Improving public health sector service delivery in the Free State, South Africa: development of a provincial intervention model

The Free State province of South Africa has recorded substandard population health outcomes at least since 2011 [1]. The dire need for effective and implementable interventions to improve public health services and outcomes is clear [2]. To avoid resource wastage, efforts should be made to systematically design health systems strengthening interventions with the potential of system-wide implementation [3‐5].

When a new Member of the Executive Council (first author) assumed leadership of the Free State Department of Health (FSDoH) in 2013 there was a clear need to address the rising burden of both communicable and non-communicable disease [6], as well as declining community trust in the public health system. The Member of the Executive Council outlined a vision to “strengthen the health system’s effectiveness, drive system changes, improve burden of disease outcomes, [and] ensure financial sustainability for better health outcomes and increased life expectancy.”

A multi-method situation appraisal of the FSDoH’s public health services was undertaken by the first author, a senior programme manager (third author) in collaboration with two health systems researchers (second and fourth authors) and a community health specialist (fifth author) in collaboration with healthcare workers in 2013. The appraisal established that the main overall system challenges underlying sub-standard performance in the province included fragmentation of services, staff shortages, financial/cash-flow problems and leadership and governance defects [1]. There was a clear need to improve programme integration and leadership, and to address organisational, human resources and financial deficiencies in this setting. The above-mentioned team collaborated in developing, implementing and evaluating a whole-system intervention to improve public health service integration and outcomes, namely the Health Systems Governance & Accountability (HSGA) intervention.

Although no universal definition or concept of integration exists [7], in 2005, the World Health Organization (WHO) explained the term as follows: “The notion of integration has a long history. Integration is supposed to tackle the need for complementarity of different interdependent services and administrative structures, so as to better achieve common goals. In the 1950s these goals were defined in terms of outcome, in the 1960s of process and in the 1990s of economic impact” [8 pp. 108–109]. A systematic review conducted in 2016 defined integration as “a variety of managerial and operational changes to health systems that bring together inputs, delivery, management, and organizations of particular service functions, in order to provide clients with a continuum of preventative and curative services, according to their needs over time and across different levels of the health system” [9 p. 2]. The WHO suggests that integration has to take place at three levels: (i) patient level, i.e., case management; (ii) service delivery level, i.e., multiple interventions provided through one delivery channel and (iii) systems level, i.e., bringing together the management and support functions of different sub-programmes, and ensuring complementarity between the different levels of care [8]. Integration is a choice that affects programme financing, planning and delivery, and, ultimately, the achievement of public health goals [10, 11].

Integration of disease control programmes such as the HIV/AIDS, tuberculosis and chronic disease programmes within the district health system and comprehensive primary health care (PHC) approach has been a priority in South Africa for decades [12, 13]. However, attempts at integration are hindered by leadership and management challenges as observed in three provinces in the complex dynamics between managers responsible for specific policies and those responsible for managing the integrated delivery of all policies [14].

The WHO [2, 15] advances that health systems are comprised of six ‘building blocks’ that need to be considered in health system strengthening: (i) service delivery; (ii) health workforce; (iii) information; (iv) medical products, vaccines and technologies; (v) financing and (vi) leadership and governance (stewardship). The building blocks framework has been found to be useful to assess in-country health system performance [16‐18]. However, there is some uncertainty as to how the framework can best be used to address the problems of fragmentation and ineffective performance of public health systems in low- and middle-income countries. Mounier-Jack et al. [19] thus recommended that researchers using the building blocks framework should adapt it and make it context-specific.

The FSDoH leadership believed that conceptualisation of the public health system in terms of the building blocks framework and improved integration of service programming and rendering through, amongst others, reconfigured facility clusters and service hubs, would help to address a broad spectrum of health system challenges. The decision was therefore made to design and develop a system-wide re-engineering intervention to address the fragmentation of healthcare provisioning using a systems approach. The working definition for a systems approach was one that “applies scientific insights to understand the elements that influence health outcomes; models the relationships between those elements; and alters design, processes, or policies based on the resultant knowledge in order to produce better health at lower cost” [20 p. 4].

