i Assessing the fidelity and determinants of implementation: A case of active TB case-finding and IPT initiation among PLHIV attending HIV clinics in Ghana By Solomon Ayertey Narh-Bana Supervisors: Prof Tobias F. Chirwa Dr Mary Kawonga A THESIS Submitted to the Faculty of Health Sciences, University of the Witwatersrand, Johannesburg in fulfilment of the requirements for the degree of Doctor of Philosophy Johannesburg, July 2022 ii Declaration I, Solomon Ayertey Narh-Bana, having read and understood what the school defined as plagiarism, do solemnly state that this thesis is my work and that I have accredited contributions, outcomes and citations from unpublished and published work of others. This thesis is being submitted for the degree of Doctor of Philosophy in Epidemiology at the University of the Witwatersrand, Johannesburg. This research has not been presented at any other university or used elsewhere. Signature: Date: 08 – 07 - 2022 iii Dedication I dedicate this piece of work to my mother, Comfort Maku Tetteh, and the memory of my late father, Nene Awate Bana Atrokpa I, who both contributed richly to my education and instilled in me the spirit of perseverance. Also, to my wife Diana Esi Narh-Bana for your support, patience and love that saw me through. Finally, to my kids Joel, Eliana and Odette for their support and enduring my continual absence from home during my study. God, I am thankful to You. iv Original Papers Arising from this PhD Thesis Four original papers were written as part of this PhD thesis. Out of the four papers, two are currently published, while the other two have been submitted and under review. The first two papers assessed fidelity at provider and facility levels and their moderating factors. The third paper explored barriers and facilitators of implementation. The fourth paper aimed to determine fidelity as a combination of facility and provider level activities of the intervention. The titles of the four articles are listed below and their contents are presented as results in chapters 4 to 7. The original papers have been attached as appendixes A to D. 1. Narh-Bana SA, Kawonga M, Chirwa ED, Ibisomi L, Bonsu F, Chirwa TF (2021) Fidelity of implementation of TB screening guidelines by health providers at selected HIV clinics in Ghana. PLoS ONE 16(9): e0257486. https://doi.org/10.1371/journal.pone.0257486 Candidate’s role: conceptualization, tool development, data collection and analysis, writing the first and subsequent drafts of the paper in light of co-author inputs, and response to reviewers and editor. 2. Narh-Bana SA, Chirwa TF, Chirwa ED, Bonsu F, Ibisomi L, Kawonga M (2021) Adherence of HIV clinics to guidelines for the delivery of TB screening among people living with HIV/AIDS in Ghana. BMC Health Services Research 21:1110 https://doi.org/10.1186/s12913-021-07121-9 Candidate’s role: conceptualization, tool development, data collection and analysis, writing the first and subsequent drafts of the paper in light of co-author inputs, and response to reviewers and editor. 3. Solomon A. Narh-Bana, Mary Kawonga, Selase Adjoa Odopey, Frank Bonsu, Latifat Ibisomi, and Tobias F. Chirwa. Factors influencing the implementation of TB screening among PLHIV in selected HIV clinics in Ghana: a qualitative study (Manuscript accepted for publishing by BMC Health Services Research) https://doi.org/10.1371/journal.pone.0257486 https://doi.org/10.1186/s12913-021-07121-9 v Candidate’s role: conceptualization, tool development, data collection and analysis, and manuscript drafting. 4. Narh-Bana SA, Chirwa TF, Chirwa ED, Odopey SA, Bonsu F, Ibisomi L, Kawonga M. Weighted fidelity of delivery of an intervention in the health facility: a case of TB screening among PLHIV in selected hospitals in Ghana (Manuscript submitted – pre- printed and under review) Candidate’s role: conceptualization, development of tools, data collection and analysis, and manuscript drafting. vi Abstract Background. The World Health Organization (WHO) 2021 global TB report, until the COVID-19 pandemic, reported that TB was above HIV/AIDS as the leading cause of death from a single agent. TB is known to account for about 33% of all deaths associated with HIV. The risk of contracting TB is approximately 10-37 folds higher among PLHIV. Ghana, faced with the TB and HIV epidemic, implemented the WHO’s TB/HIV collaborative policy guidelines to lessen the burden of TB among PLHIV. Active TB case-finding through TB screening and Isoniazid Preventive Therapy (IPT) initiation among PLHIV is one essential intervention in the policy. Since the implementation of the intervention over two decades, screening targets have not been achieved. Information on the coverage of TB screening among PLHIV HIV visiting HIV clinics in Ghana is available. This information indicated that, despite the impact of TB in HIV- positive people, TB screening coverage among HIV clients has been low and fluctuated a few years after the implementation of the TB screening intervention among HIV clients. But there is no information on fidelity and determinants of implementing the intervention of TB screening among PLHIV, which is a crucial factor in determining the success of implementation. If available, information on implementation outcomes such as fidelity (adherence to guidelines), and implementation barriers can help identify the bottlenecks to successful implementation, and inform strategies for improving implementation processes and outcomes to achieve successful intervention outcomes. Therefore, this PhD thesis aims to 1) assess the fidelity of the active TB case-finding through TB screening IPT initiation intervention among PLHIV attending HIV clinics in Ghana, and 2) explore implementation determinants of the TB screening intervention among PLHIV attending HIV clinics in Ghana. This is to provide context-relevant evidence that could inform the design of appropriate implementation strategies for supporting or strengthening implementation efforts in Ghana. The specific objectives included: determining the provider level fidelity to clinical protocols of the intervention; evaluating the influence of moderating factors on provider fidelity of the intervention; determining the facility level fidelity to the intervention; comparing the moderating factors by facility level; determine an overall implementation fidelity using a composite of provider and facility level guidelines of the intervention, and exploring the determinants of implementation of the intervention among PLHIV attending HIV care clinics in Ghana. vii Methods. The research was cross-sectional and explorative, conducted in 27 selected district hospitals with HIV or ART centres in Ghana. These 27 hospitals were purposively selected because, in the year 2018, they were identified to have started the initiation of IPT among HIV clients who have been confirmed of not having TB in Ghana. Fidelity was assessed at the provider, facility and combined levels. Items in the guideline related to the providers were grouped into three components (TB diagnosis – 10, TB awareness – 4, TB symptoms questionnaire – 2 items). At the facility level, the items were grouped into four components - intensive TB case-finding among PLHIV (ITCF) 5 items, IPT initiation (IPT) 3 items, TB infection control (TIC) 5 items and programme review meeting (PRM) 5 items. A summation of items score was used to determine fidelity scores. A computed median fidelity score was used in this study as the cut-off point due to lack of reference for determining what constituted a high-level fidelity or a low-level fidelity. Opportunity-based weighting approach was used to assess the combined fidelity. In addition, we conducted Focus Group Discussions (FGD) and Indepth Interviews (IDIs) to explore the determinants of implementation of the intervention. Results. A total of 226 HIV healthcare providers and 27 facility managers were interviewed. Also, 8 FGDs with HIV healthcare providers and 17 IDIs with TB/HIV coordinators were conducted. At the provider level, 60% (135) of the HIV healthcare provider were females, and their mean age was 34.5 years (SD=8.3). Majority were clinicians [63% (142)] and had obtained post- secondary non-tertiary educational level [62% (141)]. The proportion of the healthcare providers categorized to have implemented the intervention with high fidelity was 53% (119). Also, 56% (126), 53% (120), and 59% (134) of them implemented the TB diagnosis, TB awareness and TB symptoms questionnaire components, respectively, with high fidelity., Female healthcare providers (AOR=2.36, 95%CI: 1.09–5.10, p=<0.029), those with tertiary education (AOR=4.31, 95%CI: 2.12–9.10, p=0.040), and clinicians (AOR=1.78, 95%CI: 1.07– 3.50, p=0.045) were more likely to adhere to the guidelines compared to their counterparts, after adjusting for the clinic as cluster effect. At the facility level, the median adherence score for ITCF component was 85.7% (IQR: 85.5- 100.0), for IPT was 0% (IQR: 0-66.7), for TIC was 33.3% (IQR: 33.3-50.0), and for PRM components it was 90.0% (IQR: 70.0-90.0. Overall, the median adherence score for the facility was 62.1% (IQR: 58.6-65.1), with 17 clinics (63%) scoring above the median, categorized as viii high adherence. High adherence facilities had statistically significant lower PLHIV clinic attendees per month (256 (IQR: 60-904) than low adherence facilities 900 (IQR: 609-2622); p=0.042), and lower HIV provider workloads (28.6 (IQR: 8.6-113) than low adherence facilities 90 (IQR: 66.7-263.5); p=0.046). The high adherence facilities had most of the screening guidelines (76%, p<0.01) and questionnaire (80%, p<0.01) available on-site. Also, at the combined level, the weighted fidelity median score was 67% (IQR: 59.9 – 74.9%), and the TB screening coverage was 71.3% (IQR: 56.9 – 96.7). The weighted fidelity of delivery is statistically associated with TB screening coverage (p<0.01). Facilities with TB IE&C materials available had significantly (p=0.025) higher median fidelity score (75.4% (74.9 – 88.5)) than their counterparts (65.7 (59.4 – 72.6)). All the other moderating factors investigated have no statistical association with weighted fidelity of delivery (p>0.05) except IE&C. Participants in the qualitative study (phase 2) recognised TB screening as an essential intervention to decrease TB's burden among PLHIV. In addition, ease in TB screening for PLHIV, existing good communication and referral systems, effective goals setting and feedback mechanisms, good recognition for the need of the intervention by HIV healthcare providers, and the role played by existing chemical sellers were indicated by the HIV healthcare workers and coordinators as the main facilitators that promote the implementation of the TB screening among PLHIV. The study also identified factors such as high workload, no transportation, nature of the disease, non-reimbursement of ancillary cost, lack of staff, no staff motivation, no in-service training, old infrastructures, lack of privacy and space, inadequate resources (laboratory), non- availability of enablers for clients and health works were also identified by the participant main barriers hindering the TB screening implementation among PLHIV attending HV clinics. Conclusion. The study assessed fidelity of delivery at the different levels of health delivery, their related factors and determinants of implementation of TB screening at the HIV or ART clinics. Combined provider-facility fidelity was revealed as a practical approach to assessing implementation fidelity at the health facility. The separate fidelities showed the extent of implementation of the intervention guidelines as planned at the various levels. Hence, variations observed in the fidelity levels and related factors needed further investigation. The findings are critical in the wake of the rising public health importance of the double burden of TB and HIV in low- to middle-income countries like Ghana. ix Implementation of TB screening is critical in PLHIV in Ghana. Programme managers, facility managers and TB/HIV