Developing the national community health assistant strategy in Zambia: a policy analysis

Like other LMICs, Zambia, with a population of 13.5 million people is facing a chronic shortage of human resources for health. This challenge has been worsened by the extra demands brought about by the high burden of HIV infection, malaria, tuberculosis and other health problems. HIV prevalence in the adult Zambian population stands at 13.5%; average life expectancy in Zambia is a mere 48 years; neonatal mortality is 35 per 1,000 live births; while infant and under-five mortality are also high, at 86 and 141 per 1,000 live births, respectively [17]. The proportion of vacant posts has been estimated at 42% in rural health centres, 22% in urban health centres, and 41% in hospitals. Vacancies for nurses, the most numerous clinical care category in Zambia, stand at 55%, as compared with 63% and 64% of clinical officer and doctor positions, respectively. Distribution of the health workforce is also highly uneven, with rural areas facing a more significant shortage than urban areas [17].

There are six levels of care in the public sector and corresponding facilities, including outreach services, health posts, health centres, and level-1 district, level-2 provincial and level-3 central hospitals [18]. The health post is the lowest level health facility, while nurses are the lowest category of formal health workers. Recent estimates suggest that Zambia has about half the health workforce that it needs, with fewer than 646 doctors and 6,096 nurses [17, 19].

Because of this shortage, about 23,500 CHWs have been active providing outreach services during the past 10 years, although this has been going on for a longer period in Kalabo district of Western Province, where a CHW pilot programme was officially launched in 1983 [19, 20]. In Zambia, CHWs are conceived as “members of communities who work either for pay or as volunteers in association with the local health care system and usually share ethnicity, language, socio-economic status and life experiences with the community members they serve. They have many titles, including home based care givers (often work with faith-based programmes), health promoters, community health advisors, lay health advocates, community health representatives, and peer health educators” [19], p. 8].

Community health workers either work for health care facilities or NGOs. In both instances they help in providing basic health services in areas with geographical access challenges and where there is a shortage of professional health workers. For example, CHWs have been able to manage malaria fevers by correctly interpreting results and appropriately prescribing antimalarials [21]. Other activities include promoting proper food production, basic sanitation, and detecting risk groups for the prevention of common illness [21]. In HIV prevention, treatment and care programs, CHWs organize support groups and perform home visits for patients who miss their appointments, clearly greatly contributing to the supply of health service and in particular to the rapid expansion of antiretroviral drug delivery [22]. Despite this, there has been no comprehensive community health policy and strategy to guide the operations of this workforce, nor has there been any comprehensive structure to regulate and monitor the services they offer. The mandate to regulate their services is often left to individual institutions or organisations engaging them.

With increasing external pressures and funding, in particular for HIV/AIDS control, CHWs have been required to perform new tasks with different training and incentive packages, and this has ultimately caused further fragmentation in their operations [23]. Additional challenges that CHW programs face in Zambia include high turnover, low motivation, inadequate supervision, insufficient compensation or incentives, and low recognition by qualified health care providers, all of which limit their ability to effectively contribute towards primary health care [17, 19].

To try and address some of the challenges that CHWs face in health service delivery, the Zambian Ministry of Health (MoH) developed the National Community Health Assistant Strategy in 2010 [19]. The purpose of this strategy is to formalize and standardize the role of CHWs in the health system by creating a new group of workers called community heath assistants (CHAs). Compared to CHWs, whose training is short and not standardised, CHAs are expected to undergo a year’s standardised training programme. Furthermore, in contrast to CHWs who work as volunteers, CHAs will be put on the government payroll and registered with a regulatory body. The CHAs are expected to work below nurses, and they deliver health services on a task-shifting basis (i.e., from nurses to CHAs). Their tasks and mode of supervision are clearly specified in the strategy. The Neighbourhood Health Committees (committees comprising leaders in the community) and the District Health Management Team are responsible for selecting CHAs from the communities in the district. Implementation of the strategy started in June 2011 with a pilot phase lasting until 2013. During this first phase, 307 CHAs were trained. If successful, the strategy will be rolled out to all districts in Zambia in four phases, from 2013 to 2017. About 5,200 CHAs will be trained by 2017 [19].

