follow-up care from those discharged with chronic diseases.

The scope of work of the ‘Community health team’ to deliver care within a family medicine approach were also elaborated. The proposed model emphasizes rationalizing outpatient and community care and envisages the need for greater understanding between community and curative health teams. The importance of PHC workers to be involved in “community based long term care” for those with NCDs became a priority action (Box A: Strategies 2, 6 and 4). After these broad strategies were agreed upon, fourteen ‘Domains of PHC practice’ were listed which aimed to reflect the intervention space for PHC practice (Table A). These comprised of disease specific components as Vector borne disease control, Oral hygiene and dental care; emergency preparedness, but also on Health promotion and Community participation. Community participation was listed as a discrete domain with “intersectral action in PHC service” also coupled with it. Identification of the specific staff roles and duties required to carry out these activities formed the final component.

Criticism was also made of tokenistic approaches to address NCDs that currently existed in the health system which were not integrated within a PHC approach. An example was

Page | 336

made of the newly recently created Medical officer of NCDs post for each district by the MoH. There remained little clarity on the job description, nor any operational guidance for conducting health promotive and preventative programs. More importantly a lack of meaningful integration of the post to the activities of other primary health care workers and stakeholders led to disenfranchisement in some areas.

A similar ‘fate’ was also realized with the post of the Medical Officer of Mental Health services in Sri Lanka. The post was established at District level order to coordinate community based mental health services and improve community capacity via mental health services. However, due to limited investment in planning and development, lack of coordination structures within PHC teams and with hospital based psychiatric services, most districts inheriting the post were effective in implementing effective community based mental health programming. Most districts opted to train a community based network of volunteers (referred to as “community support officers”) to facilitated referrals from PHC centres. With the exception of two districts, the involvement of PHC workers were marginal. The network of community based mental health workers piloted across 4 district did prove effective in acting as direct conduits of referral from community to the district mental health system and in mental health promotion. However, when external funding for these programs ended the community mental health networks were weakened and dissolved within 6 months. Most participants felt the creation of such programs and posts (labeled as a ‘vertical’ strategy) should be avoided for NCD programming. Careful investment in defining PHC worker roles within the larger intervention space of NCD management were highlighted.

Box A. Proposed strategies formulated through expert group consensus to enhance Primary care worker roles and interventions

Strategy 1- Improve accessibility to appropriate primary level care 1.1.Defining catchment populations to each government primary level facility 1.2.Capacity building of primary level health care teams to address the identified deficiencies in the existing system for addressing the transition challenges 1.3.Improving confidence of patients to access care at the primary level curative institutions 1.4.Improving access to private care for those who can afford by sharing care protocols and use of the individual health record for referral to specialized services- 1.5.Promoting setting up of family medical practices (private providers) in remote areas through a purchasing system 1.6.Further strengthening of the Community Health services provided through field health staff

Strategy 2- Ensuring continuity of care 2.1. Adoption of Family medical approach to care in all primary level curative institutions 2.2. Use of appropriate information tools such as the ‘individual health record’ and other institutional records 2.3. Ensuring appropriate referral and back referral system (The individual health record will support this activity ) 2.4. MOH field staff to follow up on compliance to chronic NCD management during home visits

Strategy 3 – Being responsive to individual and community health needs 3.1. Adoption of family medical approach to address individuals health needs 3.2. Addressing individual needs will be through the standard clinical protocols. Promoting self management through standard health promotion messages addressed at individual or group level will also be important 3.2. Improving skills of primary level curative teams to understand and address community health needs. The community emphasis in curative institutions will vary depending on the community health needs. There Page | 337

may be acute conditions such in outbreaks of communicable diseases as well as chronic / endemic problems of the area that needs to be addressed. 3.3. Improving coordination and identifying ways of working together between the curative and community health services

Strategy 4 – De-institutionalizing primary care and promoting community care 4.1. Promoting community help groups 4.2. Fostering Community level organizations to take lead role in providing care 4.3. Identifying and implementing health care activities that can be carried out in a community setting (will differ from area to area)