Health systems are complex and characterised by high levels of variability, uncertainty, and dynamism [21]. Kaplan et al. (2013) [20] proposed that the use of a systems approach is beneficial in addressing complex health systems challenges by considering the dynamic interaction among the multiple elements (i.e., people, processes, policies, and organisations) involved in caring for patients and the multiple factors influencing health. These authors proposed four essential considerations or steps in designing interventions from the perspective of a systems approach, (i) identification, (ii) description, (iii) alteration and (iv) implementation. These steps were followed in the development of the HSGA intervention model in the Free State. Kaplan and Norton’s (1992) Balanced Scorecard performance measurement tool [4] was concomitantly put in use.

Details on the implementation and assessment of the intervention including the results of a questionnaire survey and focus group discussions with health managers and community representatives will be reported elsewhere. This descriptive study relates the development of an intelligible and promising intervention, the HSGA intervention model, and its formalisation into an official policy, the HSGA policy, to address fragmentation and improve public health service delivery in the Free State by the provincial health department in collaboration with its stakeholders. A participatory design involving a wide range of public health stakeholders and role players in plans and efforts to improve the integration and outcomes of public health services was followed as part of routine health service delivery management processes.

The HSGA intervention model was developed and formalised into policy in a quest to bind the employees and managers across the whole department to comply with implementation of the model in a sustainable manner and in line with the Strategic Transformation Plan 2015–2030. The model was designed with support and input from all the relevant stakeholders.

A process flow was conceptualised on how service delivery at the district level could be functionally integrated with the next levels of referral through a cascade termed Re-engineering of Admission (READ) to enable smooth referral of the patient from the community up to the tertiary level of care. Another process flow termed the Re-engineering of Discharge (RED) [40] was adopted to direct the safe referral and movement of the patient downwards through the system. These changes were reflected in a conceptual framework, the HSGA intervention model, illustrating the need for interventions to integrate the various levels of care within the reviewed policy positions and structural rearrangements (Fig. 3).

The HSGA intervention model emphasised the accountability of the system to the community and patients it served. It indicated how the different healthcare levels dovetailed into one integrated system to improve health system performance based on the six health system building blocks. It provided a framework for leadership, governance, and accountability throughout the whole health system, including community- and facility-based PHC, as well as district, provincial, tertiary, and central hospital services. All healthcare providers were required to comply with the approved referral and service protocols and to provide feedback to the referring facility clinician for further handling and down-referral for continuity of care.

In respect of the service delivery building block, the HSGA intervention model and policy also directed that all public healthcare facilities had to adhere to the norms and standards required by the Council for Health Service Accreditation of Southern Africa; healthcare services had to be provided in an integrated manner with good co-operation and communication between facilities and functionaries; patient movement up and down the health service value-chain had to be managed in line with procedural and clinical protocols; monitoring and evaluation functions had to be co-ordinated by the strategic planning and monitoring and evaluation unit at the corporate office; the District Clinical Specialist Teams had to conduct clinical oversight, clinical governance protocol audits and surveillance in district facilities; and the different levels of care had to be linked through a reciprocal transfer and referral system.

Regarding the leadership/governance building block, the HSGA intervention model and policy directed that leadership and managers had to ensure coordination and integration of governance efforts around a common mission to focus management teams’ activities on achieving optimum performance of the public health system; adopt a holistic view of the system’s operations to enable effective achievement of desired outcomes; comply with service integration imperatives and clinical governance protocols; and monitor obstacles with immediate intervention where necessary. Led by the district managers, the District Management Teams were responsible for governance at the PHC level, integration of service implementation, and compliance with policies and procedures. The District Management Teams were expected to collaborate with multiple internal and external stakeholders who play a role in disease control programme management. All the changes were streamlined in accordance with the building blocks and outlined in the draft Strategic Transformation Plan 2015–2030 [41].