coordinators should be devoted to ensuring HIV healthcare workers have regular in-service training, motivated through reimbursement of intervention-related costs and the provision of adequate space and privacy to ensure successful implementation of TB screening amongst HIV clients. Participants, especially HIV healthcare providers have accepted the intervention and therefore used the ease of the TB screening tool, feedback and supervision and their good network and communication to promote the implementation of the TB screening intervention among PLHIV attending HIV clinics. Hence, knowledge of fidelity and both facilitators and barriers of implementation, and how to address them can inform the planning and implementation of the TB screening in HIV clinics and ART centres in Ghana and similar contexts across low- and middle-income countries worldwide. Key Contribution. Most fidelity assessments were done at the provider level, however, this PhD has added fidelity assessment at the programme or facility level and especially combined facility-provider level which is missing in the implementation literature. Hence, it is an extension of knowledge and methodological innovation in implementation science. In addition, this study has demonstrated how to categorise fidelity into high or low fidelity. This research reveals where efforts can be targeted to improve implementation fidelity to increase and achieve targets of TB screening coverage among PLHIV in Ghana. In addition, it showed the barriers to be addressed by the National TB Control Programme to ensure successful implementation of the intervention in Ghana. Keywords: TB, Facility adherence, provider adherence, TB screening factors, IPT, implementation determinants, PLHIV, HIV clinics, composite, weighted fidelity, implementation fidelity levels, healthcare providers, implementation science, Ghana, sub- Saharan Africa x Acknowledgements ".... Be thankful to Him, And bless His Name. For the Lord is good;" Psalms 100: 4 - 5 I hereby show my appreciation to my Supervisors, Professor Tobias F Chirwa, Dr Mary Kawonga and Prof Frank Bonsu, for their unending support, inspiration and supervision during the period I embarked on my PhD and most importantly for their directions and contributions to the different phases of this research work throughout. Special gratitude to Dr (Mrs) Esnat Chirwa for her helpful advice and constructive suggestions. I chiefly appreciate her taking the time out of her busy schedules to review most of the draft manuscripts to refine them for publication. Dr (Mrs) Chirwa, Thank you. Mrs Selase Adjoa Odopey provided methodological support towards the qualitative research of this study and her constant inspiration in the Lord whenever the going got tough. Prof Latifat Ibisomi offered invaluable guidance, for which I am exceedingly grateful. Also, I’m thankful to Professors Margaret Gyapong, Evelyn Ansah and Irene Agyepong for their technical support. I also appreciate mightily the data management support provided by Gabriel Odonkor. I appreciate the Regional and District Directors of Ghana Health Service and the research participants. They accepted and participated in the study and cooperated with us throughout the study period. In addition, my research assistants were superb; guys, thank you. My PhD and this research would not have been possible without total funding from the TDR, the Special Programme for Research and Training in Tropical Diseases, hosted at the World Health Organization and co-sponsored by UNICEF, UNDP, the World Bank, and WHO. Thank you to Prof Jude Igumbe and Mr. Paul Bohloko for all the technical and administrative support. Now that I mentioned administrative support, I can’t forget Ms Gloria Bowes and Ms Nomonde Malahlela. Finally, I thank all the MSc Implementation Science students (2018/19 batch), PhD students and academic staff – too many to mention names – at the Wits School of Public Health. To finish, I thank my family for all the fantastic support – my wife Diana and our children Joel, Eliana and Odette for enduring my long absences from home. Thank you! xi Table of Contents DECLARATION ............................................................................................................................................ II DEDICATION ...............................................................................................................................................III ORIGINAL PAPERS ARISING FROM THIS PHD THESIS .................................................................... IV ABSTRACT .................................................................................................................................................. VI ACKNOWLEDGEMENTS ........................................................................................................................... X TABLE OF CONTENTS ............................................................................................................................. XI LIST OF FIGURES .................................................................................................................................... XIV LIST OF TABLES....................................................................................................................................... XV LIST OF ABBREVIATIONS/ACRONYMS ............................................................................................. XVI DEFINITIONS OF TERMS ...................................................................................................................... XVII 1.0 INTRODUCTION ............................................................................................................................. 1 1.1 THESIS OVERVIEW AND STRUCTURE ............................................................................................................... 1 1.2 BACKGROUND ........................................................................................................................................... 3 1.3 RATIONALE OF THE STUDY ................................................................................................................................... 5 1.4 AIM AND OBJECTIVES OF THE STUDY ............................................................................................................... 7 1.4.1 General aim of the study ...................................................................................................................... 7 1.4.2 Specific objectives ................................................................................................................................ 7 1.5 THESIS FRAMEWORK ................................................................................................................................... 8 2.0 LITERATURE REVIEW ................................................................................................................ 10 2.1 THE ECONOMY AND HEALTH SERVICE STRUCTURE OF GHANA .................................................................................... 10 2.1.1 Demographic and socio-economic information ................................................................................ 10 2.1.2 The health service structure of Ghana .............................................................................................. 12 2.2 THE GLOBAL BURDEN OF TB AMONG PLHIV ......................................................................................................... 13 2.3 THE BURDEN OF TB AMONG PLHIV IN SUB-SAHARAN AFRICA .................................................................................. 15 2.4 TUBERCULOSIS IN GHANA ................................................................................................................................. 17 2.4.1 The burden of TB and TB/HIV co-infection in Ghana .................................................................... 17 2.4.2 Tuberculosis and HIV collaborative activities in Ghana .................................................................. 18 2.4.3 Funding the needs of TB control ....................................................................................................... 22 2.5 IMPLEMENTATION FIDELITY ASSESSMENT OF HEALTH INTERVENTIONS ......................................................................... 23 2.6 FIDELITY LEVELS OF AN INTERVENTION.................................................................................................................. 29 2.7 MODERATING FACTORS OF IMPLEMENTATION FIDELITY ........................................................................................... 29 2.8 IMPLEMENTATION DETERMINANTS OF HEALTH INTERVENTIONS ................................................................................. 31 2.9 IMPLEMENTATION DETERMINANTS OF TB SCREENING AMONG PLHIV ........................................................................ 32 2.9.1 Outer setting ....................................................................................................................................... 33 2.9.2 Inner setting ....................................................................................................................................... 33 2.9.3 Intervention characteristics ............................................................................................................... 34 2.9.4 Individual Characteristics .................................................................................................................. 34 2.10 CONCEPTUAL FRAMEWORK OF THE STUDY .......................................................................................................... 35 3.0 METHODOLOGY .......................................................................................................................... 37 3.1 STUDY SETTING ............................................................................................................................................... 37 3.2 STUDY DESIGN ................................................................................................................................................ 40 3.3 STUDY POPULATION ......................................................................................................................................... 41 3.4 INCLUSION AND EXCLUSION CRITERIA ................................................................................................................... 42 3.5 SAMPLE SIZE AND SAMPLING PROCEDURE ............................................................................................................. 42 3.6 DATA COLLECTION ........................................................................................................................................... 44 xii 3.6.1 Data collection for phase 1 study ....................................................................................................... 46 3.6.2 Data collection for phase 2 study ....................................................................................................... 48 3.7 DATA MANAGEMENT, MEASUREMENT AND ANALYSIS.............................................................................................. 48 3.7.1 Measurement and analysis of quantitative data ................................................................................ 52 3.7.1.1 Provider implementation fidelity analysis ................................................................................................. 52 3.7.1.2 Programme/facility implementation fidelity analysis ............................................................................... 54 3.7.1.3 Combined implementation fidelity of the delivery analysis ..................................................................... 55 3.7.1.4 Moderating factors of implementation fidelity analysis ........................................................................... 56 3.7.1.5 Statistical tests.............................................................................................................................................. 