Strategy 5 – Fostering other sector involvement 5.1. Capacity building of primary level healthcare teams to engage other sectors involvement in health care and health development. 5.2. Fostering joint review activities for health with other sectors 5.3. Improving health infrastructure to provide an environment to work with other sectors

Strategy 6 – Accommodating Home based care 6.1. Human resource development to foster home based care

Strategy 7 – Supporting Self management (health literacy) 7.1. Identify opportunities to foster self management and incorporate same in to individual patient management plans 7.2. Identify opportunities to use wider communication practice to harness health literacy

Strategy 8 – Continuous quality improvement and patient safety 8.1. Adoption of nationally recommended quality improvement practices 8.2. Adoption of clinical and management audits 8.3. Improve supervision using nationally guided supervision tools to be used at primary level institutions 8.4. Conduction of quality and patient safety reviews using identified set of Indicators

Table A. The proposed strategies formulated through expert group consensus to enhance PHC worker roles and interventions are also listed.

Consensus areas from 3 Rounds of Expert Meetings PHC Strategies Domain of PHC* Activities Responsible PHC worker category* Strategy 1- Improve accessibility to Community  Improving skills of PHC team at appropriate primary level care Participation and primary level MOH level: Intersectoral action in curative teams to Strategy 2- Ensuring continuity of PHC service delivery. understand and care address community PHI =public health needs. health Strategy 3 – Being responsive to  Behavior change inspector individual and community health communication at needs through community PHC level PHM= Public participation  Improving health coordination and midwives Strategy 4 – De-institutionalizing identifying ways of primary care and promoting working together MOH community care between the curative and community health =Medical Strategy 5 – Fostering inter-sectoral services officer of involvement in health Health  community participation Strategy 6 – Accommodating Home PHNS = Public based care emphasis service health nurse delivery of curative Strategy 7 – Supporting Self care institutions SPHM = management (health literacy) Supervising Page | 338

Strategy 8 – Continuous quality PHM improvement and patient safety SPHI = Supervising PHI

HVs = health volunteers

*14 domains were formulated within the 8 strategies’ including : Communicable diseases , Vector borne Diseases; Oral hygiene and dental care.

Debates on new PHC model

A new model of PHC service delivery was proposed by the Directorate of Policy and Development which emphasized rationalizing outpatient and community care through better engagement of preventative (MOH units) with curative arms (District and Base Hospitals) better to serve communities. Whilst there was support for revitalizing the role of PHC system and existing work practices, the drastic structural reforms outlined in this proposed model was met with much resistance. Whilst the model did not achieve consensus support, pilot schemes for such a model were supported by attending experts.

In the ‘new’ model, curative institutions will defined catchment populations, similar to the preventative arm. All primary level curative institutions (Divisional level and primary medical care units) will have their outpatient departments providing family medicine practices (a catchment population of 10-12,000 will be looked after by 3 MOH units). Essential medicines and basic laboratory tests and will be made available and standard care protocols will be used for management of chronic conditions at the Divisional (village cluster) level. A personal health record will be introduced beginning with to those above 30 yrs and gradually to all citizens. A protocol driven referral and back referral system will be instated (although unclear as to how private sector is involved). For in patients the primary medical care unit will decide if the in-patient admission requires primary care or specialized care. Accordingly the referral will be made to the higher level or to the closest divisional hospital. Duty lists of all PHC workers will be modified to this ‘family medicine’ model. A new curriculum, service training, basic training, professional development training for all PHC workers will also be instated. A schematic diagram depicting the enhanced scope of PHC service coverage is show in Figure B.

Page | 339

Figure B. The proposed model for PHC reform aimed at expanding community service coverage. Adapted from MoH (2010) [20].

20 Policy Analysis & Development Unit (2010) Presentation, National scientific & Policy Forum. August, 2010 Page | 340