Table 4 shows the changes in the configuration of the institutions per district, including the planned establishment of 32 health posts. A health post is a facility below the level of a clinic which is staffed by lower-category nurses allowing community members to collect their chronic medicines or receive care for minor ailments like wound care. Comprehensive PHC services provided by the Ward Based Outreach Teams in the households, health posts, mobile clinics and fixed PHC facilities were established for effective and efficient PHC re-engineering, thus bringing healthcare services closer to the people. The PHC facilities were re-organised into PHC clusters, each having a group of PHC clinics and a community health centre. Certain PHC clinics were upgraded to community health centres, and certain small hospitals downgraded to community health centres.

EMS are responsible for all transportation of patients, both in emergencies and planned patient referrals in FSDoH facilities. Patient routes had to be aligned with the changed referral paths and routes according to the reviewed departmental referral policy considering that the 60-seater buses were to be operated on long distances to Bloemfontein only. Bloemfontein is the capital of the province and is where most of the higher-level referral and specialist hospitals are located. The categories of patients referred to Bloemfontein was reviewed in line with the approved service packages and patient needs. The new routes improved the flow of patients to the nearest facilities with reduced transit times. Therefore, the operational costs of Emergency Medical Services were reduced as well.

To transform mental health service delivery by 2030, the service delivery system focused on the following aspects within an agreed upon timeframe. In line with the WHO (2003) recommendations [42] regarding organisation of mental health services, the mental health system was planned to include an array of settings and levels that included primary care, community-based settings, general hospitals, and specialised psychiatric hospitals. It was further planned that the Free State Psychiatric Complex would undertake outreach visits to regional hospitals and that the latter would in turn conduct outreach services to district hospitals and selected community health centres. Further outreach services would be provided to the Kimberley Hospital Complex in the Northern Cape province and Queen Mamohato Memorial Hospital in Lesotho.

This study describes the development the Health Systems Governance & Accountability (HSGA) intervention model by the FSDoH in collaboration with the community and other stakeholders following a participatory action approach. Researchers publish the processes they use to develop interventions to improve healthcare to help future developers to improve their approaches, plans and practices. Description of the development of interventions is important because it provides future developers with the terminology and language to identify, conceptualise and explain their rationales and more effectively plan and design context-specific intervention models [43]. It is important to understand how the models that end up being designed during interventions, complement the traditional ways of doing things for the betterment of a common good such as public health. A movement toward an assets-based approach to intervention design and implementation has started to take place in public health where intervention development processes emphasise activating and drawing on the strengths and skills of individuals, communities, and through the co-production of intervention programmes and research [44].

A systematic review of studies reporting intervention development showed that they mainly took a pragmatic self-selected approach, a theory- and evidence-based approach, or a partnership approach, e.g., community-based participatory research or co-design [45]. Both theory (systems) and provider/user approaches were utilised in the development of the HSGA model in the Free State. Systems theory and the WHO health systems building blocks framework helped the developers and participating stakeholders to conceptualise and plan a whole-system intervention, the HSGA intervention model and policy. The participatory approach allowed the developers to incorporate the experiences and ideas of frontline health systems users and providers and managers across all levels of the public health hierarchy. Obtaining both user and provider perspectives is crucial in efforts to model and design context-specific public health strengthening initiatives [46‐48].