56 3.7.2 Analysis of qualitative data ................................................................................................................ 56 3.8 QUALITY CONTROL ........................................................................................................................................... 56 3.9 ETHICAL CONSIDERATIONS ................................................................................................................................. 57 3.9.1 Ethical approval ................................................................................................................................. 57 3.9.2 Informed consent ............................................................................................................................... 58 3.9.3 Confidentiality .................................................................................................................................... 58 3.10 PROTOCOL AMENDMENTS ............................................................................................................................... 58 4.0 PROVIDER-LEVEL IMPLEMENTATION FIDELITY TO TB SCREENING GUIDELINES .. 59 4.1 ABSTRACT ...................................................................................................................................................... 59 4.2 INTRODUCTION ............................................................................................................................................... 60 4.3 METHODS ...................................................................................................................................................... 65 4.4 RESULTS ........................................................................................................................................................ 73 4.5 DISCUSSION.................................................................................................................................................... 80 4.6 CONCLUSION .................................................................................................................................................. 83 REFERENCE .......................................................................................................................................................... 84 5.0 FACILITY-/PROGRAMME-LEVEL IMPLEMENTATION FIDELITY TO TB SCREENING GUIDELINES .............................................................................................................................................. 90 5.1 ABSTRACT ...................................................................................................................................................... 90 5.2 INTRODUCTION ............................................................................................................................................... 91 5.3 METHODS ...................................................................................................................................................... 93 5.4 RESULTS ...................................................................................................................................................... 100 5.5 DISCUSSION.................................................................................................................................................. 106 5.6 CONCLUSION ................................................................................................................................................ 109 REFERENCE ........................................................................................................................................................ 110 6.0 DETERMINANTS OF IMPLEMENTATION OF TB SCREENING INTERVENTION ........... 116 6.1 ABSTRACT .................................................................................................................................................... 116 6.2 INTRODUCTION ............................................................................................................................................. 117 6.3 METHODS .................................................................................................................................................... 120 6.4 RESULTS ...................................................................................................................................................... 123 6.5 DISCUSSION.................................................................................................................................................. 132 6.6 CONCLUSION ................................................................................................................................................ 134 REFERENCES ....................................................................................................................................................... 135 7.0 COMBINED FACILITY-PROVIDER LEVEL FIDELITY OF IMPLEMENTATION OF THE INTERVENTION ....................................................................................................................................... 141 7.1 ABSTRACT .................................................................................................................................................... 141 7.2 INTRODUCTION ............................................................................................................................................. 143 7.3 METHODS .................................................................................................................................................... 146 7.4 RESULTS ...................................................................................................................................................... 151 7.5 DISCUSSION.................................................................................................................................................. 155 7.6 CONCLUSIONS ............................................................................................................................................... 158 REFERENCES ....................................................................................................................................................... 158 xiii 8.0 GENERAL DISCUSSION AND CONCLUSION ......................................................................... 161 8.1 OVERVIEW ................................................................................................................................................... 161 8.2 GENERAL DISCUSSION ..................................................................................................................................... 161 8.2.1 Contextual conduct of the study ...................................................................................................... 161 8.2.3 Provider Level Fidelity ..................................................................................................................... 162 8.2.4 Programme/Facility Level Fidelity .................................................................................................. 163 8.2.5 Combined provider and programme fidelity scores ........................................................................ 165 8.2.6 Moderating factors of the implementation fidelity .......................................................................... 166 8.2.7 Determinants of implementation ..................................................................................................... 167 8.2.8 Linkage between findings and conceptual framework ................................................................... 169 8.2.9 Limitations of the study .................................................................................................................... 169 8.2.10 Strengths of the study ..................................................................................................................... 170 8.2.11 Implication for research ................................................................................................................ 170 8.2.12 Implication for policy and practice ................................................................................................ 171 8.2.12 Contribution of PhD ...................................................................................................................... 172 8.3 GENERAL CONCLUSION ................................................................................................................................... 173 REFERENCES ........................................................................................................................................... 175 APPENDICES ............................................................................................................................................ 187 APPENDIX A: ORIGINAL PAPER 1 ............................................................................................................................ 187 APPENDIX B: ORIGINAL PAPER 2 ............................................................................................................................ 209 APPENDIX C: ORIGINAL PAPER 3 ............................................................................................................................ 223 APPENDIX D: ORIGINAL PAPER 4 ............................................................................................................................ 246 APPENDIX E ETHICS CLEARANCE CERTIFICATES ........................................................................................................... 268 APPENDIX F: PERMISSION LETTERS.......................................................................................................................... 271 APPENDIX G: DATA COLLECTION TOOL FOR ASSESSING PROVIDER-LEVEL IMPLEMENTATION FIDELITY .................................... 281 APPENDIX H: DATA COLLECTION TOOL FOR ASSESSING FACILITY-LEVEL IMPLEMENTATION FIDELITY ....................................... 284 APPENDIX I: DATA EXTRACTION FORM ..................................................................................................................... 288 APPENDIX J: IN-DEPTH INTERVIEW GUIDE FOR EXPLORING IMPLEMENTATION DETERMINANTS ............................................ 290 APPENDIX K: FOCUS GROUP DISCUSSION GUIDE FOR EXPLORING IMPLEMENTATION DETERMINANTS ................................... 294 APPENDIX L: INFORMATION SHEET AND CONSENT FORMS ............................................................................................ 298 APPENDIX M: PLAGIARSM CHECKER – TURNITIN REPORT ............................................................................................ 308 xiv List of Figures Chapter 1 Figure 1: Thesis framework linking the research objectives to the papers ................................ 9 Chapter 2 Figure 2: Map of Ghana showing the 10 regions and 3 ecological belts (32) ......................... 11 Figure 3: Revised screening algorithm (source: NACP) ......................................................... 21 Figure 4: Conceptual framework for the study ........................................................................ 35 Chapter 3 Figure 5: Map of Ghana showing study sites........................................................................... 37 Figure 6: Phases of the study ................................................................................................... 40 Figure 7: Flowchart of recruitment and study profile of the healthcare providers and facility representatives.......................................................................................................................... 47 Figure. 8: TB screening activities and scores by components to asses programme implementation fidelity ............................................................................................................ 