Health facility committees and facility management - exploring the nature and depth of their roles in Coast Province, Kenya

Community accountability can be defined as listening to and responding to the views and inputs of the public, citizens, or users, and is increasingly being emphasized in health delivery in developing countries [1‐3]. A range of mechanisms have been introduced to strengthen community accountability at the facility level, including health facility committees (HFCs), patients' rights charters, suggestion boxes, customer care desks and health clubs. At the level of peripheral health facilities such as dispensaries, health centres and health posts, local committees are the most widely documented mechanism, with some information available on their implementation or impact from a wide range of countries including Kenya, Uganda, Tanzania, Niger, Nigeria, Benin, Zambia, South Africa, Peru, Mexico, Cambodia and Nepal (Molyneux et al.: Community accountability at peripheral health facilities, submitted).

Although some notable successes have been documented in committee operation (e.g. [4‐6]), a number of significant challenges have also been identified. These have included problems with the selection and functioning of committees, lack of clarity in roles and responsibilities, difficulty in sustaining voluntary membership over time, insufficient resources, and inadequate representation of and links with the wider community (Molyneux et al.: Community accountability at peripheral health facilities, submitted). However, the evidence base remains limited, with only a handful of studies documenting in detail how community accountability mechanisms have operated or influenced health services, and how their performance can be improved. The literature is particularly limited in its analysis of the depth of community involvement, which may vary from mere information provision to consultation, representation, and at the greatest depth, actual influence over decisions [7‐10].

This paper examines the nature and depth of community accountability through HFCs in Coast Province, Kenya, focusing on their role in facility management. HFCs in Coast Province have been relatively well supported and have responsibility for a component of the facility budget, so there is the potential for committees to overcome some of problems identified in other settings.

The Kenyan Government officially established HFCs in 1998 [11], although in some facilities similar community-based or NGO supported mechanisms existed before then. The official roles of HFCs are described in Table 1[12]. In Coast Province HFCs were strengthened through management training provided through the Ministry of Health and Danish International Development Agency (DANIDA) Health Services Project [13]. In 2005 HFC roles were expanded in Coast to include the management of a budget under the Direct Facility Financing (DFF) scheme, which was piloted in this province. DFF is an innovative finance mechanism to provide additional funds at facility level. It was introduced in response to a reduction in user fee levels in 2004, which significantly reduced facility income, and was argued to have undermined the relationship between committee members and communities ([14, 15]).

DFF funds were transferred directly into facility bank accounts, with public health centres and dispensaries receiving an average of USD 3,392 per year [16]. The Government continued to cover the vast majority of facility resources through in kind provision of infrastructure, trained health workers, supervision, drug kits and medical supplies. HFCs budgeted for the DFF funds within specific guidelines, implemented work plans and kept accounts. Funds could be spent on a wide variety of items including wages for support staff; maintenance of buildings, furniture and equipment; travel allowances; stationary; fuel and non-drug medical supplies. The wage bill could not exceed 30% of disbursed funds, and expenditure on HFC allowances and drugs was not allowed. DFF supervision was provided by the District Health Management Team (DHMT), with the team members most involved being the facility management nurse (FMN) and the district accountant. Additional oversight was provided by two provincial level accountants. The post of FMN was created to support links between facilities, the community and the district by strengthening the management of committees. This involved overseeing the selection of committee members, organizing training, and assisting committees in planning and continuously evaluating those plans.

DFF in theory has the potential to strengthen community accountability as it involves additional HFC training, additional support through the FMN, and provides HFCs with control of some resources. Details of the implementation and effects of DFF in general are described elsewhere [16]. This paper assesses the nature and depth of managerial engagement of HFCs at the facility level, and how this has contributed to community accountability in the context of the DFF financing mechanism. We investigate HFC characteristics and training, the perceived breadth and depth of HFC managerial roles, and the relationships between HFC members and both health workers and the wider community.

Data were collected between October 2007 and March 2008, 2 to 3 years after the introduction of DFF. Two of the 7 districts in Coast Province were purposively sampled to reflect diversity of experience with DFF implementation according to managerial views. Kwale, which was viewed as a relatively strong performer, is close to the provincial headquarters (Mombasa) and comparatively accessible. Tana River, which was perceived to have experienced more problems with DFF implementation, is 5 hours' drive from Mombasa and suffers from poor roads and infrastructure.

The sampling frame included all government health centres and dispensaries with at least one qualified health worker. Structured interviews were conducted with the health worker in-charge in a sample of 15 facilities in each district, including all health centres (5 in Kwale and 4 in Tana River), and a random selection of dispensaries (10 of the 47 eligible in Kwale, and 11 of the 25 eligible in Tana River). Health workers were asked about the facility, the functioning of the HFCs and about other community engagement mechanisms. At the same facilities we conducted structured exit interviews with a target sample of 10 community members seeking outpatient curative services per facility, obtaining a total of 292 completed questionnaires. Data were collected on patient characteristics and awareness of community engagement strategies.