Previous studies to find ways to improve health outcomes through a consultative and iterative process involving users identified barriers such as weak and unorganised referral patterns and inadequate human resources as crucial [44, 49]. The development of the ‘Better Health Outcomes through Mentoring and Assessment’ (BHOMA) intervention in Zambia was intended to specifically address such challenges [17]. Similar challenges were addressed in the development of the Free State HSGA model, but the focus was on improving integration and performance assessment as means to improve health outcomes. Both interventions were intended to provide the system-wide solutions to the challenges, rather than focusing on specific diseases. The influence that the users exerted could be thought of as a ‘user intervention’ because they would in turn support changes and improvements while constantly assessing whether they are appropriate or not. Von Hippel [44] and Altman et al. [50], reiterate that the context and support for this participatory approach to ‘user intervention’ leverages user innovation capability in influencing the manner of medicine practice or provision of health services, which is considered a novel approach in dispensing public health. Von Hippel explains that when faced with problems or challenges, the users of the products, services or beneficiaries of processes are often likely to develop solutions to their problems and sometimes offer innovative ways and new approaches and solutions to challenges facing them. Hence the essential aspects of the HSGA intervention model were premised on collective solution finding and co-creation of approaches. Users and communities therefore got to enjoy the benefits of the changes they themselves helped to bring about. That is why integration of inputs from users or beneficiaries into the development of innovative approaches or models should be considered as a way to leverage the strengths of communities served by health providers for efficient and effective interventions to be developed and new approaches to be generated to address the varied complexities of health service challenges they face [50].

In the development of the HSGA intervention model obtaining users and communities’ perspectives took the form of unannounced visits to hospitals and clinics and community meetings respectively. This worked well because during the unannounced visits and community meetings, different and diverse groups of patients and providers tended to congregate. They presented their respective views and suggestions on which processes need to be changed or improved for the service offering to improve. This was challenging to deal with because the views and recommendations of the respective users, communities and frontline health staff were often diverse and needed to be deciphered, understood by all parties, and reconciled.

In the development of the HSGA intervention model obtaining managers and providers’ perspectives took the form of learning from and incorporating their views and suggestions presented during routine management meetings, as well as engagements during routine performance assessments. This was advantageous because challenges and proposed solutions presented by the managers and providers could then be directly discussed and addressed. Defects in the referral processes and clinical procedures were identified and reviewed for translation into corrective policy positions. This was challenging because procedures and processes often had to be reviewed ‘on-the-go’ and drafters were to affect the changes on the existing process and procedure documents forthwith, for later implementation.

In the development of the HSGA intervention model obtaining other stakeholders and experts’ perspectives took the form, amongst others, of a health indaba. This was beneficial because various stakeholders at the indaba raised their views, some of which corroborated the issues earlier raised by users and communities, as well as during the initial situation appraisal. It was sometimes difficult to differentiate personal experiences from health system shortcomings relating to process or procedure defects or challenges. Nevertheless, the recommendations made were considered in the overall revisions of processes and procedures.

Using systems theory to understand organisational performance helps intervention developers to obtain a broader perspective that enables recognition of patterns, cycles and overall structures, rather than focus on specific events or aspects of the system [51, 52]. While functional integration of health services may be an attainable vision [53], certain important intersecting health system capabilities are required to bring about frontline service integration as well. These include fully functional frontline health services, adequately trained and motivated healthcare workers, availability of suitable technology, and devolving authority and decision-making processes to lower cadres of frontline managers and staff to adapt integration processes to local circumstances [54]. Therefore, during the development of the HSGA model, Kaplan’s theory influenced the assessment of how the health system was organised, as well as the identification of the various service interlinkages and interdependencies of the multiple elements (building blocks) involved in public health systems strengthening and performance. The various interlinkages and process dependencies were clearly described, broadly understood and reconfigurations entrenched through policy changes to ensure continuous improvement of the Free State public health system’s performance.

The WHO considers systems thinking as an essential ingredient in opening pathways to identification and resolution of health system challenges [55, 56]. The use of the WHO building blocks approach [2, 15] to guide the integration process of the HSGA intervention model, placed emphasis on the interactions and interdependencies to guide the management of transitions among and between the demand side (community) and the supply side (healthcare facilities and staff).