55 Chapter 4 Figure 9: A conceptual framework for provider level IF to TB screening at HIV clinics in Ghana: adapted from CFIF by Carroll et al, 2007 ................................................................... 64 Figure 10: A map of Ghana showing the 10 regions and the study sites ................................. 66 Figure 11: A typical structure of district hospital depicting where facility and provider levels belonged ................................................................................................................................... 67 Figure 12: Flowchart of recruitment and study profile of the healthcare providers at the HIV clinics ....................................................................................................................................... 70 Chapter 5 Figure 13: Boxplot showing the distribution of adherence scores of the TB screening intervention components ........................................................................................................ 103 Figure 14: Graphical representation of the facility adherence score (%) to TB screening guidelines and activities at 27 HIV clinics ............................................................................ 104 Chapter 7 Figure 15: Sample weighted implementation fidelity score calculation for one health facility ................................................................................................................................................ 150 Figure 16: Weighted fidelity score and TB screening coverage in selected HIV clinics in Ghana ..................................................................................................................................... 155 xv List of Tables Chapter 2 Table 1: Fidelity Criteria: Process of development, measurement and validation .................. 27 Chapter 3 Table 2: Basic characteristics of the district facility where the study was conducted ............. 38 Table 3: Distribution of interviews and discussions conducted............................................... 43 Table 4: Summary of the expected and actual number of interviews conducted .................... 45 Table 5: variables measured and approach to analysing various objectives ............................ 50 Table 6: List of the 19 items and their measurement for assessing provider implementation fidelity ...................................................................................................................................... 52 Chapter 4 Table 7: List of the 19 items used to determine provider implementation fidelity score ........ 68 Table 8: Characteristics of the healthcare providers by profession at the HIV clinics in Ghana, 2019.......................................................................................................................................... 73 Table 9: Descriptive statistics of the fidelity assessment items among HIV healthcare providers at the HIV clinics in Ghana...................................................................................... 74 Table 10: Summary statistics of the provider fidelity scores by the components ................... 76 Table 11: Comparison of geographical and provider characteristics between fidelity levels . 77 Table 12: Multivariable logistic regression between implementation fidelity level and background factors with district hospital (random) effect ....................................................... 79 Chapter 5 Table 13: TB screening activities and scores by component and source of information ........ 97 Table 14: Characteristics of the study participants ................................................................ 100 Table 15: Characteristics of the health facilities by ecological zones .................................. 101 Table 16: Association between facility-level factors and programme adherence level ........ 105 Chapter 6 Table 17:Total number of enrolled participants from the Northern, Central and Eastern regions .................................................................................................................................... 124 Table 18: Background characteristics of study participants .................................................. 124 Table 19: Emerging barriers and facilitators linked to CFIR domains .................................. 125 Table 20: Distribution of background factors of the selected HIV clinics ............................ 151 Chapter 7 Table 21: Distribution of weighted fidelity scores by facility characteristics ....................... 152 Table 22: Model statistics assessing the fidelity type in relation to screening coverage ....... 154 xvi List of Abbreviations/Acronyms AIDS - Acquired Immune Deficiency Syndrome ART - Antiretroviral Therapy CFIF - Conceptual Framework for Implementation Fidelity CFIR - Consolidated Framework for Implementation Research CXR - Chest X-ray DHs - District hospitals GHS - Ghana Health Service HIV - Human Immunodeficiency Virus IPT - Isoniazid Preventive Therapy IQR - Interquartile range IS - Implementation Science ITCF - Intensive TB case-finding among PLHIV MOH - Ministry of Health NACP - National AIDS/STI Control Programme NTP - National Tuberculosis Control Programme PCA - Principal component analysis PLHIV - People living with HIV PRM - Programme review meeting TB - Tuberculosis TIC - TB infection control WHO - World Health Organization xvii Definitions of Terms • Implementation: Carrying out an evidence-based intervention under real-life conditions and making it part of the everyday practice • Implementation Outcomes: The effects of deliberate and purposive actions to implement TB screening and initiating IPT in HIV care clinics in Ghana. Among the eight implementation outcomes is fidelity to the intervention guidelines. • Fidelity: The degree or extent to which the TB screening and IPT initiation had been implemented at the facility as prescribed in the NTP guidelines. • Facility (programme) level fidelity: The level at which programme guidelines for implementing the intervention are being implemented by the facility as planned by the NTP. The term programme level and facility level are used interchangeably for the purpose of this study to mean the same. • Provider level fidelity: The level at which clinical protocol guidelines for implementing the intervention are being delivered by HIV healthcare providers as the NTP planned it. • Determinants of implementation: Barriers and facilitators during the implementation of the TB screening at the HIV clinic 1 1.0 Introduction 1.1 Thesis Overview and structure This thesis is submitted using the divided block format. Each paper produced from this study represents a different results chapter. It also includes an Introduction, Literature Review, Methods and Conclusion as separate chapters. The thesis is organized and divided into eight (8) chapters explained below: Chapter 1 Introduction The introduction to this thesis is presented in this chapter. The chapter further presents a brief background to the study, discusses the study's rationale, explains the aims and objectives, and offers a general overview of how the thesis is structured. Chapter 2 Literature review We present under this chapter a brief review of the global burden of TB among PLHIV, followed by the burden in sub-Saharan Africa. The TB situation in Ghana follows, which comprises a brief on Ghana's demographic and socio information, the health structure of Ghana, the burden of TB in Ghana, and the TB/HIV collaboration in Ghana. Next, the assessment of implementation fidelity, fidelity levels and factors influencing implementation of an intervention in health facilities are discussed. The last sub-section discusses the conceptual framework that directed the conduct of the study. Chapter 3 Methods Chapter 3 provides a synthesis of the methods used for the study. Methods discussed under this section include; the design of the study, the setting where the study was conducted, the study population and sampling approaches used. In addition, we discussed briefly the data sources, the study's conduct and procedures, and the data analysis method employed. Detailed methods are described in each of the four papers representing the four different result sections under 2 chapters 4 to 7, of which the PhD student is the first author. This chapter also describes all the ethical issues of the study. Chapter 4 Results – Understanding provider-level fidelity to the intervention The fourth chapter presents the paper produced to address objectives 1 and 2 of the thesis. It evaluates the provider level implementation fidelity to the TB screening intervention and it’s relating factors. Chapter 5 Results – Understanding facility-level fidelity to the intervention This chapter presents the second paper produced from this thesis. It emanated from objectives 3 and 4 of the study, focusing on assessing facility-level fidelity to the implementation guidelines and its related factors in Ghana. Chapter 6 Results – Understanding implementation determinants to the intervention We present the third paper that sought to understand the barriers and facilitators to implementing the intervention. This paper covers objective 6 of the study, and it is purely qualitative. Chapter 7 Results – Understanding combined facility-provider fidelity to the intervention The seventh chapter presents the fourth paper, which addresses the extent to which the facility as a whole implemented the intervention. The paper combined the guidelines for the facility and HIV healthcare providers to ascertain a combined fidelity level of delivery of the intervention. This paper covers objective 5 of the study. Chapter 8 General discussion and conclusion In this ending chapter, the significance of the findings are highlighted considering what is already known and their implications for policy and practice. Additionally, it discusses the 3 contributions of this thesis to literature, outlines limitations of the research, and lessons learnt along with general conclusions drawn from the study. 1.2 Background The Global Tuberculosis Report, 2014 indicated the number of people infected with tuberculosis (TB) was 9 million in 2013, out of which 1.5 million people died; about 24% of this number were people with HIV. Globally, TB, although curable, is known to kill about 500 people daily and is the main cause of morbidity and mortality amongst people living with HIV/AIDS (PLHIV). According to the World Health Organization (WHO) 2017 TB report globally, TB accounted for about one-third of all deaths associated with HIV (1,2). In addition, the risk of contracting TB is approximately 10-37 folds higher among people living with HIV than those without HIV (3–5). In sub-Saharan Africa, the issue of TB and HIV co-infection is of great concern. Of the 22 sub- Saharan African countries that have the burden of TB, nine countries contribute to the highest- burden of TB globally (6). According to the WHO, the high prevalence of HIV and its spread within this region drive the incidence of TB. Also, in 2016 about 34% of PLHIV were diagnosed with TB. In 2014, the WHO reported an estimate of 1.1 million TB patients were co-infected with HIV (7). Most of these sub-Saharan African countries are within the low to middle-income class, facing resource constraints (8) but have to take action to attain a range of TB related Sustainable Development Goal (SDG) targets. For instance, Ghana, one of the middle-income countries, has an HIV prevalence of 1.6% (7) and an estimated 224,488 persons, made up of 34,557 (11.8%) children having HIV and AIDS. The TB and HIV co- infection prevalence in Ghana is at 14.7% (9). Effective TB control strategies are contingent on treating those who tested positive and preventing those who tested negative from contracting the disease. Tuberculosis can best be controlled by early detection through active case finding and appropriate treatment. Active TB case finding among contacts of TB and HIV patients through simple interventions such as outreach TB screening or screening of all HIV patients for TB at HIV care clinics have been used (10). Mhimbira et al. (11) conducted a review to assess the effectiveness of diverse innovations to improve the detection of TB at a significantly lower level of healthcare services. They found that outreach TB screening increased TB case detection from 24% to 52%. 