The Views of Public Service Managers on the Implementation of National Health Insurance in Primary Care: A case of Johannesburg Health District, Gauteng Province, Republic of South Africa. Dr SD Murphy Student Number: 301690 A publishable article as research report submitted to the Faculty of Health Sciences, University of the Witwatersrand in partial fulfilment of the requirements for the degree of Master of Medicine in Family Medicine (M.Med in Fam. Med) Published in BioMedCentral: Health Services Research Citation: Murphy, S.D., Moosa, S. The views of public service managers on the implementation of National Health Insurance in primary care: a case of Johannesburg Health District, Gauteng Province, Republic of South Africa. BMC Health Serv Res 21, 969 (2021). https://doi.org/10.1186/s12913-021-06990-4 Supervisor: Prof S. Moosa (MBChB, MBA, PhD) Family Medicine University of Witwatersrand Johannesburg, 2021 ii iii 1 Research Declaration I, SHANE DARREN MURPHY, hereby declare that this research is my own unaided work, except where due acknowledgement for assistance received has been made. It is being submitted for the degree of Master of Family Medicine at the University of the Witwatersrand, Johannesburg. It has not been submitted previously for any other degree or examination at this or any other University. Signed …………………………. Shane Murphy (Candidate) Date: 07/06/2021 Signed ……………………… Professor Shabir Moosa (Supervisor) Date: 07/06/2021 iv Acknowledgement I would like to thank thanks the staff of the Faculty of Health Sciences, Division of Family Medicine, University of Witwatersrand for their guidance and support during the course of this research. Secondly, I would like to thank my wife, Nisha Makan, for her unwavering support throughout the Family Medicine programme. v Dedication I dedicate this research report to the citizens of South Africa. It is our hope that we can contribute to the effective implementation of Universal Health Coverage in South Africa for the progressive realisation of the access to equitable quality care. vi Table of Contents Acknowledgement .............................................................................................................. iv Dedication .......................................................................................................................... v List of Tables ................................................................................................................... viii Nomenclature .................................................................................................................... ix The Views of Public Service Managers on the Implementation of National Health Insurance in Primary Care ...................................................... Error! Bookmark not defined. Managerial Engagement in Policy Development ............................. Error! Bookmark not defined. Managerial Views on NHI ............................................................... Error! Bookmark not defined. Perceptions of NHI implementation ................................................ Error! Bookmark not defined. Infrastructure ................................................................................................ Error! Bookmark not defined. Health Systems ............................................................................................... Error! Bookmark not defined. Corruption ..................................................................................................... Error! Bookmark not defined. Human resources ........................................................................................... Error! Bookmark not defined. Fragmentation ............................................................................................... Error! Bookmark not defined. Digital Health ................................................................................................. Error! Bookmark not defined. Discussion ............................................................................... Error! Bookmark not defined. References: .............................................................................. Error! Bookmark not defined. Appendix I: Semi-structured interview protocol ................................................................ 17 Appendix II: Participant information sheet and informed consent forms .......................... 19 Appendix III: Ethics approval – Human Research Ethics Committee (University of Witwatersrand) ................................................................................................................. 25 Appendix IV: Ethics approval from district (National Health Research Database) ............ 26 Appendix V: Research proposal ........................................................................................ 28 INTRODUCTION .................................................................................................................... 34 BACKGROUND LITERATURE ANALYSIS AND CRITIQUE ............................................ 34 The plan for UHC in South Africa ........................................................................................... 34 International experience with decentralized management and UHC implementation ............. 35 Decentralised management in South African primary care ..................................................... 37 Public health district and sub-district level managerial engagement with NHI ....................... 37 Summary .................................................................................................................................. 37 DEVELOPING THE RESEARCH QUESTION ..................................................................... 38 Research Question .................................................................................................................... 38 RESEARCH OBJECTIVES .................................................................................................... 38 1. To explore district and sub-district level public service manager’s views on NHI. .................................. 38 2. To explore the experience of district and sub-district level public service manager’s engagement in NHI policy development. ..................................................................................................................................... 38 3. To explore district and sub-district level public service manager’s perceptions of managerial and public sector readiness for the implementation of NHI. .......................................................................................... 38 vii 4. To explore district and sub-district level public service manager’s perceptions of potential challenges as well as solutions to effectively implementing NHI. ..................................................................................... 38 RESEARCH METHODOLOGY AND STUDY DESIGN ....................................................... 38 Maintaining Scientific Rigour .................................................................................................. 38 Research methods ..................................................................................................................... 39 Sampling ................................................................................................................................... 39 Sample Size ............................................................................................................................... 40 Data collection .......................................................................................................................... 40 Data Management .................................................................................................................... 40 Qualitative Data Analysis ......................................................................................................... 41 Steps in Framework Analysis27 ................................................................................................. 41 Computer Assisted Qualitative Data Analysis Software .......................................................... 41 Research Process Flow Chart ................................................................................................... 42 ETHICS .................................................................................................................................... 42 TIMING ................................................................................................................................... 42 FUNDING ................................................................................................................................ 43 REFERENCES ......................................................................................................................... 44 Appendix VI: Proofreading Certificate ............................................................................. 48 Appendix VII: Turnitin Report ......................................................................................... 49 viii List of Tables Page Table 1: Overarching questions used in in-depth interviews 7 Table 2: Respondent profiles 8 ix Nomenclature CUPS : Contracting Unit for Primary Care DHMOs : District Health Management Offices DHS : District Health System MaxQDA : “Max” Qualitative Data Analysis NHA : National Health Act NHI : National Health Insurance PHC : Primary Health Care SDG’s : Sustainable Development Goals UHC : Universal Health Coverage WHO : World Health Organisation The Views of Public Service Managers on the Implementation of National Health Insurance in Primary Care: A case of Johannesburg Health District, Gauteng Province, Republic of South Africa. S D Murphy,1 MBChB, MPH, Dip PEC, Dip HIV Man, H Dip Emerg Med S Moosa,2 MBChB, MBA, PhD 1Department of Family Medicine, School of Clinical Medicine, University of the Witwatersrand, Johannesburg, South Africa 2Department of Family Medicine, School of Clinical Medicine, University of the Witwatersrand, Johannesburg, South Africa Corresponding author: S D Murphy (murphy.shanedarren@gmail.com) 2 Abstract Background. The South African government is implementing National Health Insurance (NHI) as a monopsony health care financing mechanism to drive the country towards universal health coverage. Strategic purchasing, with separation of funder, purchaser and provider, underpins this initiative. The NHI plans contracting units for primary healthcare services (CUPS) to function as either independent sub-district purchasers or public providers and District Health Management Offices to support and monitor these CUPS. This decentralised operational unit of PHC, the heartbeat of NHI, is critical to the success of NHI. The views of district-level managers, who are responsible for these units, are fundamental to this NHI implementation. This qualitative study aimed to explore district and sub-district managerial views on NHI and their role in its implementation. Methods. Purposive sampling was used to identify key respondents from a major urban district in Gauteng, South Africa, for participation in in-depth interviews. This study used framework analysis methodology within MaxQDA software. Results. Three main themes were identified: managerial engagement in NHI policy development (with two sub-themes), managerial views on NHI (with three sub-themes) and perceptions of current NHI implementation (with six sub-themes). The managers viewed NHI as a social and moral imperative but lacked clarity and insight into the NHI Bill as well as the associated implementation strategies. The majority of respondents had not had the opportunity to engage in NHI policy formulation. Managers cited several pitfalls in current organisational operations. The respondents felt that national and provincial governments continue to function in a detached and rigid top-down hierarchy. Managers highlighted the need for their inclusion in NHI