A subset of 6 facilities from each district was re-visited for in-depth individual interviews with the health worker in-charge, and group discussions with a representative range of HFC members. The 6 facilities were purposively selected to include only those where the in-charge had been in post for at least 1 year, and to encompass variation in facility type, accessibility and indicators from the structured survey. Between one and nine HFC members participated in group discussions, often including the chair and treasurer. Finally, 7 in-depth interviews were conducted with members of the DHMTs: two District Medical Officers of Health (DMOHs), two FMNs, one District Health Administrative Officer (DHAO), one district health accountant and one provincial facility grants accountant.

In-depth interviews with health workers and HFC members included an exercise to stimulate discussion of the nature and depth of their roles in facility management tasks. To develop the exercise, we drew on frameworks that distinguish between depths of community involvement from simple information sharing with communities or their representatives, consultation with these groups, or - suggesting most depth of involvement - communities or their representatives having a tangible influence on health policy or practice [1, 9]. A number of managerial tasks were identified, including employment of professional health workers; employment of support staff; setting the level of user fees; deciding how DFF funds are spent; deciding how user fee funds are spent; and disciplining health workers. Six cards, each with one of these managerial tasks, were given out in turn and respondents were asked to discuss what they thought their roles were regarding the task. In each situation, respondents were asked to decide whether their role entailed (1) making the final decision, (2) being consulted but with the final decision made by someone else, or (3) having no role at all. We deliberately selected situations where we expected committee members to have a variety of roles based on government guidelines (see Table 1). These included some activities where we expected HFCs to make the final decision (allocating DFF funds; allocating user fee revenues; employing casual workers); and some where we expected them to have no role (replacing the facility in-charge; disciplining health workers; setting user fees).

Quantitative data were double entered using Fox-pro D-base IV, MS Access or MS Excel, and imported into STATA version 9 for analysis. We used the STATA svy commands to adjust for clustering at the facility level and stratification by facility type and district, and variation in sampling probability across facilities. Notes were taken during qualitative interviews, and where possible, interviews were digitally recorded. Discussions were transcribed and imported into N-Vivo 7 for coding and analysis. A coding scheme was developed from a thematic framework and from reading a sub-set of the transcripts to identify the main themes.

Informed consent was obtained for all interviews, and the study was approved by the Ethical Review Committees of the Kenya Medical Research Institute and the London School of Hygiene and Tropical Medicine.

We used a mix of quantitative and qualitative data to investigate the functioning of HFCs as a community accountability mechanism, drawing on the views of health workers, HFC members, patients and district level managers. We found that in a context where HFCs have control over some facility resources through DFF, and receive some training and supervision, they are generally functioning well in both sampled districts (though there were some problems, discussed below). Committees were in place in all facilities and meeting regularly. Selection of members had generally followed DFF guidelines, allowing community involvement and ensuring at least one female per committee. Members usually had good relationships with health workers, and were perceived to play an important role in representing community interests and helping oversee facility management. There were also indications of true depth of involvement with, for example, committee members and health workers agreeing that HFCs made the final decision about employment of subordinate staff and the use of user fee revenues. It is possible that some respondents aimed to present HFCs in an artificially positive light in order to encourage the continued flow of funds for DFF and health system strengthening in general, and the continued provision of HFC allowances! However, problems were also freely raised by respondents, suggesting that this type of bias may be limited.

Two factors were perceived to have facilitated these positive findings. First, FMNs who formed part of the DHMT had specific responsibility for supporting committees and for links between the community and the health service more generally. Their role appeared to have been very important in guiding HFCs in the early stages and in resolving any disputes that arose. Secondly, the introduction of DFF was viewed as having played a crucial role in empowering HFCs to meet, plan and implement activities, increasing their sense of ownership and motivation, and in many facilities transforming committees from dormant entities to active stakeholders. DFF both provided resources which the HFC could manage, and freed up user fee revenues which could then be used to pay allowances to committee members. Stakeholders indicated that without these allowances such high levels of HFC activity would have been very unlikely.

These findings are quite likely to be representative of Coast Province in general, given there was considerable variation between the two sampled districts in terms of accessibility, infrastructure, cultural practices, levels of poverty, and perceived ease of DFF implementation. However, it is unlikely that the results can be generalised to other provinces in Kenya, which had not benefited from FMNs and DFF, though the latter is scheduled to be introduced nationwide under the Health Sector Services Fund from late 2010. In addition, since the early 2000s, Coast Province had benefited from the broader DANIDA-funded programme of health system strengthening. This included training of health workers and managerial staff, construction and renovation, provision of equipment and supplies, strengthening the drug delivery system, enhancing community health activities (including seed funds for income generating activities and strengthening of HFCs), and strengthening of health management information systems (HMIS) and financial management [13]. While some other provinces have also received donor support, it has generally not been this comprehensive.