The Zambian BHOMA study [17] proposed a framework to evaluate a complex health system intervention applying systems thinking concepts. It formed the basis for intervention development using systems thinking, where emphasis was placed on understanding the complete system and how the various parts interacted with each other to create a functional whole as opposed to the individual parts that made up the system [57]. In the development of the HSGA model in the Free State, system theory and the WHO building blocks framework were used to as a ‘template’ to assess the effectiveness and operational efficiencies of various components or divisions within the health services platform. This approach was appropriate because we could discern areas of weakness and how they could be addressed.

The HSGA intervention model was directed at three levels of care, i.e. (i) the community and sub-district levels of care; (ii) the district and regional level of care and (iii) the provincial level of care. At the community and sub-district level, care was to be delivered by Ward Based Outreach Teams and community health workers; at the district or regional level, by local PHC clinics, community health centres and district and regional hospitals; and at the provincial level, by provincial hospitals and the central hospital. Within and between these levels, transitions in the form of READ and RED [40] were intended to be managed efficiently and effectively in an integrated manner to mitigate morbidity and mortality. Austin et al. [49] acknowledged that the movement of a patient from the home or community setting into the health system during admissions into a healthcare facility or hospital and vice versa, constitutes transitions to avoid the effects of care-risk factors that can lead to morbidities, excess stays and even re-admissions if those transitions are ineffective and inefficient. In the Free State, the READ and RED transitions were conceptualised and applied to improve referral up and down the referral hierarchy. The effectiveness and efficiency of these transitions were based on the strength of the referral system which the HSGA intervention model sought to integrate and improve. Austin et al. [49] pointed out that both the system and patient factors exert influence on the transitions and the capacity to address complex barriers to safe admission and discharge procedures and workflows. The development of approaches like the HSGA intervention model to address and improve complex transitions effectively and efficiently require continued testing and assessment of their effectiveness in view of the complexity of the transitions.

Austin et al. [49] corroborate the importance of iterative, internally driven intervention, monitoring, and wide dissemination during the implementation process as was applied in the Free State and will be reported in another paper. In the current study monitoring or evaluation was conducted using Kaplan and Norton’s (1992) Balanced Scorecard performance measurement tool [4]. This decision was based on its successful application in healthcare settings and organisations elsewhere [31‐33]. Our experience aligns with that of Oliveira et al. [31] that the Balanced Scorecard is a valuable management tool to enhance flexibility, collaboration, innovation and adaptation that contribute to improve healthcare integration and outcomes.

A limitation of this descriptive study is that it is first authored by the former executive leader of the FSDoH, and an element of bias is thus inevitable. However, inclusion of authors who are not employees of the department helped to mitigate this bias. A strength of the current study is that the health system as a whole, and not just disparate elements within it, were considered in the development of the HSGA intervention model. The study provides useful insight on practical issues to consider during the development of public health interventions geared towards health systems strengthening and moves away from the traditional practice of addressing only individual behaviour change. Furthermore, this study attempts to provide a systematic approach to health systems intervention development with community and stakeholder involvement and aligns with Wight et al.’s [3] and Kaplan et al.’s [20] recommendation that a systematic approach to intervention development, as well as rigorous evaluation, are required. Another strength of the HSGA intervention model is that it was developed during normal service delivery processes and mostly during routine management planning meetings and discussions.

Public health practice should be evidence-based and grounded in theory and systems thinking. The WHO’s health system building blocks framework is a useful guide to conceptualise and assess the status of a health system and its performance and is a valuable tool to align or configure services according to communities’ actual realities and needs. Based on the building blocks concept, the HSGA intervention model was developed by the provincial health department during routine service delivery conditions to bring about desired reforms where gaps and inefficiencies had previously been identified. Reconfiguration of the service platform provided an enabling framework for future efforts to enhance integration, improve service delivery and improve the public health system’s performance. Thus, the development of the HSGA intervention model and its formalisation into an official policy provided an opportunity to improve healthcare service integration, health system performance, and improve health outcomes in the Free State public health sector. The study contributes to the body of knowledge on health systems strengthening and illustrates how to draw on both theory and participatory approaches to develop an intelligible intervention to improve public health system integration and performance.