policy formulation and training and development for them to oversee the implementation strategies. Conclusion. It appears that managers are not adequately capacitated to implement NHI mechanisms such as strategic purchasing. NHI policy implementation appears to function in a rigid top-down hierarchy that excludes key stakeholders in the NHI implementation strategy. The findings of this study suggest an inadequate decentralisation of healthcare governance within the public sector necessary to attain UHC. District managers need to be engaged and capacitated to operationalise the planned decentralised purchasing-provision function of the DHS within the NHI Bill. Keywords: Universal Health Coverage, National Health Insurance, Managerial Capacity, Qualitative research, Decentralised governance, Primary care. 3 Background South Africa, one of the most unequal societies in the world, is an upper middle-income country that suffers a crippling quadruple burden of disease.1 At the center of this crisis, South Africa’s fragmented healthcare system fails to cope with this disease burden, and further propagates societal inequity.2 The extant bipartite health system was institutionalised through historical ethnic and socioeconomic deprivation and marginalization. The Apartheid regime established 14 separately-operating health departments3 and passed legislature that favoured the privatisation of healthcare. These exclusionary structures have systematized the view of health as a commodity in the post-apartheid era and continue to drive excessive medical specialisation coupled with the bloom of myriad independent medical schemes. The dawn of democracy in 1994 heralded a social solidarity that sought to establish the basic human right to the progressive realisation of the access to equal and quality healthcare.4 However, the prevailing apartheid legacy has continued to drive social inequity - South Africa spends nearly 9% of its gross domestic product on healthcare,5 with more than 80% of this allocation used to provide healthcare services to the top socioeconomic quintile.6 The country’s fragmented healthcare system remains disease-oriented and perpetually disadvantages the poor with unequal access to care and catastrophic out-of-pocket expenditure. The South African government is introducing a new healthcare financing and service delivery system - National Health Insurance (NHI) – to reorganize healthcare governance to achieve universal health coverage (UHC). UHC is advocated by the World Health Organisation7 as a means for governments to improve health levels, reduce the burden of disease, cross-subsidise risk and enhance social equity through increased access to comprehensive services, increased population coverage, and financial protection. NHI intends to create a single financial pool for national risk-pooling and cross- subsidisation that will generate funds through an increased GDP expenditure on health, new payroll taxes, and increased income tax and value-added tax.4 This fund looks to subsume the fragmented health insurance schemes and create a uniform, equitable social insurance scheme that standardises healthcare services across South Africa. UHC, through NHI, flips the view of health as a commodity to health as a basic human right, and represents a considerable reorganization of the healthcare service delivery model. Although UHC is a laudable social construct, South Africa has failed several previous endeavours to implement structural and financial health reforms.2 The NHI Bill acknowledges these failures to translate theoretical frameworks into actionable plans, and posits the devolution of healthcare governance from provincial health structures to district and subdistrict managers (DMs) as critical to successful NHI implementation.4 The National Health Act of 2003 (NHA)2 created the structures of the current public health system with national and provincial departments of health that sought to provide redress to the fragmented healthcare system. Within this legislative framework, the national health council oversees policy formulation and national priority setting, while provincial government is legislated (through the intergovernmental fiscal relation system) to receive the bulk of financing and is responsible for healthcare service delivery through the district health system as an agent of the province. Operational control of districts remains vested in provinces, with DMs appointed by provinces and local government facilities and staff being moved under provincial control. This top- heavy model has resulted in a detached governance structure that fails to meet both organisational standards of compliance13 as well as the needs of South African citizens.2 4 One of the central premises of the NHI Bill is that the devolution of governance to DMs will synergise top-down policy with the insight and leadership of DMs who face the complex realities of healthcare service delivery. To facilitate the reorganisation of the healthcare system, the NHI Bill repeals the previous legislative framework in the NHA and vests control of personal public services with DMs within their districts. Strategic devolution of governance, coupled with commensurate managerial capacitation, intends to empower DMs with the authority and expertise to implement gatekeeping (of entry and access to standardised healthcare services), establish clear referral pathways and reorient the service delivery towards comprehensive and integrative services through primary healthcare (PHC) re-engineering.4 The Bill intends for the NHI fund to finance Contracting Units for Primary healthcare Services (CUPS) that will function either as sub-district purchasers, funded by the NHI fund, or as current sub-district public providers contracted with NHI to provide personal services. Contracting Units for PHC Services (CUPS) are the preferred organisational unit (comprising an integrated network of district hospital, clinics or community health centres, and private providers) with which the NHI fund will directly contract to provide PHC services in a subdistrict. District Health Management Offices (DHMOs) will facilitate and coordinate PHC services and strategic purchasing2 at the district level, directly accountable to the national department. Current DMs will function within DHMOs to support and monitor these CUPS.14,15 Longstanding UHC schemes such as the National Health Scheme (NHS) in the United Kingdom (often thought of as the archetypal UHC system) and the Medicare system in Canada have invariably improved national health and economic outcomes within their respective countries.16 Stakeholders maintain that accountable and delegated management is the greatest predictor of successful UHC implementation16 and both systems have reiterated the need for a ‘strategic vision’ at the level of district organizational structures,16 as well as the capacitation of local health managers (DMs) to improve their ability to enforce implementation. In the developing world, regions such as Asia, Latin America and sub-Saharan Africa have shown that inadequate decentralization of health system governance is crippling to UHC implementation strategies.7,17,18 In contrast, UHC successes in countries such as Thailand and Indonesia, nations with low gross national income per capita, have been attributed to sustained federal investment in devolving governance to the district structures of healthcare.10 Similarly, the Cuban health care system, which relies predominantly on CUPS, has shown that the decentralised approach to healthcare allows targeted strategies that address community- specific needs, increase the equity of care, and promote managerial flexibility and accountability.17 In the South African context, the first phase of NHI (2012-2017) piloted health services strengthening initiatives targeted at PHC – the “heartbeat of NHI”.19 Ten interventions were implemented across eleven pilot sites. A recent external evaluation of phase 1 was released in 20192 which showed mixed results across the sites. The authors noted several shortfalls in critical governance components, similar to those in several countries already implementing UHC,7,17,20 due to an inadequate decentralision of governance. The findings of this review support previous studies showing that a recurrent obstacle to NHI implementation is its dependence on the detached and centralized governance model.21 A recent report by the Office of Health Standards Compliance13 (OHSC) showed that primary healthcare clinics yielded a 47% compliance across seven domains of quality. The domain of leadership and governance scored 47% - classified as ‘non-compliant’ and ‘requiring urgent intervention’. Another prominent concern is that the lack of transparency and accountability, inherent in a centralized model of governance, allows corruption (in the form of irregular expenditure) to remain unchecked - the OHSC audited nearly 45% of health care expenditure as irregular.13 5 It is evident that the effective devolution of governance will play a vital role in the mechanisms intended to facilitate the implementation of NHI. DMs need to be fully aware of, and engaged in, the NHI structures, processes and rollout plan. Despite the growing appreciation that the success of UHC interventions is critically dependent on decentralized management, a thorough literature review yielded no studies evaluating this. We aimed to explore the views of DMs on NHI, as well as their engagement in policy development and implementationvand health system reorganisation as mandated by the NHI Bill. Methods This was an exploratory descriptive study with a focus on applied policy. Our study was conducted within the Johannesburg Health District, one of Gauteng province’s seven health districts, between 2020 and 2021. This district was selected as it is a major district with geopolitical proximity to provincial and national governance structures. Delays and challenges in obtaining ethics approval from several district ethics committees hindered a planned evaluation of other Gauteng districts. A list of DM personnel was obtained from the district ethics committee to facilitate sampling of data rich participants. Sampling was conducted through a framework of stratified random sampling so that a balanced mix of DMs from all backgrounds would be sampled. Ethics committee approval of the study was obtained from the University of Witwatersrand’s Human Research Ethics Committee (M191046), as well as the Research Committee of the Johannesburg Health District (GP_202006_048). In-depth interviews were selected as the method of data collection to allow respondents to develop their own description of their perspectives and experiences. The principal investigator (SAM) conducted all interviews. Both researchers had undergone several courses on qualitative interviewing skills through their university and conducted previous qualitative work. Further, the senior researcher has previously published several qualitative studies.22,23 Email invitations which included a participant information sheet were sent to prospective participants. Due to the COVID-19 pandemic, interviews were conducted with consenting participants by telecommunication software as regulated by the ethics committee. The four core questions used in the interviews are described in Table 1. Interviews were recorded via a dictaphone and transcribed verbatim into electronic text format. The concepts of ‘information power’ and ‘informational redundancy’24 informed pragmatic sampling saturation. Saturation was reached following seven in-depth interviews, although a total of ten interviews were conducted. Transcripts were sent back to participants for member-checking. No changes were made following member checks. Transcripts were then imported into computer-assisted qualitative data analysis software (MaxQDA, 2020)25 to facilitate robust and explicit data analysis. Data analysis followed the five-step approach of Framework Analysis26: familiarization, framework formulation, coding and indexing, charting, and mapping and interpretation. Peer-checking of the thematic analysis, as well as the greater framework analysis, was performed by the second researcher. The researchers met regularly to cross-examine analyses. No divergent interpretations arose during this process. Measure to ensure the trustworthiness of this work are listed under table 2 below. Table 1: Overarching questions in the interview schedule Nº Question 1. What are your views on National Health Insurance? 