4 However, they deduced that this strategy had little or no effect on treatment compared to screening at HIV care clinics. On the other hand, according to Date et al. (12), TB screening among PLHIV improves the probability of survival, quality of life, and lessens the spread of tuberculosis in an area. It also increases the chances of putting HIV clients who show no active TB on Isoniazid Preventive Treatment (IPT) for latent TB infection (5,6,13). IPT is a preventive therapy given to HIV clients who do not have active TB. Its effect is known to supplement the impact of ART on reducing the infection of TB. In 2004, the WHO instituted a TB/HIV collaborative policy guideline to control the TB/HIV epidemic (1,7). Amongst the goals of the TB/HIV joint activities is to lessen the trouble of TB in PLHIV (14). The intervention activities for this goal were intensified TB case-finding through screening for TB, ensuring that HIV clients who screened negative for TB are initiated on Isoniazid preventive therapy (IPT), and ensuring the control of TB in all health facilities (14). In addition, high-quality anti-tuberculosis treatment is given to PLHIV, who have a confirmed diagnosis of active TB. Many countries where TB/HIV co-infection were of public health concern, including Ghana, adopted and implemented this policy intervention, as recommended by the WHO (15). Despite evidence that this TB/HIV control intervention has the potential of reducing the burden of TB in PLHIV, progress on IPT implementation worldwide has been slow and unable to achieve and sustain desired results (16). Globally, about one-third of HIV positive-tested clients are screened for TB at HIV care clinics (16–18), contrary to the policy, and only 17% of HIV positive adults screened negative for TB are put on IPT (16). Despite the immense challenge posed by the TB/HIV diseases to healthcare systems in sub-Saharan Africa, only 9% of PLHIV had been screened actively for TB, out of which only 3% were put on IPT (19). Therefore, the existence of intervention of known efficacy is not enough to ensure their implementation. Implementation is a way of carrying out an evidence-based intervention under real-life conditions and making it part of everyday practise (20,21). Therefore, implementation is critical in achieving and sustaining the desired result of health interventions. The definition of implementation science (IS) by the National Institutes of Health (NIH), as “the study of methods to promote the integration of research findings and evidence into healthcare policy and practice”, makes an essential contribution to understanding implementation processes and barriers, and developing strategies to improve implementation 5 outcomes (21). Various implementation outcomes (or implementation quality constructs) are known to influence the effectiveness of an intervention (22). Implementation of intervention with high quality is expected to yield the desired programme and clinical outcomes (16). Implementation outcomes commonly discussed in the IS literature include, fidelity (whether the implementation was implemented as planned), reach (whether it reached the intended audience) and acceptability (whether it was acceptable to implementers and recipients) (21) 1.3 Rationale of the study Ghana's TB/HIV policy goals were to reinforce the health system to act in response to the TB and HIV epidemic, lessen the catastrophe of TB amid PLHIV and lessen the catastrophe of HIV among TB patients. Each of the goals has specific collaborative activities. For instance, to reduce or lessen the burden of TB in PLHIV, joint activities including TB prevention and infection control among PLHIV, timely diagnosis and treatment of HIV clients among TB patients, and prevention of TB disease in HIV clients were in place. Effective TB control strategies for minimising the burden of TB are through early detection, then prevention of TB among PLHIV. The study's intervention of interest and focus was the TB burden reduction through TB screening in PLHIV visiting HIV or ART clinics in Ghana. Ghana initially implemented the intervention to lessen the burden of TB in PLHIV over a decade ago, and the IPT component was supposed to have been added in 2018 (23). A review of literature on program indicators for Ghana shows that during the early stages from 2008 to 2012 of the implementation of the TB screening intervention, a high proportion of PLHIV (70% – 96%) were screened for TB at HIV care (24). Unfortunately, this observed success has not been sustained as programme data further revealed that in the years 2013 and 2014, approximately 80% (24) and 41% (25) of PLHIV, respectively, were not screened for TB as part of care contrary to the TB/HIV collaborative guidelines. In 2014, the policy, as a result, was revised with different TB screening coverage goals set at 56% for 2015, 64% for 2016, 70% for 2017, 80% for 2018, 85% for 2019, and increased to 90% in 2020 (24). The reason for such slow and fluctuating progress is unknown since no research has been conducted to that effect. Achieving the set targets will require full implementation of the intervention. If the intervention is implemented partially, it might affect the set targets and have a limited impact on curbing 6 the dual TB and HIV epidemic (16). It has been shown that the degree of how well an intervention is implemented can determine its potential impact (26). Furthermore, digression from the implementation policy may be counter-productive and lead to a waste of resources to the country. Although the information on the TB screening coverage among HIV clients visiting the ART clinics in Ghana is available, information on how well the intervention is being implemented and the determinants of implementation are not readily available in Ghana. The availability of such information will be useful in identifying and addressing gaps and bottlenecks to improve implementation outcomes in later years to achieve set targets of the intervention outcomes. A review by Proctor et al. (2010) set clarity on the taxonomy of the eight implementation outcomes and their measurement and the salient stage at which each implementation outcome could be assessed (21). Most of the intervention components of interest in this study have been implemented for over a decade now. Therefore, the choice of evaluating fidelity for this study is appropriate because it agrees with what Proctor et al. proposed as an implementation outcome to be assessed at mid to later stages of the intervention (21). Fidelity to an intervention and factors that influence implementation is essential and critical implementation issues worth considering, especially when the intervention is not yielding expected results (27–29). Also, the most significant gap in the IS literature is how we measure fidelity, factors that influence fidelity in different contexts, and the mechanisms through which these factors influence fidelity. To address some of the gaps in IS, there is the need to assess how well the TB prevention intervention is being implemented and investigate the determinants of implementation to establish the barriers and facilitators of the implementation. The implementation outcome to be researched is fidelity at the provider and the facility levels. To minimize the problem of TB amongst PLHIV, the operational strategies of the TB/HIV collaboration programme need to be implemented at HIV care clinics according to the policy guidelines. Likewise, the providers have a role in implementing the interventions as intended in the guidelines—the healthcare providers work in health facilities. The health facilities are the district hospitals within which the HIV care clinics are located. However, the literature on fidelity often describes either provider or programme (facility) level fidelity separately. However, this PhD study argued and concentrated on the complementary roles of the health facility and the providers in implementing the intervention that can influence the outcomes. In the health system, the facility level is where intervention programmes are operationalized. The 7 intervention at this level is implemented by the facility operational managers and respective healthcare workers. The facility is therefore a good representative of the programme activities. Information on how well the facility and providers undertake their roles is crucial, but such is not available. WHO has recommended TB screening among PLHIV as a critical component of the intervention package of HIV care due to the high susceptibility of the PLHIV to TB and the spillover effect on the non-HIV families and community members. The study, therefore, aimed to assess implementation fidelity and implementation determinants of TB screening amongst PLHIV visiting the HIV clinics in Ghana. Combined fidelity was evaluated at both provider and programme (facility) levels for 27 district hospitals across Ghana. In addition, barriers and facilitators of the intervention to improving TB/HIV programme performance and reducing the problem of TB in PLHIV in Ghana were explored. The study makes a methodological contribution to measuring fidelity in health at a health facility where intervention is implemented. It also provided helpful information to link TB programme outcomes, such as TB screening coverage and the proportion initiated on IPT during the intervention (20,30). 1.4 Aim and objectives of the study 1.4.1 General aim of the study To assess the fidelity and determinants of the implementation of the TB screening intervention among PLHIV visiting HIV and ART care clinics in Ghana. The TB screening is the key intervention activity for the active TB case-finding and IPT initiation policy. In addition, the TB screening among PLHIV is a critical intervention component of the HIV care package due to its potential impact on reducing the TB burden. 1.4.2 Specific objectives 1. To determine the provider level fidelity to clinical protocols of the intervention among PLHIV attending HIV care clinics in Ghana. 8 2. To evaluate the influence of moderating factors on provider fidelity of the intervention among PLHIV attending HIV care clinics in Ghana. 3. To determine the facility level fidelity to the intervention guidelines among PLHIV attending HIV care clinics in Ghana. 4. To compare the moderating factors by facility level fidelity of the intervention among PLHIV attending HIV care clinics in Ghana. 5. To determine an overall implementation fidelity using a composite of provider and facility level guidelines of the intervention among PLHIV attending HIV care clinics in Ghana. 6. To explore the determinants of implementation of the intervention among PLHIV attending HIV care clinics in Ghana. 1.5 Thesis framework This PhD research includes both provider and facility level analyses, as shown in Figure 1 below. The providers implement an intervention within health facilities; however, interventions in the health facilities are executed from the facility levels cascading down to the provider level. Therefore, guidelines of the implementation of the intervention operate hand in hand to achieve results. 