A number of important limitations were also noted in HFC operation. First, many respondents reported insufficient clarity in HFC roles, with variation in views across stakeholders on the extent to which they could exercise their powers. For example during the card exercise on HFC roles many different views were expressed on the appropriate depth of their involvement in decisions on how to spend DFF funds and in setting user fees. In both scenarios, there were lengthy discussions and in several cases an inability to agree on the HFC role. Lack of clarity in roles, including depth of involvement, has been shown to be an important challenge in other settings (for example Gilson and Erasmus 2006; Lowensen et al., 2004; Khumalo, 2001; cited in Molyneux et al.: Community accountability at peripheral health facilities, submitted). In our case, this lack of clarity may have been exacerbated by the brevity of training, the training of committee officials only in Kwale, and the lack of clear documentation on DFF at a facility level.

Lack of clarity on roles can have important negative implications for the relationship between HFCs and health facility in-charges, a relationship that is crucial to the sustained involvement of committees [5, 17] and ultimately their potential for positive influence on health facility performance. While relations were generally good, there were some indications in our study of tensions and mis-trust between health workers and other HFC members. However, to some degree this may be inevitable if HFC members are performing their oversight role appropriately.

As has been noted elsewhere [18, 19], there were concerns about the ability of some less educated members to fulfil the more technical roles such as budgeting. With the nationwide scale up of DFF, new guidelines have been gazetted for HFCs which restrict the number of community members to 5, one of whom must have knowledge and experience in finance and administration and three of whom should be women [20]. The rest of the HFC is to be made up of nominated members (the facility in-charge, a representative from the provincial administration, the District Medical Officer for Health or their representative, and the Area Councillor). The implications of these changes for effective community accountability are unclear. While greater education may improve capacity for technical roles, nominated members may inhibit community members from contributing confidently to discussions, and more highly educated community members may be less accessible to some groups of the general population. If the target number of women on HFCs is achieved in future, this may increase the prominence of issues relevant to women and children. However, as observed during our HFC group discussions, the presence of women on committees does not necessarily mean that they will speak freely in front of male members, especially in communities where this is not viewed as a traditionally female role.

HFCs are only one of the mechanisms used to enhance community accountability and engagement in Kenya, with others including public meetings, and the Government's Community Strategy, which aims to deliver services at the community level through community health workers and village health committees. However, it remains a concern that links between the HFC and the broader community are currently so inadequate. Although HFC members were clear that one of their roles was to represent the broader community in health facility operations, less than half of facility users had ever heard of the HFCs. Fewer still knew the chair or any other member of the committee, even if not by name. Women and less well-educated respondents were particularly unlikely to know about the HFC or know its members. Moreover, we only interviewed community members who were using the health facilities, and it is likely that those not using the facility were even less likely to be aware of HFC operations. Engagement by HFCs of the broader community at any depth - even simple information sharing - therefore remains weak. There were also a few reports of poor relationships between committee members and the wider community, including some lack of trust by community members in HFC members' handling of finances.

Strengthened trust and communication is likely to require greater transparency and clarity in HFC selection, roles and activities and on facility finances. Mechanisms to enhance this might include additional public meetings, disseminating information through CHWs or the use of public noticeboards at facilities. Noticeboards (or blackboards) had in fact been supplied to all Coast Province facilities, and at the time of data collection were clearly visible to clients in 25 of the 30 facilities. Facility staff were supposed to complete the blackboard table with monthly data on health and utilization (for example, number vaccinated, under-weight, births and deaths, etc), and accounts (income, expenditure, cash in hand, cash in bank). However, the current utility of these boards was unclear as very few were fully completed, it was unclear whether community members could interpret some of utilization data, and the financial information was limited to bank account totals with no information on how facility funds had been spent [16]. The boards might have been more useful in supporting community engagement if they had instead included the names of all committee members, a brief description of HFC roles, user fee charges, facility income from each source, and expenditure by category for each quarter.

There is potential for HFCs to function well and play an important role in some aspects of facility management, particularly in a context where funds are available that can be used flexibly, albeit within centrally defined guidelines. This motivates HFCs to meet and to contribute local perspectives to decision-making for the facility.

For the potential of HFCs to be met fully, it is essential that all actors are clear about the roles of HFCs, how committee members are selected, and - most fundamentally - the extent and limits of their powers in different areas. In our setting the key areas requiring greater clarity were depth of involvement in decisions over spending DFF money, and over increasing user fee charges; both of which are important areas in ensuring that health services are locally acceptable and accessible. Improved clarity in roles in turn requires strong training, documentation and supportive supervision from the district. Failure in any or all of these areas can undermine effectiveness of operations and contribute to conflicts.

In addition, HFCs should be representatives of a broader community, requiring appropriate selection from and links to those communities. It is essential that mechanisms for communicating with the wider community are carefully considered when establishing HFCs, and maintained over time, to ensure that accountability to the broader population is achieved.