2. Can you tell me about your engagement in NHI policy development? 3. What is your perception of the implementation of NHI? 4. Is there any other view or thought you would like to express? 6 Table 2: Measures to ensure study trustworthiness Component of Trustworthiness Measures Credibility This study will used validated Framework Analysis method. Critical reflexivity was used during interviews with thick descriptions of data to facilitate internal validity. Transferability Detail of the research setting with an audit trail is provided for future research and cross comparison. Confirmability Computer assisted qualitative data analysis software was used to leave a paper trail alongside reflexive commentary. Quotes are used for in vivo description. Dependability Study design and methods are made explicit so that researchers can emulate the study design. Results Ten district managers were interviewed for the study. Participant ages and work experience within healthcare services ranged between 35-65 years and 6-42 years respectively. However, the duration of their current position as a manager ranged from one month to six years. Six participants were female and four were male. Seven participants were DMs, and three participants were sub-DMs. All the research participants were employed within the Johannesburg Health District in Gauteng. The respondent profiles are summarised in table 3 below. Table 3: Respondent profiles Identifier Ethnicity Gender Designation Duration in Current Position Respondent 1 (R1) African Male DM* 1 month Respondent 2 (R2) African Male DM 3 months Respondent 3 (R3) Coloured Female DM 6 months Respondent 4 (R4) White Female DM >1 year Respondent 5 (R5) African Female sub-DM# >1 year Respondent 6 (R6) Indian Male DM >1 year Respondent 7 (R7) African Male sub-DM >1 year Respondent 8 (R8) African Female DM 4 months Respondent 9 (R9) African Female sub-DM >1 year Respondent 10 (R10) Coloured Female DM 3 months Footnote: *District-manager; #sub-district manager Three main themes were identified, represented in the figure below: managerial engagement in NHI policy development (with two sub themes), managerial views on NHI (with three sub-themes), and perceptions of current NHI implementation (with six sub-themes). 7 Figure: Summary of findings Managerial Engagement in Policy Development Communications received and engagement with NHI rollout The respondents described ambiguous experiences regarding their engagement in policy development. Respondents cited uncertainties around roll-out plans as well as their own managerial responsibilities that they attributed to a near absence of communication around NHI from central governance structures. Several respondents were aware of other policy implementations such as PHC re-engineering and Ideal Clinics4 as part of the primary care improvement and greater NHI communication and implementation from national and provincial government, but struggled to see the bigger picture, often leaving them with more questions than answers: The third respondent (R3) stated, ‘You don’t get the total information, you get an introduction of what you might do for their target.’ The seventh respondent (R7) declared, ‘[We are] at the coalface of service delivery… to engage and to request for our inputs. I think it was not done properly.’ The fifth respondent (R5) said, ‘Basically, we hear NHI from the media; we haven’t really had people come in and talk to the people.’ Organisational culture within the department of health Some respondents depicted their absence of engagement and the one-sided narrative around NHI as a recurrent organisational practice. Other respondents felt disillusioned by the lack of senior managerial engagement and commitment to NHI: 8 The seventh respondent (R7) expressed, ‘But what can I say, because the government has decided they will implement what they have decided to do.’ The tenth respondent (R10) declared, ‘But really, I would not know why the department is not giving it the attention that it needs.’ Managerial Views on NHI Managerial views on NHI policies and documents A prevailing theme across all of the interviews was that respondents were unfamiliar with NHI documents, policy and law. This finding supports the overarching theme of poor managerial engagement described above. The second respondent (R2) asked, ‘on that… what was the meaning of that?’ The fourth respondent (R4) said, ‘I don’t know, I really don’t have an idea...’ The interviewer frequently needed to describe the content and concepts contained within the NHI policy document to the managers. Consequently, interviewees often lacked the insight to provide meaningful commentary. The fifth respondent (R5) said, ‘Maybe, I’m not so sure… So I don’t know, I don’t want to comment on something that I’m not so sure of… how is it going to work, or what.’ The sixth respondent (R6) stated, ‘I’m not sure by facility, or … by municipality, or the service provider… then to, I suppose…So I’m not too sure, I haven’t looked at the detail.’ The need to address social disparities Despite a lack of familiarity with NHI policies, most managers viewed NHI as a mandatory social intervention to redress social inequities and unequal access to quality healthcare. This was coupled with the impression that NHI would redistribute resources from the private health facilities to serve as a panacea to current public sector challenges: The first respondent (R1) maintained, ‘The issue of overcrowding on public institutions, I think it will also reduce that… which will actually reduce on the number of litigations.’ The first respondent (R1) added, ‘So that they are equally empowered to render services, or better services to our citizens.’ The fourth respondent (R4) declared, ‘I think NHI aims to ensure that the issue of social solidarity between the healthy and the sick, those who have and those who have not.’ 9 Views on the current healthcare system There was a singular deviant case that saw the NHI as a scheme that was not applicable to the South African setting. The participant regarded the current private health system as adequately suited to the South African context and felt that more meaningful interventions could be made towards improving existing public health services. Perceptions of NHI implementation Concerns around the current healthcare infrastructure and policies All participants regarded the current public healthcare system as not ready to implement NHI. Respondents expressed concerns around infrastructure that had aged and not been maintained, The first respondent (R1) maintained, ‘You cannot be sitting with a hospital that is built fifty years back… but 90% of them are very old, so they are not ready.’ The sixth respondent (R6) declared, ‘The second thing is, remember, our health information system, our epidemiological data, is limited; how do they begin to develop appropriate plans.’ Furthermore, respondents emphasized that the health system was not standardized and access to quality care varied across the country: The fifth respondent (R5) asked, ‘What about the areas where people still need to travel about a hundred kilometres to access a clinic?’ Respondents expressed several concerns around current performance and quality appraisal mechanisms. They saw current indicators as invalid and unreliable - often misrepresented to appease executives: The second respondent (R2) maintained, ‘There will be no questions to say, how did you reach this target, how did you do whatever. As long as you are presenting, to say, I’m at 55%, when the target was 50%.’ Views on readiness to implement NHI Most respondents felt disempowered and lacked clarity on their role in the NHI. They expressed concerns around poor quality and one-way communication, as well as not knowing what the requirements or procedures would be. Further, respondents described doubt around the readiness of 10 the health system to implement the NHI policies due to experiences with several long-standing policies and laws that had not yet been effectuated: The fifth respondent (R5) said, ‘We are still far, because even the tools for the hospitals… it’s still on the trial, on the draft.’ Concerns around corruption within the governance structures A major theme that emerged was that of corruption throughout all tiers of governmental structures. Distrust towards government seemed to be magnified when managers discussed the proposed increased fiscal expenditure on health in the NHI Bill: The third respondent (R3) affirmed, ‘Well, I definitely don’t think that the current structure should govern it.’ The fifth respondent (R5) declared, ‘When you look what is really happening in terms of government, there are so many corruptions, so many things, every day… when you listen to the news, it’s all about corruption, people who are enriching themselves.’ Respondents suggested a reform of managerial structures that fostered good governance, transparency and accountability: The ninth respondent (R9) said, ‘… but if ever you don’t have a consequence management approach, and accountability, things can just fall through, without anybody really paying attention or taking responsibility.’ Human resource challenges Respondents held that challenges around good governance were exacerbated by deficits in human resource management at national and provincial levels. Respondents expressed concerns around staff retention and ineffective devolution of management. Further, respondents described an absence of consensus orientation or succession planning towards NHI implementation: The first respondent (R1) said, ‘What we’re supposed to be doing now, is to do succession planning and align it to National Health Insurance.’ Several respondents highlighted challenges around understaffing within the district, resulting in overworked and demotivated employees: 11 The first respondent (R1) declared, ‘… and she said to me, it’s because of workload; I’m frustrated before nine o’clock, because I have to stand and attend to more than fifty people… then I ask in private, in private I’ll be allocated only five…’ Fragmentation and incoordination of services Several respondents expressed that healthcare service delivery systems functioned in a siloed fashion. Respondents felt that patient care was poorly coordinated between levels of care and referral pathways were poorly established. Further, beyond thehealth system, managers described the need for them to redirect energy and resources to ancillary supportive services to compensate for sub- optimal service delivery from other national and provincial departmental counterparts: The fourth respondent (R4) stated, ‘So we do have the generators now, but we do have lots of unforeseen problems, to ensure that there’s still uninterrupted electrical supply at the clinics.’ The sixth respondent (R6) said, ‘But apart from making sure that the service standards are complied to… we need to make sure that other systems are in place… adequate for the work we have to do for that particular community.’ Potential avenues to facilitate NHI implementation Participants emphasized the need for national-level government to facilitate the digitalisation of healthcare to map community profiles which would allow for tailored health service planning as well provide an avenue of communication to all stakeholders. Respondents also spoke to the development of a booking system to optimize the flow of patients throughout facilities and improve user experience: The third respondent (R3) maintained, ‘If we had to look at a first step, I would say, going electronic is very much needed… especially that we can start keeping track of our patients’ The eighth respondent (R8) said, ‘But if proper SMSs or appointments were sent… reduced waiting time will definitely improve services.’ Respondents held that national government needed to expand multilateral collaborative efforts alongside in-house managerial capacitation to nurture comprehensive and inclusive NHI implementation strategies that reduced duplication and fragmentation of health services: The first respondent (R1) declared, ‘… trained, specifically on National Health Insurance; and not from one discipline, it must be a multi-tasked or disciplinary team.’ The fourth respondent (R4) maintained, ‘So I think it’s to listen to the community, hear what all the stakeholders have to say… to what could be better solutions.’ 12 Lastly, managers felt that national government should iteratively appraise NHI strategies through shared international learning and reflection to guide efficient local NHI implementation: The sixth respondent (R6) declared, ‘I think we’re very fortunate in the sense that we can learn from lessons learnt in other countries.’ Discussion In this study we have described the evidence-based need for the Ministry of Health to raise the awareness, readiness and engagement of DMs to operationalize the pillars of UHC - effective health systems management is determined by a strategic managerial group identity across the levels of health service management. The NHI, as a radical reformation of the South African health sector, requires unison of vision, participation and consensus orientation, transparency and accountability. The NHI bill4 states that the DHS is responsible for forming and collaborating with implementation structures such as DHMOs and CUPS. Capacitated DHMOs and CUPS are the bedrock of strategic purchasing that aims to address health system fragmentation, improve healthcare infrastructure, as well as enable accountable and transparent financial management. The results of this study show that managerial engagement in NHI policy development and implementation remains21 inadequate. Managers failed to appreciate the implications of the NHI Bill at the operational unit of PHC – the DHS – and lacked insight into the decentralised nature of contracting and financial management in strategic purchasing, the use of capitation, or that private providers may be in the expected mix of personal services provided. The deficits in managerial awareness and capacity, coupled with their exclusion from engagement of policy formulation and rollout, is contradictory to the apparent purpose of the NHI Bill to empower the DHS to assume responsibility for the needs of predefined municipal areas. Joint learning from Asian,10 Latin American17 and African18,27,28 regions has shown that incomplete operationalization of the DHS hinders UHC implementation strategies15 and leads to wasteful expenditure. Conversely, success stories from countries such as Cuba,17 Thailand and Indonesia7 show that UHC can be successfully implemented in the face of fiscal austerity, when sustained investment in decentralised governance strategies that enabled CUPs and strategic purchasing. Similarly, several African countries such as Uganda and Ghana have effectively used decentralised governance strategies to facilitate UHC success.18 Notwithstanding the overall lack of familiarity with NHI policies, most managers viewed NHI as a mandatory social intervention to redress social inequities. However, there was an underlying notion amongst the respondents that NHI is a panacea to current public sector challenges by simple redistribution of resources from the private sector. This paradigm fails to appreciate the strategic and collaborative governance needed to address health system fragmentation and has resulted in corruption, wasteful expenditure and ultimately UHC failure in countries such as Kenya and Nigeria.12,14 Despite the abovementioned concerns, select studies have demonstrated that NHI intervention pilots have the potential to markedly improve health system performance2 within resource constraints. These achievements were most evident in districts where governance was effectively assigned to the respective DHS. It is clear that when the policies within the NHI Bill are effectively applied, definitive progress can be made towards the attainment of UHC. Sustained federal investment in the DHS is necessary to operationalise NHI policies and provide redress to the fragmented and inequitable healthcare system. 13 Study limitations Only one district (of seven) in Gauteng was sampled for this study - a large urban district that is well- capacitated and in close contact with national and provincial governing bodies. The findings of this study do not accurately reflect all districts of South Africa, especially rural areas. Recommendations This study should be repeated across the spectrum of rural and urban districts to corroborate findings and expand on contextual nuances for community-specific health system planning. Similarly, the qualitative findings of this paper could inform the constructs (by means of principal component analysis) for future quantitative questionnaires which, once validated, could be used for widespread evaluation of managerial decentralisation. Further, this could be coupled with explanatory models to guide strategic interventions targeted at healthcare governance. Recent studies have illustrated the efficacy of brief training interventions that empower DHS managers to effectively govern healthcare services.2,15 It is recommended that future strategies for NHI implementation include specific engagement and capacitation of the managers of all 52 districts within South Africa to prepare for contracting at that level. In the current (second) phase of NHI implementation, National policy should employ validated frameworks15,20,21 of health system governance and utilise existing bodies, such as the OHSC, to establish and enforce accurate indices of decentralised governance and health within each DHMO and their respective CUPS. Managerial capacitation could be targeted through human resource for health interventions7,10,11 such as managerial fellowships, mentoring, internships and succession-planning. Directly engaging DMs will coordinate simple top- down assumptions with bottom-up insights (into the complex realities of service delivery) to focus and synergise available resources towards improving health outcomes. Declarations Ethics approval and consent to participate This study is approved by the University of the Witwatersrand’s Human Research Ethics Committee (reference: M191046) as well as the Research Committee of the Johannesburg Health District (reference: GP_202006_048). All methods were carried out in accordance with the guidelines and regulations. Due to the COVID-19 pandemic, interviews were conducted with consenting participants by telecommunication software as regulated by the ethics committee. Data availability statement The dataset of this work is available in the Zenodo repository, https://doi.org/10.5281/zenodo.4765125, and can be accessed on reasonable request to the corresponding author. Consent for publication Not applicable. Conflict of Interests None. 14 Acknowledgements The corresponding author would like to thank all of the district managers who willingly agreed to participate in this study. Author Contributions Funding sources None. 15 References: 1. Bradshaw D, Nannan NN, Wyk VP, Laubscher R, Groenewald P, Dorrington RE. Burden of disease in South Africa: Protracted transitions driven by social pathologies. S Afr Med J. 2019 Dec 5;109(11b):69–76. 2. Day C, Zondi T. Measuring National Health Insurance: towards Universal Health Coverage in South Africa. South Afr Health Rev. 2019;2019(1):55–68. 3. Katuu S. Transforming South Africa’s health sector: The eHealth Strategy, the implementation of electronic document and records management systems (EDRMS) and the utility of maturity models. J Sci Technol Policy Manag. 2016 Oct 3;7:330–45. 4. National Health Insurance Bill B11-2019 | South African Government [Internet]. [cited 2019 Aug 25]. Available from: https://www.gov.za/documents/national-health-insurance-bill-b-11-2019-6- aug-2019-0000 5. WHO | South Africa [Internet]. WHO. [cited 2019 Mar 11]. Available from: http://www.who.int/countries/zaf/en/ 6. Coovadia H, Jewkes R, Barron P, Sanders D, McIntyre D. The health and health system of South Africa: historical roots of current public health challenges. Lancet Lond Engl. 2009 Sep 5;374(9692):817–34. 7. Yeoh E-K, Johnston C, Chau PYK, Kiang N, Tin P, Tang J. Governance Functions to Accelerate Progress toward Universal Health Coverage (UHC) in the Asia-Pacific Region. Health Syst Reform. 2019 Jan;5(1):48–58. 8. Greer SL, Méndez CA. Universal Health Coverage: A Political Struggle and Governance Challenge. Am J Public Health. 2015 Jul 16;105(S5):S637–9. 9. Amado LA, Christofides N, Pieters R, Rusch J. National health insurance: A lofty ideal in need of cautious, planned implementation. South Afr J Bioeth Law. 2012 Jun 14;5(1):4-10–10. 10. Hort K, Jayasuriya R, Dayal P. The link between UHC reforms and health system governance: lessons from Asia. J Health Organ Manag. 2017 May 15;31(3):270–85. 11. Brinkerhoff DW, Bossert TJ. Health governance: principal–agent linkages and health system strengthening. Health Policy Plan. 2014 Sep 1;29(6):685–93. 12. Fryatt R, Bennett S, Soucat A. Health sector governance: should we be investing more? BMJ Glob Health. 2017 Jul 1;2(2):e000343. 13. OHSC 2016-17 ANNUAL INSPECTION REPORT [Internet]. Office of Health Standards Compliance. [cited 2019 May 21]. Available from: http://ohsc.org.za/ohsc-media-response-to-the- article-by-the-business-day/ohsc-2016-17-annual-inspection-report-2/ 14. Dodd R, Palagyi A, Jan S, Abdel-All M, Nambiar D, Madhira P, et al. Organisation of primary health care systems in low- and middle-income countries: review of evidence on what works and why in the Asia-Pacific region. BMJ Glob Health. 2019 Aug 1;4(Suppl 8):e001487. 15. Fusheini A, Eyles J. Achieving universal health coverage in South Africa through a district health system approach: conflicting ideologies of health care provision. BMC Health Serv Res. 2016 Oct 7;16(1):558. 16 16. Reich MR, Harris J, Ikegami N, Maeda A, Cashin C, Araujo EC, et al. Moving towards universal health coverage: lessons from 11 country studies. The Lancet. 2016 Feb 20;387(10020):811–6. 17. Atun R, de Andrade LOM, Almeida G, Cotlear D, Dmytraczenko T, Frenz P, et al. Health- system reform and universal health coverage in Latin America. The Lancet. 2015 Mar;385(9974):1230–47. 18. Ifeagwu SC, Yang JC, Parkes-Ratanshi R, Brayne C. Health financing for universal health coverage in Sub-Saharan Africa: a systematic review. Glob Health Res Policy. 2021 Mar 1;6(1):8. 19. Langlois EV, McKenzie A, Schneider H, Mecaskey JW. Measures to strengthen primary health- care systems in low- and middle-income countries. Bull World Health Organ. 2020 Nov 1;98(11):781– 91. 20. Ramesh M, Wu X, Howlett M. Second Best Governance? Governments and Governance in the Imperfect World of Health Care Delivery in China, India and Thailand in Comparative Perspective. J Comp Policy Anal Res Pract. 2015 Aug 8;17(4):342–58. 21. Marais DL, Petersen I. Health system governance to support integrated mental health care in South Africa: challenges and opportunities. Int J Ment Health Syst. 2015 Mar 11;9(1):14. 22. Bidwell P, Laxmikanth P, Blacklock C, Hayward G, Willcox M, Peersman W, et al. Security and skills: the two key issues in health worker migration. Glob Health Action. 2014 Dec 1;7(1):24194. 23. Poppe A, Jirovsky E, Blacklock C, Laxmikanth P, Moosa S, Maeseneer JD, et al. Why sub- Saharan African health workers migrate to European countries that do not actively recruit: a qualitative study post-migration. Glob Health Action. 2014 Dec 1;7(1):24071. 