Since the CHA strategy was the first policy framework aimed at integrating CHWs into the health system in Zambia and since it was only launched in 2010, little analysis has been performed, to date, on how power dynamics among various actors and contextual factors shaped the process of producing the strategy and its content. Current studies on the CHA strategy in Zambia have focused more on considerations relevant to selecting and compensating workers at the community level [24].

This paper, therefore, seeks to fill this knowledge gap by addressing the following questions. First, which actors contributed towards the development of the CHA strategy, and what were their roles? Second, how did the process of agenda setting and developing the CHA strategy evolve? Third, to what extent did actors’ differential power and other contextual factors shape the development and content of the policy? These questions are not only relevant to the MoH in Zambia but also to organizations and other countries that intend to introduce health reforms aimed at strengthening community participation in health care services.

The “policy making process” refers to the way in which policies are initiated, developed or formulated, negotiated, communicated, implemented, and evaluated. This process often occurs in stages, which include problem identification and issue recognition (i.e., focusing on getting issues onto the policy agenda), and policy formulation, which includes actor involvement in formulating policy, how policies are arrived at, agreed upon, and how they are communicated. The other stages are policy implementation as well as policy monitoring and evaluation [25].

The policy analysis framework of Walt and Gilson will guide analysis of the findings [26]. This framework recognizes that the health policy process involves four elements, namely (i) the context within which the policy is formulated and executed, (ii) the actors involved in policy making, (iii) the steps associated with the development process, and (iv) the policy content. In this paper, the analysis approach is guided by the assumption that actors, such as the CHA strategic team, directly influence the course that any policy process will take, as well as the final content of the policy through their actions and values [27]. Further, we assume that actors’ views are influenced by the context in which they live and work, while their role in policy development is dependent on their power [25, 26].

The study was conducted in Lusaka (the capital city of Zambia) and Kapiri Mposhi district, located in the Central Province of Zambia (Figure 1). Lusaka was selected because this is where the strategic team members were based. Furthermore, this is where the MoH Headquarters – the body responsible for coordinating all health policies – is located. The other study area, Kapiri Mposhi district, was chosen because it is one of the rural districts where the CHA strategy is being piloted. Training of CHAs was conducted in Ndola district which is located on the Copperbelt Province, about 100 km north of Kapiri Mposhi.

A qualitative case study methodology was used to analyse the CHA development process in Zambia. The case study methodology is an empirical approach that investigates contemporary phenomena within a real-life context, where the boundaries between phenomena and context are not clearly evident and in which multiple sources of evidence are used [28]. The case study approach was considered as appropriate for this study because the CHA strategy was developed within a complex context, which involved social interactions, and which was dependent on multiple local and global influences. This approach enabled us to capture the relations among actors in the process of developing the strategy, as well as the realities of the context in which the strategy was formulated.

The study used purposive sampling to select respondents for the key informant interviews based on their role in developing the CHA strategy and coordinating CHA activities: these included strategic team members, MoH staff, and community members. NGOs working with CHWs were not part of the key informant interviews as they did not participate in developing the CHA strategy and they were not familiar with the strategy at the time of data collection.

All interviews were recorded digitally and later transcribed verbatim. Data analysis was conducted manually and followed a thematic framework analysis approach, which is used to classify and organise data according to key themes, concepts and emergent categories [33]. First, interviews and transcripts were thoroughly read. This was followed by the development of codes and categories, a process that was guided by the elements of the Walt and Gilson policy analysis framework (content, context, actors, and process) [25], as well as the case study approach. This approach was relevant in analysing the complex relationships among actors as well as the context.

