Introduction

Collaborative governance as practiced in the health sector brings different stakeholders together for consensus-oriented decision making to address issues that affect service provision, responsibility over deliberations, and accountability. The structures for collaborative governance are provided by governments that legitimize and encourage the active participation of citizens on an individual basis and as representatives of specific entities [1, 2]. Collaborative governance provides an opportunity to support community empowerment and to contribute to and influence the goals and practices of services. This includes enhancing the capacity of services and community members to improve overall health and quality of life, to nurture cordial relationships between health facility staff, and to increase the extent to which services are responsive to users and their communities [3–5]. Although these collaborations are intended to ultimately contribute to responsive health care, there is limited evidence and lack of clarity about operational structures and processes needed for effective implementation [7–12].

Governance is a key determinant of performance of health care facilities [1–3]. Governance refers to the complex mechanisms, processes and institutions which enable actors to interact at constitutional, collaborative and operational levels to articulate their interests, mediate differences, and exercise legal rights and obligations in order to influence outcomes of public policy [4–6]. Health institutions characterized with poor governance face performance challenges, including in relation to their capacity to address or satisfy stakeholders concerns [3, 7]. One way of overcoming these governance challenges is involving citizens in the provision of the services they utilize. Both governance and citizen engagement can be looked at in a multilevel perspective. Ambiola and colleagues apply one such framework for primary health care governance at operational, collective and constitutional governance levels depending on influence. This present article focuses on collective governance level [8]. Governance attributes include accountability, transparency, participation, consensus orientation, equity and inclusion, effectiveness, efficiency, intelligence and information and power [6, 9]. ‘Good’ or ‘bad’ governance is measured in terms of the extent to which these attributes are implemented [1, 10].

Health care services in Kenya are organized into six levels of care. Level one is the community, level two is the dispensaries, level 3 is the health centers, level four is the sub-county and county hospitals, level five is the county referral hospitals, and level six is the national referral hospitals. Levels one, two and three constitute primary health care services, with formal facility based health care services provided at levels two and three. These primary health care facilities constitute the majority of Kenya’s health facilities that should be accessible to majority of the population and are essential to attain key health outcomes [11]. There has been global effort by various multinational and bilateral bodies since the Alma-Ata Declaration of 1978 on the effective implementation of primary health care services [12, 13]. Despite this, researchers have demonstrated that many countries in Low and Middle Income Countries (LMIC), including Kenya, have weak primary health care systems [14].

In 2010, Kenya initiated governance reforms to support devolution, following a referendum to split power between national and county governments. Devolution is a form of decentralization whereby authority is restructured and there is co–responsibility between national and county governments [15, 16]. The aim in limiting the role of centralized government was to enhance oversight, accountability, promote individual freedom and public participation as aspects of collaborative governance [28–31]. Devolution began to be implemented in 2013, when the newly elected government took office following general elections.

The constitution has many provisions for community participation, including participation in the management and oversight of public facilities. Counties receive 15% of equalization funds of the countries’ budgetary allocation to be distributed among the ten service departments at county level, including education, development, agriculture and health. The Department of Health receives funds directly or indirectly from Local Authority Transfer Funds (LATIF), Constituency Development Funds (CDF), Health Sector Service Funds (HSSF) and hospital management service funds. Engaging community members in governance structures, such as through health facility committees (boards) or other stakeholder fora, is set out in the constitution and is often a donor requirement, for example, to use Health Sector Service Fund (HSSF) [15, 17, 18].

The National Health Sector Strategic Plan II of 2014–2018, issued by the Ministry of Health, refers to the absence of accountability mechanisms on committees and facility boards, suggesting that there was already, at time of writing, a problem with planning mechanisms, transparency, monitoring, supervision and reporting by collaborators. Further, the document reported limited implementation of a rights-based approach in health service delivery. The plan emphasizes the need to establish mechanisms for collaboration with all health-related partners by holding stakeholder forums, creating committees and boards, and carrying out dialogue days with community members at facility level.