24. Vasileiou K, Barnett J, Thorpe S, Young T. Characterising and justifying sample size sufficiency in interview-based studies: systematic analysis of qualitative health research over a 15-year period. BMC Med Res Methodol. 2018 Nov 21;18(1):148. 25. MAXQDA: Qualitative Data Analysis Software | Windows & Mac [Internet]. MAXQDA - The Art of Data Analysis. [cited 2019 Aug 25]. Available from: https://www.maxqda.com/ 26. Srivastava A, Thomson SB. Framework Analysis: A Qualitative Methodology for Applied Policy Research [Internet]. Rochester, NY: Social Science Research Network; 2009 Jan [cited 2019 May 22]. Report No.: ID 2760705. Available from: https://papers.ssrn.com/abstract=2760705 27. Sakyi EK, Adzei FA, Atinga RA. Managerial problems of hospitals under Ghana’s National Health Insurance Scheme. Clin Gov Int J. 2012 Jul 27;17(3):178–90. 28. Adeloye D, David RA, Olaogun AA, Auta A, Adesokan A, Gadanya M, et al. Health workforce and governance: the crisis in Nigeria. Hum Resour Health. 2017 May 12;15(1):32. 17 Appendix I: Semi-structured interview protocol For the Researcher: Name of Researcher: Shane Darren Murphy Participant/Interview Number: Participant Age: Participant Gender: Participant Ethnic Group: Participant Religion: Participant Qualifications: Participant Role: Guidelines for each Interview: • Greet participant and introduce yourself directly to the participant. Discuss your purpose (manage expectations), and establish rapport. Only the researcher and participant should be present. Run through the consent form with the participant to ensure he/she understands all of the study’s details. • Maximum interview duration of 1 hour to avoid researcher/participant fatigue. • Location: Office, or other chosen quiet location of participant. • Timing: During office times at an agreed upon time between researcher and participant. • Jot down some important notes during the interview that interpret emotional or ethnographic milieu. Ensure that the audio-recorder is working/batteries are charged and operational. • Pay full attention to the participant. Be respectful. • It is critical to enable the interviewee to raise own issues. Ask open-ended questions with follow-up, probing, specifying and interpreting questions. • Remember that the template is not a questionnaire, but an agenda for the interview guide. Interview questions (numerical) and prompts (alphabetical): 1. What are your views on National Health Insurance? a. Can you tell me your views of the NHI green/white papers or the bill? b. What are views on a single pool of funding (where funds are pooled together to contract health services)? c. What are your views on strategic purchasing (separation of funder (purchaser) well as the provider (sub-district public services), from the department of health and the NHI contracts directly with you, the provider)? d. What are your views of the Contracting Unit for Primary care? e. What are your views on contracting with the private sector? f. What are your views on capitation (per capita) payment? 2. Can you tell me about your engagement in NHI policy development? a. What is the engagement you have had from province/national government? 18 b. How do you get your information on the NHI? 3. What is your perception about the implementation of NHI? a. What is your perception about the readiness of the overall public service to implement NHI? b. What is the readiness of the district health service to implement PHC contracting? c. What is your view on readiness of contracting at a subdistrict level? d. What is the best way to organise PHC services to contract with the NHI? e. What are the challenges of the district health service managers to cope with contracting in the NHI? f. What are your solutions for district health service managers to cope with contracting in the NHI? 4. Is there any other view or thought you would like to express? Closure: Thank the participant for their time. 19 Appendix II: Participant information sheet and informed consent forms STUDY INFORMATION DOCUMENT Study title: The views of public service managers on the implementation of national health insurance in primary care. Greeting: To the Manager at a District or Sub-District level in Gauteng Province Introduction: I, Shane Darren Murphy, am doing qualitative research on the views, engagement and readiness of district and sub-district managers on the implementation of NHI at the primary care level. Research is a process used in seeking new knowledge. This study intends to explore the views and perceptions of public service managers on NHI implementation. Invitation to Participate: I am asking you to take part in a semi-structured interview. What is involved in the study: This could include but would not necessarily be limited to such features as: 1. A single one-on-one interview of 45-60 minutes 2. Audio-recordings will be made of the interview for transcription and data analysis. Risks of being involved in the study: There are no foreseen risks to yourselves or other clinicians as no identifying data will be collected. Should you feel vulnerable or victimized at any stage, please contact the lead researcher, Shane Murphy at murphy.shanedarren@gmail.com or the study supervisor, Professor Shabir Moosa, at shabir@profmoosa.com. Benefits of being in the study: In this study, as in most studies, there is no direct benefit to yourself as the participant. This is an opportunity to improve insight into the views of, and engagement with public service managers in implementing NHI. The Participant will be given pertinent information on the study while involved in the project and after the results are available. Participation is voluntary. Please note that refusal to participate will involve no penalty or loss of benefits to which you are otherwise entitled as a primary clinician. You may discontinue participation at any time without penalty, or loss of benefits to which you are otherwise entitled. There is no requirement to provide a reason for withdrawing and any data collected on such a person will in default be destroyed, unless you specifically consent to its retention. No Reimbursements for participation in this Study will be provided. 20 Confidentiality: Personal information will be treated in the strictest confidence and will only be available to the principal investigator (Shane Murphy) and his supervisor (Professor Shabir Moosa). The only exceptions - and all of them are rare - would normally be: 1. personal information may be disclosed if required by law 2. the Human Research Ethics Committees of the University may exceptionally require personal data to respond to a formal complaint, or for a compliance audit. All data collected in the course of the study will be securely retained for two (2) years, if a scientific publication arises from the study and six (6) years, if there is no publication. Thereafter it will be destroyed accordingly. Anonymity of data collected from interviews and audio-recordings will be maintained. Contact details of researcher: Shane Darren Murphy Email: murphy.shanedarren@gmail.com Contact Number: 074 893 1153 Outputs Framework analysis will be conducted on the data corpus through Computer-Assisted Qualitative Data Analysis Software. Contact details of HREC administrator and chair: This study has been approved by the Human Research Ethics Committee (Medical) of the University of the Witwatersrand, Johannesburg (“Committee”). A principal function of this Committee is to safeguard the rights and dignity of all human subjects who agree to participate in a research project and the integrity of the research. If you have any concern over the way the study is being conducted, please contact the Chairperson of this Committee who is Professor Clement Penny, who may be contacted on telephone number 011 717 2301, or by e-mail on Clement.Penny@wits.ac.za. The telephone numbers for the Committee secretariat are 011 717 2700/1234 and the e-mail addresses are Zanele.Ndlovu@wits.ac.za and Rhulani.Mukansi@wits.ac.za 21 PARTICIPANT CONSENT SHEET: Interview The views of public service managers on the implementation of national health insurance in primary care. 1. I have been given a Participant Information Sheet which explains the nature and processes involved in this study, which is attached hereto; 2. I was given time to read it, or had it read to me, in the language I best understand; 3. I was given time to ask any questions I wanted to and found any answers given to me to be reasonable and satisfactory; 4. I believe I fully understand why the study is being conducted and what the intended outcomes will be; 5. I understand that there will be no immediate benefit to me, should I agree to participate, nor will I receive any payment; conversely, participation will not cost me anything but my time; 6. I understand that, even if I initially consent to take part in the study, I may subsequently withdraw at any time and would not be required to give any reasons; if that happened, any data collected about me for the purposes of the study would immediately be destroyed, unless I give consent for it to be retained 7. I have been given a range of contact details, listed below. If I require further information or become concerned about any aspect of this study I am free to speak to any of these contacts. Contact details of researcher: Shane Darren Murphy Email: murphy.shanedarren@gmail.com Contact Number: 074 893 1153 Contact Details of the supervisor: Professor Shabir Moosa Email: shabir@profmoosa.com Contact Number: 082 446 6825 Professor CB Penny, Chairperson of the Human Research Ethics Committee (Medical) at the University of Witwatersrand, on telephone no. 011 717 2301, or by e-mail at Clement.Penny@wits.ac.za. Ms. Z Ndlovu or Mr Rhulani Mkansi, Committee Secretariat, telephone nos.: 011 717 2700 or 1234, or by e-mail at: Zanele.Ndlovu@wits.ac.za or Rhulani.Mkansi@wits.ac.za Name of Participant: ________________________________________ Date: ________________________________________ Place: ________________________________________ Signature ________________________________________ 22 Witnessed by: Name of Witness: ________________________________________ Signature: ________________________________________ Date: ________________________________________ 23 PARTICIPANT CONSENT SHEET: Audio-Recording Study Title: The views of public service managers on the implementation of national health insurance in primary care. I have been given a Participant Information Sheet which explains the nature and processes involved in this study, which is attached hereto. 