Then, the codes and categories were cross-checked with the interview transcripts in order to ensure that they were applied to relevant responses found within and across the interviews. The focus was placed on identifying, summarising, and retaining the patterns and similarities, differences, and new emerging themes. Data were then triangulated with other sources such as the information gathered through informal discussions and review of documents. Although the process is presented as a linear process, it is important to stress that this was an iterative process that involved continuous shifting back and forth from participants’ narratives to the researcher’s interpretation of what the informants meant [34].

Ethical clearance to conduct the study was obtained from the University of Zambia Biomedical Research Ethics Committee [IRB 0001131 of IORG 0000774, reference number 009-10-11]. Verbal consent was sought from the key informants in the study. This was preceded by an explanation to the participants regarding the research objectives and aims. Furthermore, informants were notified that they were free to withdraw from the study at any point. Study participants were also assured of confidentiality during and after the study period.

The policy agenda consists of issues that policy makers agree to consider at a given time. Agenda setting is a process through which actors negotiate to have their interests considered by policy makers.

Having refined the concept, in 2009, the MoH formally appointed a strategic team to facilitate the process of formulating the strategy. Its mandate was to engage stakeholders and regularly update senior management and the human resources technical working group at the MoH on the process. It was also required to collaborate with four sub-committees, namely curriculum, logistics, monitoring and evaluation, and budgeting (Figure 2).

The strategic team was appointed by the Permanent Secretary for the MoH. The team mainly consisted of staff from various directorates/departments under the MoH headquarters. For example, there were representatives from the Directorate of Human Resources and Administration, Public Health and Research, and Technical Support Services as well as Nursing Services. Others were from the Planning, Health Promotion, and Child Health Services departments. Only two members were drawn from the MoH directorate which is not under the MoH headquarters, the National Malaria Council Centre. The strategic team was chaired by the Directorate of Human Resources and Administration.

Once the strategic team was in place, the MoH conducted a situational analysis with 18 implementing partners, 76 CHWs and the District Medical Officers (DMOs) to understand the roles, scope and challenges of CHW-driven programmes. The GHWA funded the situation analysis. However, it was not clear who from the MoH coordinated the analysis and the partners that were interviewed, or how respondents were selected.

The situational analysis showed that DMOs recommended standardisation of recruitment, education requirements for recruitment, training and supervision of CHAs, and remuneration, with monetary incentives being the most preferred type of incentive by the DMOs and CHWs. Implementing partners recommended standardisation of training, remuneration and supervision for CHAs. All respondents unanimously supported the need for providing periodical refresher courses to CHAs. However, implementing partners were silent on the type of remuneration to give to CHAs or on the educational requirements for recruitment. Further, only about half of them supported standardisation of recruitment processes for CHAs [19].

Having concluded the analysis, the MoH organised a consultative meeting to negotiate with unions and professional health bodies on the need to develop a new group of health workers in Zambia. Staff in the strategic team from the Directorate of Human Resources with the assistance of those from the Directorate of Nursing Services at the MoH headquarters chaired the consultative meeting. Although it was termed consultative, the meeting was in practice strategically organised to enable the pioneers of the CHA strategy to push the concept through.

To convince the unions and nursing bodies on the relevance of developing the cadre, the strategic team members reported that, apart from revising the title, they also had to change the emphasis regarding the area of operation for CHAs when presenting the concept to the stakeholders. They stressed that CHAs would be based in rural areas, the areas where the pioneers of the strategy knew that most professional workers would not want to work from. This is how one of them explained how they managed to convince the stakeholders:

“We did our homework before presenting the idea to them. For this one, we presented the strategy like it was going to be purely in the village. We asked them if they were willing to go and work in the village. They all said no. Then we reminded them that most of their grandmothers were in the villages. So we asked them “who should provide health services to them?” They all responded “Another cadre! … another cadre!” So we told them that the same cadre is the one we were now calling Community Heath Assistants.” (Strategic team member).