9 This section ends chapter 1 of the thesis. The next chapter offers a well-organised review of existing literature mostly related to the focus of the study. Implementation determinants Understanding the determinants of implementation Programme (facility) Level Understanding the adherence to implementation guidelines Provider Level Understanding the extent of implementation of guidelines Facility factors Moderating factors responsible for facility level adherence Provider factors Moderating factors responsible for provider level fidelity Composite fidelity of implementation Understanding adherence at both provider and facility level Understanding the determinants of implementation Both level analysis AIM 2: Objective 6 (Paper 3) Understanding facility adherence Facility level analysis AIM 1: Objective 3 & 4 (Paper 2) Understanding provider fidelity Provider level analysis AIM 1: Objective 1 & 2 (Paper 1) Methodological Innovation Both level analysis AIM 1: Objective 5 (Paper 4) Figure 1: Thesis framework linking the research objectives to the papers 10 2.0 Literature Review This chapter contains seven main sub-sections of reviewed literature. It starts with information on Ghana's demographic, socio-economic, and health service structure. It continued with a literature review on the global TB burden, followed by sub-Saharan Africa and Ghana's TB burden. After which, a review on fidelity levels of intervention, determinants of implementation in health and TB screening among PLHIV. Finally, the section ends with information on the frameworks adapted for conducting the study. 2.1 The economy and health service structure of Ghana 2.1.1 Demographic and socio-economic information Ghana is a sub-Saharan African and a lower-income developing country found alongside the Gulf of Guinea and the Atlantic Ocean in the West Africa sub-region. Ghana covers 238,535 km2 and is bordered in the west with Ivory Coast, east with Togo, north with Burkina Faso, and with Gulf of Guinea and the Atlantic Ocean in the south (31). There are ten subnational government administrations known as regions within Ghana when planning the study. For local governance, there were further 216 administrative subdivisions of the regions called districts in 2017. This number has grown by splitting into 260 districts in 2018 and 2019. These districts are made up of ordinary, municipal and metropolitan. As shown in Figure 2 below, the ten regions were further regrouped into three ecological zones, namely Savannah, Forest and Coastal zones (32). 11 Figure 2: Map of Ghana showing the 10 regions and 3 ecological belts (32) It is estimated as of 2019 that Ghana has a mid-year population of 30,083,000, of which 29% were less than 15 years of age and 57.9% were between 15–64 years (31). However, the 2010 census in Ghana showed that most (71.2%) people were Christian, followed by Muslim (17.6%), and then the traditionalist (5.2%). The country has over seventy ethnic groups, with the populous groups being the Akans (47.5% ), the Mole-Dagbon (16.6%), the Ewe (13.9%), the Ga-Dangme (7.4%), the Gurma (5.7%), the Guan (3.7%), the Grusi (2.5%), the Kusaasi (1.2%), and the Bikpakpaam also known as Konkomba (3.5%) (31). 12 Starting 2018, the country had an annual population growth rate of 2.16%, a birthrate of 30.2 births per 1,000 population, a rate of death at 6.8 deaths per 1,000 population. In the same period, the fertility rate was 3.96 children born per woman, the infant mortality rate was 39.01 deaths per 1,000 live births, the contraceptive prevalence rate was 33%, and a life expectancy was 67.4 years (33). Ghana’s economy has various rich resource bases. These consist the manufacturing and exporting materials such as hydrocarbons industrial minerals. In addition, Ghana is the second- biggest producer of gold (after South Africa) and the second-largest producer of cocoa in Africa. Ghana currently has a nominal GDP of $67.077 billion and an inflation rate of 9.2% (2020 estimates) (33). Government total health expenditure for the year 2016 was 4.4% of GDP while 37.8% of total health expenditure for the same year was from out-of-pocket of the individuals (34). In 2015, Global Fund signed a grant of US$ 248 million for Ghana to increase substantially the number of people receiving prevention, treatment and care for TB, HIV and malaria (35). With Ghana’s universal health care system available, more than 12 million Ghanaians have enrolled in the National Health Insurance Scheme by 2019 (36). 2.1.2 The health service structure of Ghana The Ghana Health Service (GHS) is the mandatory institution responsible for healthcare provision and delivery in Ghana (37). The GHS is a Ghanaian government body established under the GHS and Teaching Hospitals Act 1996 as part of Ghana's Health Sector Reform and Medium Term Health Strategy and a distinct entity under the Ministry of Health (37). This distinctiveness made GHS different and not part of the overall Civil Service of Ghana. The Ministry of Health is in charge of policy formulation and enactment, while GHS is responsible for service provision. There are different levels of management teams under the GHS spanning from the regionals through districts to facilities. The team is made of a different calibre of health personnel headed by a substantive head of the respective institutions at each level. Consequently, the teams also vary considerably across regions, districts and facilities. 13 The health delivery system has different levels of facilities: health posts, health centres and clinics, district hospitals, regional hospitals, and tertiary hospitals (38,39). In 2017, there were ten administrative regions with 216 subdivisions called districts. At the district level, each is expected to have a functioning hospital known as the district hospital owned by the government of Ghana. So far, of the 216 districts, 140 of them have functioning district hospitals in 2017 (source: Ghana Health Service, DHIMS2) (38). The district hospitals are the highest facilities for clinical care serving a population between 100,000 to 200,000, with usually about 50 to 60 beds. In addition, they provide curative care, health promotion activities, and preventive care to the people and serve as referral points from health centres and polyclinics around its catchment areas. Treatment techniques involving surgery, laboratory, and other diagnostic techniques are performed at the district hospital. They run both out-patient and in-patient services. Healthcare provision is administered mainly by state-owned facilities, which form about 61% of total health facilities, with private facilities making 31% and faith-based health facilities making up the remaining 7% (38,39). Since 2003, the country had a universal health care system after the introduction of the National Health Insurance Scheme (NHIS) under the management of the Ghana National Health Insurance Authority (NHIA) (40). 2.2 The global burden of TB among PLHIV Tuberculosis (TB) disease is a key cause of illness and among the top 10 causes of mortality worldwide. Since 2007, it has been ranked above HIV/AIDS, a disease with the primary cause of mortality from a distinct infectious agent (41,42). Globally, TB occurs in all parts of the world and affects all age groups (42). In 2018, about 10 million people were estimated to have contracted tuberculosis(TB) globally, with men being the most affected ( 5.7 million), followed by women ( 3.2 million) and children ( 1.1 million) (41,42). Also, about a quarter of the global population has latent TB, and between 10-15% of these will progress to active TB disease over their overtime (42,43). Of the total global TB problem, 87% of new cases were accounted for by eight nations in 2018. India recorded the maximum number of new cases, while China, Indonesia, the Philippines, Pakistan, Nigeria, Bangladesh, and South Africa followed (41,42). Everyone is at risk of 14 contracting TB; however, adults in their productive years are primarily affected, and most (95%) of these cases and deaths occur in developing countries (41,42). Among all the TB infected persons, 8.6% are living with HIV (41), and TB is also the dominantly presenting illness among people living with HIV (44). People living with HIV are between 15 and 22 times more likely to get TB than those without HIV(3,4,44). TB is also causing more deaths among HIV infected people, with a quarter of all TB deaths being HIV related (41,44). In 2018, for example, about 16% (251 000) of all TB deaths occurred in persons living with HIV (42). Although this was a 60% reduction from 620 000 in 2000 (41), the current number is relatively high. According to the Global Fund report (2019) on assessment and best practices for the application of TB and HIV activities jointly, TB screening among PLHIV and ITP initiation coverage amongst people living with HIV (PLHIV) have been below targets over many years (45). In terms of spatial distribution, sub-Saharan Africa had the most problem of TB/HIV comorbidity, with approximately 84% of all TB/HIV deaths in 2018 borne by the region (44). Also, considering the distribution of TB/HIV deaths by African sub-regions, with the exception of Northern Africa, all the other sub-regions continued to increase the incidence of TB/HIV deaths over time (46). TB is a treatable, curable, and preventable disease (42,47); consequently, between 2000 and 2018, over 58 million lives were saved as a result of timely TB diagnosis and treatment (47). In 2016 alone, 83% of people were newly diagnosed with TB, and 54% of multidrug-resistant (MDR-TB) patients globally were successfully treated (47). Prominent global targets and milestones for reducing TB burden include Sustainable Development Goals (SDGs) (48) and the End TB Strategy by WHO (49). The third SDG goal consists of ending the global TB epidemic by 2030 (41,48). The End TB Strategy comprises set targets at 90% drop in TB deaths and an 80% drop in the TB incidence rate between 2015 and 2030; meanwhile, in 2020, a 35% decrease in TB deaths and a 20% decrease in TB incidence rates (41,49) were expected. However, the 2019 WHO report on TB indicated that most countries and regions with high TB issues were not in the position of attaining the 2020 End TB Strategy milestones (41). Universally, the TB incidence rate annually had declined by 1.6% from 2000 to 2018, with a cumulative decrease of only 6.3% between 2015 and 2018 (41). The worldwide decline in mortalities associated with TB was 11% between 2015 and 2018 (41). However, concerning 15 the WHO Africa region, there have been a 12% drop in TB incidence and a 16% drop in TB deaths between 2015 and 2018 (41). To reach the milestone of End to TB strategy or achieve the TB screening targets towards lessening the burden of TB globally, the extent to which the implementers implement the intervention needs assessment in furtherance of linking outcomes effectively to the intervention. 