1. I was given time to read it, or had it read to me, in the language I best understand; 2. I was given time to ask any questions I wanted to and found any answers given to me to be reasonable and satisfactory; 3. I believe I fully understand why the study is being conducted and what the intended outcomes will be; 4. I understand that there will be no immediate benefit to me, should I agree to participate, nor will I receive any payment; conversely, participation will not cost me anything but my time; 5. I understand that, even if I initially consent to take part in the study, I may subsequently withdraw at any time and would not be required to give any reasons; if that happened, any data collected about me for the purposes of the study would immediately be destroyed, unless I give consent for it to be retained 6. I have been given a range of contact details, listed below. If I require further information or become concerned about any aspect of this study I am free to speak to any of these contacts. Contact details of researcher: Shane Darren Murphy Email: murphy.shanedarren@gmail.com Contact Number: 074 893 1153 Contact Details of the supervisor: Professor Shabir Moosa Email: shabir@profmoosa.com Contact Number: 082 446 6825 Professor CB Penny, Chairperson of the Human Research Ethics Committee (Medical) at the University of Witwatersrand, on telephone no. 011 717 2301, or by e-mail at Clement.Penny@wits.ac.za. Ms. Z Ndlovu or Mr Rhulani Mkansi, Committee Secretariat, telephone nos.: 011 717 2700 or 1234, or by e-mail at: Zanele.Ndlovu@wits.ac.za or Rhulani.Mkansi@wits.ac.za Name of Participant: ________________________________________ Date: ________________________________________ Place: ________________________________________ Signature. ________________________________________ 24 Witnessed by: Name of Witness: ________________________________________ Signature: ________________________________________ Date: ________________________________________ 25 Appendix III: Ethics approval – Human Research Ethics Committee (University of Witwatersrand) 26 Appendix IV: Ethics approval from district (National Health Research Database) 27 28 Appendix V: Research proposal The Views of Public Service Managers on the Implementation of National Health Insurance in Primary Care Shane Darren Murphy Family Medicine Registrar Research proposal for a degree by coursework and research report 2019 Supervisor: Prof S. Moosa 29 DECLARATION FORM I, Shane Darren Murphy, declare that this Research Report is my own, unaided work. It is being submitted for the Degree of Master of Medicine at the University of the Witwatersrand, Johannesburg. It has not been submitted before for any degree or examination at any other University. ____________________________________ Shane Darren Murphy On this 4th day of March 2019 30 Table of Contents Content Page Number List of Tables 1 List of Figures 2 List of Abbreviations 3 1. Introduction 4 2. Background Literature Analysis and Critique 5 3. Developing the Research Question 10 4. Research Objectives 10 5. Research Methodology and Study Design 10 6. Ethics 16 7. Timing 16 8. Funding 17 9, References 19 Appendices Appendix A: Research Process Flow Chart 23 Appendix B: Interview Schedule 24 Appendix C: Consent Forms 27 31 Nomenclature List of Abbreviations Abbreviation or Acronym Full Form HIV Human Immune-Deficiency Virus GDP Gross Domestic Product UHC Universal Health Coverage SDGs Sustainable Development Goals NHI National Health Insurance WHO World Health Organisation NHS National Health Scheme PICO Population, Intervention, Comparison, Outcome CAQDAS Computer-Assisted Qualitative Data Analysis Software HREC Human Research Ethics Committee DHMO District Health Management Office CUP Contracting Unit for Primary Care 32 List of Figures Figure Page Number Figure 1 Gantt Chart 17 33 List of Tables Table Page Number Table 1 Criteria for Qualitative Trustworthiness 11 Table 2 Categories of Data Management 13 Table 3 Task Chart 16 Table 4 Budget for Funding 17 34 INTRODUCTION South Africa is an upper middle-income country in the developing world suffering a crippling quadruple burden of disease.1 The current health system is overwhelmed with the highest level of Human Immuno-deficiency Virus (HIV) infections in the world,2 800% of the global average of injury-related deaths,3 high levels of infant and maternal mortality, and a growing burden of noncommunicable diseases - in-keeping with the delayed contemporary epidemiological model of transition.4 Aggravating the current health crisis is South Africa’s two-tiered public and private health systems - institutionalised through a history of ethnic and socioeconomic deprivation and marginalization. Socio-economic inequities perpetually disadvantage the poor who are disproportionately affected by South Africa’s quadruple burden of disease with catastrophic household expenditure on health. To combat this growing burden of disease, as well as the gross disparity and social inequity across the country, South Africa is currently implementing a new financing and service delivery system - National Health Insurance (NHI) - for universal health coverage (UHC) in South Africa. UHC is advocated by the World Health Organisation for governments to improve health levels, reduce the burden of disease, cross-subsidise risk and enhance social equity.5 UHC falls under Target 3.8 of the United Nations Sustainable Development Goals (SDG's) - an overarching target that supports the other objectives of SDG 3.6 UHC is a laudable social construct, however, South Africa has failed several previous endeavours to implement structural and financial health reforms.7 The stumbling block common to all these endeavours was the inability to transform well-constructed policy into practicable plans of action. The NHI White Paper of 2018 acknowledges the failures of transition from theoretical frameworks to workable plans and highlights the need for engaged, delegated and accountable management to successfully implement NHI.7 The role of district and sub-district level managerial awareness, engagement and alignment with the principles, as well as the processes, of NHI implementation has been highlighted in the analysis of several countries implementing UHC.5,8–11 The NHI Bill of 2019 prioritises primary health care (PHC) strengthening by delegating greater responsibilities to district- and sub-district level managers to facilitate appropriate decision-making and accountability for health service provision. The aim of this study is to explore the district and sub-district level managers views on NHI, their engagement in policy development, and their perceptions of readiness to implement NHI. BACKGROUND LITERATURE ANALYSIS AND CRITIQUE The plan for UHC in South Africa In South Africa, UHC has taken the form of the National Health Insurance scheme (NHI). The NHI policy is rooted in the South African constitutional mandate to achieve a progressive realisation of the right to healthcare.7 The NHI will create a single financial pool for national risk-pooling and cross-subsidisation. The NHI fund proposes to generate funds through an increased GDP expenditure on health, new payroll taxes, increased income tax as well as increased value-added tax. The NHI fund will serve as a monopsony purchaser with a 35 purchaser-provider split and strategic purchasing of health services to increase equitable population coverage and service access while containing costs.7 The NHI policy, as per the White Paper 2018, consists of three phases spanning 14 years.7 During the second phase of implementation (2017-2022), the NHI Policy and Bill emphasise delegated managerial authority to district and sub-district managers to enhance the oversight of NHI implementation in primary health care. The National Health Act of 2004 created the structures of our health system with national and provincial departments of health.7 While creating health districts, the Act vests operational control of districts in provinces, with district managers appointed by provinces and local government facilities and staff being ‘provincialised’ - i.e. being moved under provincial control. There is fair devolution of operational control to district and sub-district level, depending on local capabilities and political tensions. Metropolitan municipalities have resisted 'provincialisation', citing better management capabilities in local government-run facilities than provincial government-run health facilities.7 In dealing with the challenges of provincial management, the NHI Bill creates District Health Management Offices (DHMOs) as health district structures accountable to the National Department to facilitate and coordinate PHC services at the district level. The DHMOs are expected to establish Contracting Units for PHC Services (CUPS) that are the preferred organisational unit (comprising an integrated network of district hospital, clinics or community health centres, and private providers) with which the NHI fund will directly contract to provide PHC services in a subdistrict.7 These ±200 CUPS are expected to identify certified and accredited public and private providers to contract and fund PHC services. The NHI Bill states that CUPs will possibly receive capitation payments directly at the sub-district level7 - as opposed to current line-item budgets. District and sub-district health system managers play a vital role in the governance of structures and processes that facilitate service provision within the current healthcare system. Further, the proposed plan in the NHI bill to decentralise health service decision-making and contracting to the level of the DHMO highlights the need for district and sub-district managers to be fully aware of and engaged in the NHI structures, processes and rollout plan. The proposed activities are critically dependent on decentralized managerial autonomy to effectively oversee NHI action plans. International experience with decentralized management and UHC implementation The National Health Scheme (NHS) in the United Kingdom is often thought of as the archetypal UHC system.12 Founded in 1948, the NHS has significantly improved national health and economic outcomes. Conventionally, the unit of contracting within the NHS was per solo private general practitioner. National experience with this system has shown that the large gap between purchaser and provider within the NHS has driven inequity amongst communities. Further, the previously centralized purchasing strategy has created a disconnect between the purchaser and the needs of the community, i.e. centralised purchasing units are not sensitive to local needs. To counter this, accountable and delegated management has become a focus of conversations as stakeholders have attributed the shortcomings of the NHS to poor local management.12 There are growing and consolidated efforts to tackle the five areas through which health management has adversely affected the whole system: accountability, participation, transparency, integrity and policy-making capacity.13 Stakeholders maintain that accountable and delegated management is the greatest predictor of equitable UHC implementation.12 The NHS has subsequently begun to reinforce a ‘strategic 36 vision’ at the level of the district organizational structures of the NHS, as well as capacitate local health managers to improve their ability to enforce implementation.13 The current Canadian healthcare system (Medicare) originated in a single Canadian province with the aim of providing care based on need rather than ability to pay.14 The concept of cost- sharing and co-ordinated local health teams gradually spread across the country and was standardized in federal law in the Canadian Health Act of 1984. The origins of this healthcare system have resulted in highly decentralised administrative and healthcare service structures providing Canada with some of the best health outcomes in the world. Currently, local health activists are calling on policy makers to renew the focus on decentralised community- centered care in all parts of the country to reduce the health inequities experienced by marginalized groups. Hort et al.10 conducted a systematic review of UHC implementation in eight countries across Asia. Inadequate health systems management was found to be a crippling factor to UHC success. The authors noted eight core areas of deficiency: participation, consensus orientation, transparency, accountability, responsiveness, efficiency, equity and inclusivity and legal frameworks. Thailand, however, showed a marked reduction in out-of-pocket payments made by households, improvement in health coverage interventions such as antiretroviral therapy and significantly reduced gaps in child mortality.9 Thailand used primary care units (CUPs) consisting of a doctor, a nurse and a community health worker to care for entire families in teams with the doctor at the district hospital responsible for the family. 