The findings of this study show that the policy development process was driven by the need to address the human resources for health shortage in the country. Our analysis also reveals, however, that it was just a few actors from the MoH headquarters who worked to place the issue on the policy agenda and who participated in formulating the strategy. Agenda setting, which is a phase when issues get selected and prioritised to come into the policy agenda [35, 36], was mainly informed by the Department of Human Resources and Administration from the MoH. Actors from this department took up the role of “policy entrepreneurs” or pioneers of the idea of developing a new group of workers [36].

The study showed that agenda setting may not always be a straightforward process, as some stakeholders may be sceptical of new ideas. Scepticism may result in delays in confirming agenda items [36, 37]. For example, when developing the HEP in Ethiopia, which is one of the first programmes in Africa to formalise HEW services, not all partners in Ethiopia supported the initiative as they feared that HEP would not positively contribute towards maternal health [38]. Similarly, the health professional bodies in Zambia did not support development of the CHA strategy at first as they did not understand the concept and were pessimistic about its potential benefits. The concept was only placed on the agenda through several negotiations and consultations between the pioneers of the concept and health professional bodies. Negotiations and consultations are an important part of agenda setting as they facilitate refining of the policy issue by “softening up” stakeholders initially opposed to the issue, and by expansion of the policy issue within or beyond the local context and actors [36].

International actors who participated in the process and who subsequently contributed towards refining the concept to an agenda item included the GWHA. The influence of international actors and global context on national policies in LMICs has been widely documented [39], with international bodies potentially playing the role of indirect actors in national policy reform. These actors may not only influence agenda setting but also the process and content. For example, documentation on the performance of HEWs in Ethiopia partly motivated the MoH to develop the CHA programme [19, 40]. The other research evidence used in developing the strategy was from Malawi. The only data from the Zambian context was from the National Situation Analysis which mainly focused on gaps in human resources for health and challenges faced by CHWs [19]. Furthermore, the GHWA also recommended that Zambia should develop a strategy similar to the HEP in Ethiopia and that it should be implemented quickly [23].

Review of the strategy showed that its content also reflected some of the key messages from the Global Consultation on Community Health Workers organised by the GHWA in 2010. These included the recommendation that countries should consider integrating CHWs into national health workforce plans, as well as the need to ensure a regular and sustainable remuneration stipend for CHWs [13]. This consideration of the views from global actors in developing the policy, by interest groups wishing to influence the policy-making process, could be viewed as one strategy of securing a recognised position with the donors/funders [41], which may in turn guarantee financial and technical support from the agencies during the policy implementation.

Although central agencies and policy entrepreneurs could be useful in policy making, literature suggests that their involvement may limit participation of local stakeholders in policy formulation [5, 42]. Limited stakeholder involvement could work against the ideal policy making process which should accommodate relevant government bodies, civil society, community structures, private-not-for-profit, and health professional groups in health policy reform [13]. While stakeholder consultation is recommended, the findings of the study suggest that consultation may in practice be deployed strategically in order to advance the interests of key drivers of the policy, or donor conditionalities. Strategic consultations could include the reported side-lining of those with opposing views from participating in the CHA formulation process, and the inclusion of contentious issues in the CHA strategy without involvement of several strategic team members.

Our findings therefore support other studies on health policy reforms that argue that actors and interest groups exercise different degrees of power in influencing the course of the health policy reform process [5, 42‐44]. The degree to which actors are able to influence the course of policy making process as well as the content of the policy depends, amongst others, on their perceived or actual power [25, 27]. This power may be exercised by making sure that issues that reflect the values and beliefs of those in top management, or of the pioneers of the idea, reach the policy agenda. It may also seek to ensure that opposing views in health reforms are not given a platform [5, 45].