2.3 The burden of TB among PLHIV in sub-Saharan Africa The sub-Saharan Africa region is known to have the highest burden of both HIV and TB infection (50). Whiles accounting for only 12% of the global population, sub-Saharan Africa (SSA) accounted for 68% of the global HIV prevalence in 2010, with South Africa and Nigeria leading the count (51). Also, nine countries in sub-Saharan Africa were part of the 22 countries with the high burden of TB globally (52). Such countries included and in no particular order the Democratic Republic of Congo, Kenya, Ethiopia, Mozambique, Nigeria, South Africa, Uganda, Tanzania and Zimbabwe (52). Still, in SSA, about 33% of the people have latent TB infection (53), with a 10% chance for the immune-competent individuals to develop active TB and a 50% chance for those infected with HIV to advance to active TB later in life (54). The primary cause for the reappearance of TB in Africa is the relationship between TB and (HIV/AIDS) infection (55). The risk of dormant Mycobacterium TB advancing to active TB in HIV clients immunologically deficient is high (55). Compared with those not infected, the comparative risk of TB incidence among people infected with HIV is from 20 to 37 folds, subject to the stage of the HIV disease (52,56). According to the WHO tuberculosis report (2011), approximately 39% of all new TB infections in sub-Saharan Africa were attributable to HIV, while the global rate for the same variable was only 13% (52). In 2010, 82% of the global PLHIV incident TB cases were recorded in the WHO Africa region (51). In 2014, about 1.2 million people living with HIV worldwide developed tuberculosis, and 74% resided in sub-Saharan Africa (56). Also, half the number of people with TB in sub-Saharan Africa are infected with HIV (41). HIV and TB co-infection are characterised by high mortality rates globally (50), with TB standing out as the leading cause of death among people living with HIV (56). For example, in 2010, for instance, of about 1.40 million deaths from TB globally, 25% were attributable to HIV, representing a 13 16 percentage point increase from 12% in 2000 (52). Concerning TB mortality, about 390,000 people lose their lives from HIV-associated TB globally in 2014, representing a 32% reduction since 2004, with most of these deaths occurring in sub-Saharan Africa (56). As part of HIV and TB management and control plans, the WHO recommends a combination of antiretroviral treatment (ART) and the three I’s to lessen the morbidity and mortality resulting from TB between people living with HIV (57). The three I strategies are (i) Intensified case-finding of TB (ICF), which is the critical component among the interventions, (ii) Infection control (IC), and (iii) Isoniazid preventive therapy for TB diagnosed negatives (57). However, the primary implementation bottlenecks of TB/HIV activities in sub-Saharan Africa in which studies were conducted have been more of limited resources and technologies to identify and eliminate active TB, high treatment dropout rates and side effects mainly due to hepatotoxicity and the risk for INH-resistant TB after IPT administration (58,59). Another major implementation bottleneck in SSA is the suboptimal level of incorporating TB and HIV Care (58), even though the implementation of the joint TB/HIV collaborative activities saved about 8.4 million people globally (56). However, it is characterised by low HIV counselling and testing uptake in TB clinics, a high workload, and a significant amount of cross-referrals between both services, usually not within the same health facility (60). Some of these bottlenecks have resulted in low TB treatment success rates after registering smear-positive TB cases under the DOTs strategy within the sub-Saharan African region (61). In a systematic review of the treatment success rate for pulmonary TB patients who were adults located within sub-Saharan Africa between 2008 and 2018, the pooled treatment success rate was 76% (61). This falls below the global rate of 83% and the WHO target of 90% (61). The pooled cured rate was 64.5%, and the pooled treatment completion rate was only 13.7% (61). There are also wide variations with the country-specific TB treatment success rates in SSA. Studies have shown a TSR of 80% in South Africa (62), 90.1% TSR in Ethiopia (63), 39% TSR in Uganda (64), 70% TSR in Zimbabwe (65), 57.7% TSR in Nigeria (66) and 85% TSR in Ghana (67). Irrespective of the enormous challenges that TB and HIV diseases posed to healthcare systems and the WHO recommended policy interventions in sub-Saharan Africa, only 9% of PLHIV had been screened for TB. Moreover, only 3% were put on IPT (19). To identify and address the gaps to improve the implementation process, country-specific information assessed the extent or degree to which the potential implementers of the various interventions, such as the 17 active TB case finding among PLHIV is implemented and the factors that influence the implementation are required in urgency. 2.4 Tuberculosis in Ghana 2.4.1 The burden of TB and TB/HIV co-infection in Ghana Tuberculosis is highly endemic across all regions in Ghana. The country is one of the documented countries with the most burdened TB/HIV in the world. Ghana recorded 44,000 TB cases and 15,800 TB deaths in 2018 (67). The incidence rate was estimated to be 148 per 100,000 population, with men being the most affected; male to female infection ratio of about 2:1, and males over 45 years recording the highest-burden (67). The TB case notification rates vary across the different regions, with the lowest being 23.7 cases per 100 000 population in the Northern region (now Northern, Savannah and North-East regions), with the highest being 77 cases per 100 000 population in the Volta region (now Volta and Oti regions) (67,72). The estimated number of people who developed TB and were co-infected with HIV (HIV-positive TB incidence) in 2018 was 8600 cases and 29 cases/100 000 population was the incidence rate. Also, 870 cases were multi-drug resistant tuberculosis (MDR/RR-TB0 in 2018, having an incidence rate of 2.9/100 000 population (67). The total TB case fatality ratio as of 2018 was 39%, with about 64% of all TB-patient households experiencing an impact of major and sudden healthcare expenditures (67). In Uthman et al.'s (2009) study aimed at determining the spatial distribution of TB/HIV deaths in Africa, they found that Ghana was in the medium-endemic category (between 11.5-30.5 deaths per 100 000 population) per year (46). The Centre for Diseases Control and Prevention Division of the Global HIV and TB reported that in 2017, the TB incidence was estimated as 152/100 000 population with a mortality of 36 per 100 000 population (73). Interventions to accelerate the decline in the burden and incidence of TB and TB among PLHIV and mortality are essential. But what is more important is the availability of reliable information on the barriers and enablers of the implementation and the degree of implementation of the intervention. 18 2.4.2 Tuberculosis and HIV collaborative activities in Ghana In July 1954, Ghana established the Ghana Society for the Prevention as the first TB control programme to support and supplement the government’s efforts to combat TB (74). The WHO DOTS Strategy was later adopted by Ghana in 1994, which was grounded on the commitment of politicians; sputum smear microscopy diagnosis, standardization of supervised treatment, regular drug supply and recording and reporting system (74,75). The Ghana National TB Control Programme (NTP) currently implements the DOTS Strategy as the primary intervention for TB control (74,75). In response to the global dual epidemic of TB/HIV, Ghana recognised high tuberculosis and HIV dual infection in the country and that each of these diseases fans the impact of one other. After the policy guideline of WHO on TB/HIV joint activities in 2004, a mechanism for commissioning the TB/HIV collaborative activities was introduced in Ghana as early as 2005 at the national level through the establishment of a national TB/HIV coordinating council (76). The established body demarcated roles and responsibilities for both the TB and HIV control programmes, intending to reduce or minimise duplications of efforts in the fight against both diseases and to coordinate scarce budgets. As a result, the two programmes enhanced their services to ensure that something is done to reduce HIV morbidity and mortality in the country by the NACP. In contrast, the NTP ensured that something is also done to reduce the morbidity and mortality of TB in Ghana. The coordinated efforts of NTP and NACP led to the implementation of a national policy in line with the WHO TB/HIV collaborative policy guidelines in 2007 to fight and halt the emerging epidemic (15,24). Three-fold policy goals were set; “to strengthen the health system to respond to the TB/HIV dual epidemic, decrease the burden of TB in people living with HIV/AIDS, and decrease the burden of HIV in TB patients” (15,76). Three TB/HIV integrated service models were also ongoing at various healthcare delivery facilities (76). Firstly, screening of TB clients for HIV at TB clinics and linked to care; secondly, screening of PLHIV for TB at HIV clinics at every visit and treated and thirdly, some levels provide one-stop service delivery for TB/HIV, and the others make referrals between programmes (76). The policy made sure referrals systems were in place to remove bottlenecks for the patients and provider, screening tools and algorithms to facilitate this collaboration, and programme guidelines and manuals for training reflect TB/HIV co-infection management (76). 19 To control the burden of TB, one of the guideline approaches was the active or intensified case finding which means “the systematic identification of people with presumptive active TB, in a predetermined target group, using tests, examinations or other procedures applied rapidly” (77). What was crucial to this active case finding is the systematic pro-activeness by trained health workers to conduct screening among the predetermined high-risk population compared to screening in response to individual complainants seeking healthcare. These predetermined high-risk populations included immigrants from areas with high rates of TB, groups with increased rates of TB infection, including PLHIV, the homeless and those on drugs, etc. On the other hand, TB case finding responded to individuals' complaints during care-seeking was termed a passive case finding (77). The passive case finding was the old approach of TB case finding, which led to missed TB cases in an earlier and usually less infectious stage at the facility level (7,78). Compared with the passive, health care providers actively screened for TB without clients complaining of symptoms by themselves. The fact that HIV weakens the immune systems make PLHIV prone to TB. It is important in the TB/HIV double epidemic fight to reduce the TB fanning effect within HIV clients. A person with latent TB infection can develop TB if the person becomes HIV positive. WHO therefore recognised the screening of PLHIV for TB components of its recommended policy as key in the fight for TB control. Thus, this fundamental approach required all PLHIV to have routine TB at every clinical encounter (3,15). In addition, they are supposed to be given education and counselling on TB at the clinic. Through screening for TB, the client aided with a simple questionnaire seeking whether clients experience symptoms such as fever, loss of weight, cough irrespective of the duration, and night sweats in the past defined period. This was followed by a confirmatory diagnostic test with either X-ray or GeneXpert or rapid diagnostic for all suspects (positive response to one or more of the symptoms), initiating appropriate treatments and ensuring TB infection control at all HIV care clinics in Ghana (16,79). Before 2018, all the HIV clients screened negative for TB were not put on IPT because it was not a national policy (15,24). At the HIV care clinic from 2007 to 2017, IPT use was only supported in specialised cases where treatment of a patient is supervised through to finish (23,24). In 2018, the Ghana Health Service, through the NTP and NACP, planned to begin the implementation of the IPT use in phases at HIV clinics for all HIV clients who have screened negatives for TB, starting in some selected HIV care clinics with the hope of spreading to all HIV care clinics by the end of the year 2020 (79). 