10The CUPs served as gatekeepers to the greater healthcare system. Each CUP (a primary care centre and a district hospital) shared the outpatient budget for its population. Low costs were facilitated by low wage costs, reliance on public providers and strong work- ethic. Authors have attributed the success of UHC in Thailand, a nation with a low gross national income per capita, to federal investment in organisational capacity and management at the district level of healthcare.9 A key finding of the ‘decentralization’ approach used in Indonesia was that, although consistently driven by domestic political interests, it opened the way for municipal policy experimentation tailored to local the local context.10 This enabled increased attention on community needs with targeted and cost-effective interventions. A scoping review of health system reform and UHC in Latin America showed gaping disparities in health between Latin American countries.15 The only countries that had implemented UHC were Brazil, Cuba and Costa Rica. These countries had markedly improved health outcomes in the face of multiple sociopolitical and economic challenges that the authors attributed to the improved organisation and management of health systems after UHC implementation.15 Brazil’s central government funds municipalities that directly contract family health teams to deliver care to defined populations. This system has increased the access to care of marginalised groups and improved their health outcomes significantly.15 However, a plateau in the uptake of UHC is linked to poor quality referrals and a lack of complementarity between the public and private sector.15 Conversely, the decentralized Cuban health care system that relies predominantly on CUPs provides Cuba with some of the best health outcomes globally. Further, the decentralized management of healthcare in Cuba has shown to focus attention on the needs of the community, speed up development programmes, increase equity of care and promote management flexibility and accountability.15 While this review highlights the favourable health system outcomes associated with addressing structural fragmentation and decentralising decision-making to the district level, it further infers a welcome propensity to focus on management associated with UHC implementation.15 37 In an attempt to address sub-optimal health service delivery in Nigeria, the National Health Act was signed into law in 2014 to facilitate the progressive implementation of UHC.16 While this change is still in the grassroots phase, critics have claimed that policy failure has already occurred despite multilateral collaborative partnerships to facilitate its implementation. This has resulted in ongoing healthcare crises due to human resources, supply chain and infrastructure deficits at the district level. The authors describe the in-coordination of services as well as the lack of a shared strategic vision and plan amongst managers as key reasons for failure.16 Further, the authors claim that a lack of planning when decentralising management resulted in a lack of organizational control and co-ordination with conflict arising from the unclear assignment of authority as well as personal ambitions. Decentralised management in South African primary care A systematic review by Fusheini and Eyles17 showed the vital role that the district health system, as a decentralised mechanism of healthcare administration, had on the outcome of the NHI pilot projects. The authors showed a strong correlation between levels of decentralised management to the district level with improvements in service availability and readiness, equity and coverage in all pilot districts. Qualitative research by Marais and Pertersen18 showed that new policies supporting the incorporation of mental health care into primary level care were impeded by the inadequate managerial capacity at district and provincial levels. The 2010 Negotiated Service Delivery Agreement identified health system strengthening as a principal element in South Africa’s healthcare reform.19 However, there has thus far been minimal open discussion about this.20 Currently, health system management has been viewed as near-absent with critics citing that the biggest obstacle to NHI implementation is its dependence on government oversight at a detached and centralized level.20 The most recent report by the Office of Health Standards Compliance showed that primary healthcare clinics yielded a 47 % compliance across seven domains of quality.21 The domain of leadership and governance scored 47% - classified as non-compliant and requiring urgent intervention and complete re-inspection. Leadership corruption is a prominent concern for NHI Implementation. While formal research on corruption in South Africa is sparse, some provinces have had nearly 45% of their health care expenditure audited as irregular. Authors claim that poor management and corruption have a reciprocal relationship and propose that capacities and competencies for implementation should be addressed as a matter of urgency.22 Decentralising management to the district levels will facilitate the transparency and accountability of local health administration. Public health district and sub-district level managerial engagement with NHI An extensive search of multiple databases (PubMed, Academic Search Premier and EMBase) with extension to Google Scholar returned no international or local literature on the awareness, readiness and engagement of district public health service managers prior to or during UHC implementation. Summary The above literature shows the need to align the awareness, readiness and engagement of district management by the Ministry of Health, responsible for implementation with national 38 policy and legislation. Effective health systems management is determined by a strategic managerial group identity across the levels of health service management. The NHI, as a radical reformation of the South African health sector, requires unison of vision, participation and consensus orientation, transparency, accountability and the eradication of corruption. In the South African context, poor leadership, corruption, lack of accountability, fragmentation and incoordination within and between governmental departments is a prominent concern for the failure of the successful implementation of the NHI. District and sub-district level management is fundamental to the mechanics of successful NHI implementation in South Africa. DEVELOPING THE RESEARCH QUESTION Qualitative research mandates the use of what, why and how questions that focus on 'process'23 as research questions emerge throughout the ensuing research project. Research Question What are the views, experiences of engagement and perceptions of readiness for NHI implementation amongst district and sub-district managers of primary care in Gauteng Province, South Africa? RESEARCH OBJECTIVES 1. To explore district and sub-district level public service manager’s views on NHI. 2. To explore the experience of district and sub-district level public service manager’s engagement in NHI policy development. 3. To explore district and sub-district level public service manager’s perceptions of managerial and public sector readiness for the implementation of NHI. 4. To explore district and sub-district level public service manager’s perceptions of potential challenges as well as solutions to effectively implementing NHI. RESEARCH METHODOLOGY AND STUDY DESIGN A research paradigm of interpretivism has been employed to explore the value-laden and subjective views of managers in this setting. Interpretivism uses the inductive reasoning of qualitative methodology to analyse nuanced contextual phenomena. Phenomenology was chosen to guide the research as it analyses and typologically classifies subjectively experienced social phenomena. Maintaining Scientific Rigour Qualitative research has come under scrutiny from quantitative proponents stating that its methodology is neither sufficiently systematic nor rigorous to serve as scientific evidence.24 The rebuttal to this is the concept of Trustworthiness – a four-component strategy that parallels quantitative descriptors of scientific. The table below expands on these four components as elucidated by Shenton24 in relation to this research topic. 39 Table 1: Criteria for Qualitative Study Trustworthiness Quantitative Criterion Component of Trustworthiness Description Measures to Improve Rigour Internal validity Credibility Research methodology integrity. This study will use the validated Framework Analysis method. The researcher is familiar with the sample population. Critical reflexivity will be used during interviews with thick descriptions of data to facilitate internal validity. External validity Transferability Application of results to alternate settings. Adequate detail of the research setting with an audit trail will be provided for future research and cross comparison. Objectivity Confirmability Removal of the researcher’s predispositions from the analysis of the data. Computer assisted qualitative data analysis software will be used to leave a paper trail alongside reflexive commentary. Quotes will be used to illustrate participant views without researcher influence. Reliability Dependability The replicability of results if the study were repeated. Study design and methods are made explicit so that researchers can emulate the study design. Research methods Semi-structured interviews will be used for intimate exploration of the participant’s unique perception. Focus groups will not be employed due to practical limitations of limited number and dispersed participants as well as the potential for skewed power relations between members of the group. Observational studies are not tenable and will have low data-yield efficiency. Secondary data analysis is not possible due to the absence of existing data. Sampling Purposive, stratified random sampling will be used in this research project. The target population is district and sub-district level public service managers in Gauteng province in South Africa. All district and sub-district managers within Gauteng province will be included. The source population will be identified from the Human Resources register of the Gauteng Department of health. No exclusion criteria will be applied. To ensure the adequate representation of the target population, managers will be classified into either ‘male’ or ‘female’; as well either ‘sub-district’ or ‘district’ level manager. The study population (samples) will be randomly selected in a stratified and proportionate fashion from each pool (see figure 5.1 below). Purposive sampling will facilitate data rich samples. District and sub- district managers will be contacted individually and requested to engage in a semi-structured interview at their offices. 40 Figure 1: Proposed stratified random sampling strategy Sample Size No absolute criteria exist for the determination of sample size within qualitative research with a broad range suggested.25 Data saturation will serve as a pragmatic strategy to guide the final sample size although the sample will likely consist of 10-20 participants. Data collection Semi-structured interviews will be the method of data collection. The interview schedule (Appendix B) will facilitate exploration of participant perspectives while ensuring the attainment of research objectives. The interview schedule will incorporate anonymised demographic details. Interviews will be recorded by digital dictaphone to allow for optimal recall and analysis of interview content. Separate consent will be obtained for interview and recording. The audio recordings (to reduce bias) will be transcribed into a Microsoft document and cross-checked by the supervisor against recordings. Transcripts will be returned to participants for member checking.26 Data Management Strong emphasis is placed on the management of data within this research proposal to enhance analysis, reduce workload and ensure participant confidentiality with adherence to ethical guidelines. The table below details how data will be managed against the four categories of data management.25 Table 2: Categories of Data Management Category Data Management Type of data An audio recorder will be used to capture verbal data collected in the interviews in digital format and subsequently transcribed into a workable Microsoft Word format for further analysis in Computer-assisted qualitative data analysis software. Storage All audio and Microsoft Word files will be password protected and stored on the researcher's personal computer as well as a university-linked Google Drive folder. Only the researcher and the researcher's supervisor, upon request, will have access to this data. Cataloguing and discovery Anonymity and confidentiality will be maintained through the numerical cross labelling of individual interviews (recordings and transcriptions). 