In addition, the negotiations and compromises with nursing and statutory bodies highlight the political nature of the policy process. Our findings resonate with the view that “policy change is not simply a technocratic process based on rational analysis, and that knowledge alone is not sufficient for policy change. Policy is profoundly political” [45], p. 8]. This means that policy is set by the various ways in which people exert control, influence, or power over each other [42]. Perhaps a more fundamental problem relates to the fact that, although the importance of politics in policy reform has been acknowledged, there is also broad agreement that political issues are rarely analysed by the people involved at all stages of the process, and particularly with respect to the interactions between international donor agencies and recipient country governments or institutions responsible for policy reforms in LMICs [43, 46]. One possible explanation for this insufficient political analysis could be that although many actors may have their own clear political agendas which they want to pursue in relation to the policy, these often remain unstated, and therefore invisible to other actors. This is likely to keep particular political issues out of the official rhetoric associated with the policy.

The lack of involvement of CHWs in formulating, finalising, and launching the strategy showed the lack of voice and power of the community-based health workforce in the contemporary policy development process and reform [5]. This insufficient involvement may not be unique to this study, as it has been documented that public participation in deciding the policy agenda in LMICs has often been limited [43]. Perhaps this limited involvement could be explained by the political will or technocratic model which assumes that decisions by senior managers or leaders or a reform champion are necessary and sufficient for policy change and that these leaders are rational actors maximizing the public interest [47], which, in the context of this study, could include addressing the chronic human resources for health gap in Zambia. Based on the findings of the study, one may argue that as desirable as engaging multiple stakeholders may be, achieving broad consultation in formulating policy may not be feasible in situations where it is difficult for stakeholders to agree on key policy contents due to different interests, or where key policy developers may have specific issues to advance.

Advancing the views of a limited group of stakeholders in the policy formulation process has the potential of producing a policy that does not reflect the interests of the majority of stakeholders, or consider key issues within the given context. For example, although it has been widely documented that movement of health workers from one job to another is a major problem affecting human resources for health globally [2], the CHA strategy did not comprehensively consider how to address staff attrition or professional development of CHAs. This policy gap could limit the possibility of building a sustained and effective strategy as a sustained CHW strategy requires viewing the CHW position not as an end-point but as a means to an end, and accepting a high degree of turnover which in turn requires putting in place measures to address the anticipated turn over [15]. Furthermore, although several studies have shown that acquired skills are quickly lost, and therefore refresher training is of crucial importance for sustaining the quality of performance and sustainability of CHWs [1, 48], the CHA strategy appears not to have adequately considered general professional development and progression matters. This limited consideration of vital issues in the strategy may resonate with Walt and Gilson’s view that much health policy wrongly focuses on content of reform and neglects the context within which policy is developed [26].

Apart from contributing to policy gaps, limited participation could potentially affect implementation of the CHA strategy, as collaborative planning with all relevant stakeholders at the initiation of the design process of any CHW programme, whether for HIV or other service delivery, is a key factor for the success of the programme’s development as well as its implementation [9]. Limited stakeholder involvement could result in weak collaborative systems and referral networks between CHAs and other health service providers at community level. For example, studies conducted on the HEP in Ethiopia showed that poor referral systems affected the work performance of health extension workers [49, 50]. It has therefore been suggested that for community-based workforce to work effectively, apart from formalising their services, there is a need to create an appropriate working environment and conditions within the health sector. It has been argued that while the renewed attention to CHWs is welcome, scaling-up of CHW programmes without considering ways of simultaneously strengthening health systems may comprise the quality of the CHW services [48].

Trustworthiness of the research findings was enhanced by triangulating data obtained from the key informant interviews with those obtained from documentary review and informal discussions. Although most of the information was similar, some disagreements were observed between interviews and documentary review regarding placement and remuneration for CHAs. Furthermore, trustworthiness was achieved by compiling, documenting, and including all data during the analysis process.

The study was unable to include all the relevant actors. Although efforts were made to interview all the 15 CHA strategic team members, only 9 were interviewed as others were not available during the study period. NGOs working with CHWs were not included in the study because they were not part of the process of developing the CHA strategy. Lack of inclusion of these stakeholders is a limitation as they could have given some important perspectives of the process. Despite these limitations, the rich description of the policy process in this study provides a valuable contribution to the knowledge base on policy reforms for community-based health workforce in LMICs.