27 facilities from the ten Ghana regions with digital X-ray facilities and GeneXpert for conforming TB test results were identified to begin 20 the IPT initiation. In 2019, facilities with any digital X-ray machine or GeneXpert will implement the IPT, and in 2020 all HIV care clinics with or without any of the devices will initiate the IPT (23). At the facility level, each facility implementing the intervention has to ensure amongst others the availability of providers trained on TB/HIV, availability of IPT, lab and testing equipment, protocol and guidelines for the providers, tools and ensuring meetings (15,24). Every HIV care provider is expected to go through the clinical protocol for every HIV client seen. The main activities on the clinical protocol include TB education, counselling, screening for TB using the screening algorithm, requesting for confirmatory test and initiation of IPT or appropriate treatment (1,7,15). Nonetheless, TB screening coverage among PLHIV were low and fluctuated overtime since the implementation of the TB screening intervention at HIV and ART clinics. Such an abysmal performance is most likely to be attributed to the extent to which the intervention is implemented, unavailability of resouses and logistics and, possibly inadequate training for the healthcare workers. The two programmes revised the screening algorithm in 2017 to include IPT initiation before implementing the IPT administration, as shown by Figure 3. Every HIV client is screened at the ART/HIV clinic for clinical signs and symptoms of TB. Confirmatory tests are further conducted before appropriate treatment is given. The guideline suggests that IPT can only be given if the PLHIV has a normal chest X- ray. But in 2018, and as a high TB/HIV burden country (80), about half (50%) of the PLHIV were screened for TB with none being put on IPT (45). The reason behind the failure or low uptake of IPT in the country could be similar to what was found Uganda setting due to context similarities. Pre-IPT counselling and logistical supplies at the HIV clinics coupled with training of healthcare providers, supervision, and monitoring of the implementation (81). The national TB survey conducted in 2013 in Ghana indicated that TB prevalence in Ghana was 290 per every 100,000 persons compared to WHO estimates (72/100,000 persons) in the same year (9,79). Available reports from the national HIV Sentinel Survey show that HIV prevalence declined from 3.6% in 2003 to 1.6% in 2014 (82). There was, however, a slight increase to 1.8% in 2015 and 2.4% in 2016 (82). The current TB/HIV dual epidemic in Ghana is of public health concern due to its population health and social impact (83). Sector reports from the TB and AIDS control programmes show that the working and active population between 30 to 39 years are mostly affected (9,82). Studies conducted in Ghana have shown that TB caused 40-50% of deaths in HIV positive people (83,84). 21 Figure 3: Revised screening algorithm (source: NACP) 22 In 2017, according to the global tuberculosis report 2018, the number of new people enrolled in HIV/ART care clinics was 31,058, of which 1,456 was notified as a TB case (80). The ambitious targets set by the TB/HIV programme could be achieved based on the intervention being implemented as planned by the NTP. But efforts to achieve the desired outcomes are seemingly not informed by an understanding of the constraints that may hamper successful implementation (e.g. fidelity to the intervention by the facility and the health care provider) (20–22,26,85) 2.4.3 Funding the needs of TB control Progress in TB control all over the world requires adequate funding that is sustainable for longer years. The primary funding for each country towards TB care and control is by domestic resources for its health services budget, and by donors, and bi-national and multinational donor partnerships. For instance, the BRICS (Brazil, Russia, India, China, and South Africa) and upper middle income countries, which account for almost half the world’s TB cases funds almost all their funding needs from domestic sources to support the TB control. Nonetheless, the main source of funding for TB control for low to middle income countries relies heavily on international donor funding, mainly the Global Fund. TB control funding grew considerably since 2002, showing impressive cost-benefit gains across the globe (68). Conventionally, an intervention is considered cost-effective if the cost per year of life saved is less than the gross domestic product (GDP) of a country (69), which is the case in many countries (70)., The average cost incurred by TB patient in low- and middle-income contours is catastrophic. Such catastrophic cost out-of-pocket leads to impoverishment of household and prevents access to healthcare. Hence, the TB management and treatment care has been free under the Ghana National Tuberculosis Programme. In Ghana, the current strategy for controlling the TB relies much on implementing and scaling up best known practices, while addressing the problems of most affected populations such as PLHIV. The principle underlying implementation is strong coalition with civil society organizations and communities (Stop TB Partnership, Ghana) and working in partnership with other state agencies such as Attorney General’s Department, to ensure protection and promotion of patients’ rights, ethics and equity under National Health Insurance Scheme (NHIS), Food and Drugs Authority (FDA) and Public Health Act. There is available TB screening tools/questionnaire and diagnostic tools which had led to the shift from passive TB case finding to active case finding. Ghana piloted the award winning evidence 23 based WHO guideline, Systematic Screening for active tuberculosis (principles and recommendations). During the piloting, the cost of the strategic plan was assessed through WHO planning and budgeting tool. According to the national TB health sector strategic plan for Ghana from 2015 – 2020 (71), it was estimated that an amount of USD 358,817,198 to implement the six year National Health Sector Strategic Plan. The Government of Ghana provided 30% of the annual budget needs while the Global Fund-financing mechanism provided 6% of the total funding needs for the first three years. Hence, the successful implementation of the plan depended on a continuous stable political climate in the country and increased, predictable and sustained funding from other developmental partners. 2.5 Implementation fidelity assessment of health interventions Fidelity of an intervention is one of the eight implementation outcomes that can be examined to assess an intervention's applicability, progress, or performance in a real-life setting. Several terms and phrases have been used to describe implementation fidelity assessment in healthcare interventions. Terms such as fidelity, integrity, adherence (although this is a component of fidelity), treatment fidelity, intervention fidelity have all been used to describe fidelity of implementing healthcare programmes or interventions (21,27,86–93). All these terms share a similar definition of implementation fidelity as the extent to which program developers deliver a healthcare intervention to recipients as intended or planned (27,87,88,90,91). In addition, implementation fidelity is usually defined in specific dimensions of a healthcare intervention that need to be measured or assessed (21,27,86,88,89,91–97). These dimensions are adherence to the content of the intervention, dose, delivery quality, responsiveness of the participant, and differentiation of the program (27,86,91,95). Adherence means how the healthcare providers follow the intervention protocol (27,86). Adherence is the main measure of implementation fidelity (27), and it focuses on the number of prescribed activities as predefined in an intervention guideline or by the intervention protocol (91). The intervention’s content is the active ingredients; the diagnosis procedure, screening, drug, treatment, skills, or knowledge that the intervention seeks to deliver to its participants (27,86). Dosage or exposure means the quantity of the intervention that the participants received; in other words, whether the frequency and duration of the intervention are as comprehensive as agreed by its designers (27,86). Quality of delivery on the other hand 24 is how a staff member or provider carries the program (27,86). Participant responsiveness is concerned with the responsiveness and or engagement of participants to an intervention (27,86). The fifth is program differentiation which is “identifying unique features of different components or programs” and identifying “which elements of programmes are essential”, deprived of which the programme will not have its intended effect (27,86). There have been recent advancements in the methods used to gather and analyse fidelity data. These methods include provider self-reports, observations, project documentation, client records, and participant surveys. The self-report approach comprises data collected directly from the provider or intervention recipient using checklists or verbal reports (98). Questionnaires and interviews are examples of self-reports. With the questionnaire method, the participants record their answers. On the other hand, with the interview method, the interviewer records the responses given by the participants (98–100). In self-report fidelity assessment, implementers are mostly asked to specify whether they offered the identified constituents of the intervention guideline and ask the intervention recipients if they received the specified intervention activities. Instead of inferring by observation, participants describe their own experiences. Hence self-report approaches are not costly and do not waste time compared to observational techniques, but self-report data faced are potentially affected by validity and accuracy issues (99). For example, participants may report biasedly about the implementers based on their feelings towards the implementer (91), and provider ratings of fidelity can be skewed positively due to social desirability bias. Such bias can also be limited by providers' or participants' ability to recall accurately or provide correct information (86). Despite this, self-report interviews allow study participants to define and or come out with their understanding and practice. Behavioural observations of an implementation process turn to produce a more objective assessment of the program. It allows observers to examine whether implementers delivered the recommended content, used the correct delivery technique, engaged the recipients during the delivery, or expressed interest in providing the intervention (86,91). Observational procedures may consist of checklists, scaled rating forms, or qualitative descriptions of project implementation (86). However, observational methods are expensive, require more observers, and implementer reactivity to observation can change implementation fidelity (91). For example, although some implementers may be more adherent to the intervention guidelines, they might be anxious while under observation, which can influence the level of adherence or competence (91). Therefore, the effect of reactivity may underestimate the status of implementation fidelity (91). 25 Facility and administrative records, such as registers, case count, and client records, are crucial in assessing implementation fidelity (86). For example, TB screening coverage among HIV positive clients can be assessed using HIV clinical care registers. In addition, a review of clinical records can determine if certain intervention elements were delivered by providers (86). Irrespective of the method used, developed fidelity tools can take on either dichotomous, Likert scale, open-ended responses, or any combination subject to the kind of intervention, nature of the data, and type of analysis (27,86). In assessing fidelity, the fidelity data should be compared to what the implementation guideline specified as core to be delivered to assess the level at which each component of the intervention is provided (86). Another fidelity assessment approach that has shown more statistical power in its process involves continuously analysing implementation fidelity data using the full range of implementation data (20). In this case, percentages can be used to assess the level of fidelity achieved and correlated with outcomes (20). Combined scores can also be created with program data to reflect the overall level of fidelity based on the sum of all fidelity items or dimensions (101). Mowbray et al. reviewed literature extensively on the steps to develop, measure, and validate fidelity criteria (101). As shown with examples from literature in Table 1, three key steps in creating fidelity criteria are identified as (1) identification of critical components of the intervention, its data sources and operational definitions, (2) collection of data to measure the indicators through an appropriate method, and (3) assessing the indicators regarding their validity and reliability (101). Whenever the health workers perform TB screening activities, some level of quality is compromised. This quality could also be in terms of numbers screened against what was to be screened or in terms of a