41 Personal data will be limited to age, gender, ethnicity and qualifications. Identifying data such as name, surname and date of birth will be omitted. Archiving and deposit Once the research project is completed, the collected data will be stored on the researcher’s personal computer and the above-mentioned Google Drive account for six years, as required by the ethics council of the University of the Witwatersrand. Qualitative Data Analysis There are multiple methods of data analysis in qualitative research. Thematic analysis is based on phenomenological inductive reasoning. Iterative processes enable robust content description and the identification of themes within the transcribed material. Content analysis, on the other hand, applies similar thematic identification throughout the data but is dependent upon the quantification of these coded themes to ascribe meaning to the data. While content analysis is viewed favourably by proponents of quantitative methodology, qualitative purists hold that the value of a theme is not related to its frequency within the dataset.25 Framework analysis, which will be used in this research proposal, builds on thematic analysis. Framework analysis is particularly useful for applied policy research and allows contextual, diagnostic, evaluative and strategic analysis of policy implementation.27 Themes identified in the data will be iteratively coded and categorised into progressively larger categories to form themes. An objective coder from the faculty of Health Sciences at the University will be approached to reduce researcher bias. Steps in Framework Analysis27 'Familiarisation' will be the first step in the data analysis process. The researcher will immerse himself in the audio recordings and transcriptions in an iterative fashion to identify nuanced patterns of the motivation within the participants. Secondly, thorough data mining will identify codes and themes within the dataset. Thirdly data will be coded and indexed both vertically within and horizontally between interviews with the aid of Computer-Assisted Qualitative Data Analysis Software (described below). Indexing will allow the comparison of perceptions between participants. Next, a thematic framework will be allowed to emerge from the data which collates themes from the transcripts to synthesise answers to the research questions. Lastly, mapping and interpretation will be used to categorise, contrast and compare themes to delineate interconnections, agreements and disagreements between datasets. Data will be represented in the form of tables and mind maps developed in CAQDAS software, as well as a written discussion. Computer Assisted Qualitative Data Analysis Software Computer-assisted qualitative data analysis software (CAQDAS) is a relatively novel technology that enables organised and coherent data analysis within qualitative research. Further, CAQDAS leaves an explicit audit trail proving the validity and reliability of the research. CAQDAS will not be used to mine textual data for themes as theme identification is inherently nuanced and requires continuous critical reflection. MaxQDA28 will be used to enhance the methodological rigour of data analysis as well as provide an audit trail. 42 Research Process Flow Chart A flow chart highlighting the research process can be found under Appendix A. ETHICS No statutory or professional regulatory body codes are breached by this research proposal. The Participant Information Sheet and Informed Consent forms for interviews as well as audio-recordings can be found under Appendix C. They are based on formats proposed by the Human Research Ethics Committee of the University of the Witwatersrand. Application will be made to the University of Witwatersrand Human Research Ethics Committee for approval of the research proposal prior to any data collection. Concurrent application will also be made to the National Health Research Database and district research committees. Both centres will serve to ensure ethical integrity of this proposal. Full informed consent, taken in person, will allow the attainment of study objectives while ensuring participant protection. The intimate nature of the interview (between participant and researcher) poses the possibility of the skewed power relationship. A “critical consciousness”,29 defined as the awareness and continuous deliberation of researcher and participant interaction during interviews, will enable reflexive adherence to ethical principles at the micro-level of this research project. TIMING The Gantt Chart and figure below illustrates the proposed timing of this research project. Table 3: Task Chart Task Start Date End Date Duration (Days) Task One: Formalisation of Research Proposal 2019/08/31 Task Two: Postgraduate Faculty Review 2019/09/01 2019/09/30 30 Task Three: Formal HREC and NHRD submission 2019/10/01 2019/11/30 60 Task Four: HREC feedback, corrections and re- submission 2019/12/01 2019/12/31 30 Task Five: Data Collection 2020/01/01 2020/01/31 30 Task Six: Data Analysis 2020/03/01 2020/03/28 27 Task Seven: Draft Submission to Supervisor 2020/04/01 2020/04/10 9 43 Figure 2: Gantt Chart FUNDING The researcher will be the sole funder of this project. Table 4: Budget for Funding Expense Budget Justification Personal Computer N/A Equipment already available Microsoft Office Software N/A Equipment already available Dictaphone N/A Equipment already available CAQDAS Software License R 800 Annual subscription required for the use of desired software. Transport R 3 500 Expense calculated using AA vehicle rates calculator30 Salaries/Wages N/A Sole researcher involved in data collection and analysis Gratuity N/A None Transcription Services R2 000 University-linked transcription services Editing/Proofreading Services N/A Existing subscriptions to both Grammarly31 and WriteCheck32 held by researcher Total R 6 300 The budget is planned for R 6300.00. There are no foreseen travel, accommodation, or overhead costs. 2019/08/23 2019/09/12 2019/10/02 2019/10/22 2019/11/11 2019/12/01 2019/12/21 2020/01/10 2020/01/30 2020/02/19 2020/03/10 Task One: Formalisation of Research Proposal Task Two: Postgraduate Faculty Review Task Three: Formal HREC and NHRD… Task Four: HREC feedback, corrections and… Task Five: Data Collection Task Six: Data Analysis Task Seven: Draft Submission to Supervisor 44 REFERENCES 1. WHO | South Africa [Internet]. WHO. [cited 2019 Mar 11]. Available from: http://www.who.int/countries/zaf/en/ 2. South Africa | UNAIDS [Internet]. [cited 2019 Mar 11]. Available from: http://www.unaids.org/en/regionscountries/countries/southafrica 3. Burton R. Maternal health: There is cause for optimism. S Afr Med J. 2013 Jul 5;103(8):520-521–521. 4. Progression of the epidemiological transition in a rural South African setting: findings from population surveillance in Agincourt, 1993–2013 | BMC Public Health | Full Text [Internet]. [cited 2019 Mar 11]. Available from: https://bmcpublichealth.biomedcentral.com/articles/10.1186/s12889-017-4312-x 5. Universal health coverage (UHC) [Internet]. [cited 2019 May 18]. Available from: https://www.who.int/news-room/fact-sheets/detail/universal-health-coverage-(uhc) 6. Goal 3 .:. Sustainable Development Knowledge Platform [Internet]. [cited 2019 Mar 14]. Available from: https://sustainabledevelopment.un.org/sdg3 7. Department Of Health - National Health Insurance [Internet]. [cited 2019 May 22]. Available from: http://www.health.gov.za/index.php/component/phocadownload/category/383 8. Kwon S. Thirty years of national health insurance in South Korea: lessons for achieving universal health care coverage. Health Policy Plan. 2009 Jan 1;24(1):63–71. 9. Ramesh M, Wu X, Howlett M. Second Best Governance? Governments and Governance in the Imperfect World of Health Care Delivery in China, India and Thailand in Comparative Perspective. J Comp Policy Anal Res Pract. 2015 Aug 8;17(4):342–58. 10. Hort K, Jayasuriya R, Dayal P. The link between UHC reforms and health system governance: lessons from Asia. J Health Organ Manag. 2017 May 15;31(3):270–85. 11. Yeoh E-K, Johnston C, Chau PYK, Kiang N, Tin P, Tang J. Governance Functions to Accelerate Progress toward Universal Health Coverage (UHC) in the Asia-Pacific Region. Health Syst Reform. 2019 Jan 2;5(1):48–58. 12. Campbell S, Roland M, Wilkin D. Improving the quality of care through clinical governance. BMJ. 2001 Jun 30;322(7302):1580–2. 13. Fryatt R, Bennett S, Soucat A. Health sector governance: should we be investing more? BMJ Glob Health. 2017 Jul 1;2(2):e000343. 14. Martin D, Miller AP, Quesnel-Vallée A, Caron NR, Vissandjée B, Marchildon GP. Canada’s universal health-care system: achieving its potential. The Lancet. 2018 Apr;391(10131):1718–35. 45 15. Atun R, de Andrade LOM, Almeida G, Cotlear D, Dmytraczenko T, Frenz P, et al. Health-system reform and universal health coverage in Latin America. The Lancet. 2015 Mar;385(9974):1230–47. 16. Adeloye D, David RA, Olaogun AA, Auta A, Adesokan A, Gadanya M, et al. Health workforce and governance: the crisis in Nigeria. Hum Resour Health. 2017 May 12;15(1):32. 17. Fusheini A, Eyles J. Achieving universal health coverage in South Africa through a district health system approach: conflicting ideologies of health care provision. BMC Health Serv Res. 2016 Oct 7;16(1):558. 18. Marais DL, Petersen I. Health system governance to support integrated mental health care in South Africa: challenges and opportunities. Int J Ment Health Syst. 2015 Mar 11;9(1):14. 19. Progress towards the negotiated service delivery agreement of the health sector | South African Government [Internet]. [cited 2019 May 22]. Available from: https://www.gov.za/progress-towards-negotiated-service-delivery-agreement-health-sector 20. Gilson L, Daire J, Gilson L. Leadership and governance within the South African health system. South Afr Health Rev. 2011 Jan 1;2011(1):69–80. 21. OHSC 2016-17 ANNUAL INSPECTION REPORT [Internet]. Office of Health Standards Compliance. [cited 2019 May 21]. Available from: http://ohsc.org.za/ohsc-media- response-to-the-article-by-the-business-day/ohsc-2016-17-annual-inspection-report-2/ 22. Exploring corruption in the South African health sector | Health Policy and Planning | Oxford Academic [Internet]. [cited 2019 May 21]. Available from: https://academic.oup.com/heapol/article/31/2/239/2355603 23. Campbell JP, Daft RL, Hulin CL. What to study: Generating and developing research questions. Vol. 6. Sage Publications, Inc; 1982. 24. Shenton AK. Strategies for Ensuring Trustworthiness in Qualitative Research Projects. Educ Inf. 2004;22(2):63–75. 25. Green J, Thorogood N. Qualitative Methods for Health Research. SAGE; 2018. 441 p. 26. openspecs-exchange. [MS-OXRTFEX]: Rich Text Format (RTF) Extensions Algorithm [Internet]. [cited 2019 Aug 25]. Available from: https://docs.microsoft.com/en- us/openspecs/exchange_server_protocols/ms-oxrtfex/411d0d58-49f7-496c-b8c3- 5859b045f6cf 27. Srivastava A, Thomson SB. Framework Analysis: A Qualitative Methodology for Applied Policy Research [Internet]. Rochester, NY: Social Science Research Network; 2009 Jan [cited 2019 May 22]. Report No.: ID 2760705. Available from: https://papers.ssrn.com/abstract=2760705 46 28. MAXQDA: Qualitative Data Analysis Software | Windows & Mac [Internet]. MAXQDA - The Art of Data Analysis. [cited 2019 Aug 25]. Available from: https://www.maxqda.com/ 29. Ethics in Qualitative Research: A View of the Participants’ and Researchers’ World from a Critical Standpoint - Dilmi Aluwihare-Samaranayake, 2012 [Internet]. [cited 2019 Mar 11]. Available from: https://journals.sagepub.com/doi/abs/10.1177/160940691201100208 30. Vehicle Rates Calculator - page