From the beginning of measures for devolution nationwide, all counties introduced policy documents, guidelines and proposed structures to implement collaborative governance. The Uasin Gishu Strategic Plan 2013–2018 emphasizes a rights-based approach and the need to ensure the inclusion of minority and marginalized groups in governance by conducting quarterly health stakeholder meetings and encouraging regular facility committee meetings [19]. In addition, the County Projects Implementation and Management Act of 2014 provided a legal framework to implement collaborative governance through health facility committees [19]. The document outlines the mechanisms for the establishment, administration and functioning of these committees. Under the act, the health facility committee should comprise of five to nine members whose roles and responsibilities are shown in Table 1. Community members from the catchment villages covered by a given health facility were to elect representatives, with the election process presided over by the public health officer from the facility and the area chief (the government administrator who handles community issues within a geographical boundary such as at the sub-county level).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Positions and responsibilities of health facility committee.

https://doi.org/10.1371/journal.pone.0248914.t001

The contribution of collaborative governance in terms of accountability, community participation and consensus orientation, transparency, equity and inclusion in the delivery of primary health care has not been fully realized in Uasin Gishu County. At the time of the study I report on here, the county department of health was unclear on the need for and contribution of collaborators, including the health facility committee members. A review of the literature reveals mixed findings on engaging communities in health, and in Kenya, few studies have been conducted. For example, McCollum and colleagues compared experiences of health system governance under decentralization in Kenya and Indonesia, and found weak accountability structures, limitations in community engagement and technological capacity in both countries [20]. Indonesia devolved the health system 15 years earlier, but still experienced challenges similar to Kenya, suggesting that the Kenyan government re-examine health governance to avert prolonged health governance challenges. Published studies on the implementation of governance at primary health care level reveal evidence gaps in terms of roles of social accountability, public-private partnerships and intersectoral collaboration. In this article, therefore, drawing on data from Uasin Gishu County, we examined the implementation of collaborative governance at primary health care facilities, looking at the structures in place for such operations as selection processes, the demarcation and undertaking of roles and responsibilities, and the challenges faced after implantation of devolution in Kenya in 2013. We illustrate the potential of collaborative governance to provide information to enable the implementation of more inclusive, responsive and accountable governance systems and sustainable collaborative efforts by all stakeholders.

Materials and methods

Study setting

The study was conducted between April 2016 and December 2017 in Uasin Gishu County, one of 47 counties in Kenya. This county was selected purposively for its strategic location in providing health services in Western Kenya [23], in an area that is politically volatile. hot spot. The county hosts the second largest referral hospital in Western Kenya and the neighbouring countries like Uganda, Sudan, Burundi, located in a metropolitan region, but serving both urban and rural populations. For administrative purposes, the county includes six sub-counties and 30 wards, with 146 health care facilities including 90 primary health care facilities [19, 24, 25]. The governance structure of health care services in Kenya is organized into three streams: partnership, governance and stewardship [2]. Partnerships involve stakeholders in health related issues in a given jurisdiction; governance involves different community members in decision making and the management of health facilities; stewardship relates to people working in the government health ministry itself and their participation in decision making. These are set out in Fig 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Governance structure for the Department of Health, in Kenya, national and county governments.

Source: Government of Kenya¹. Key: Shading shows from where participants are drawn.

https://doi.org/10.1371/journal.pone.0248914.g001

Study population

Participants were purposely selected from stakeholders including NGOs, professional bodies, private service providers, and the media, from health stakeholder forums, and community members of primary health care committees (Fig 1). Five primary health care facilities were selected purposively from different sub-counties, allowing for maximum variation and potential to provide rich information on interactions between facility staff and local communities. The criteria for facility selection included work load, location and potential to obtain information; senior officers of the Department of Health advised on this and facilitated entry of the first author (JS) (Table 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Primary health care facility.

https://doi.org/10.1371/journal.pone.0248914.t002

Results

i.) Demographic characteristics

The demographic profiles of the 60 participants representing individuals service providers and community members in collaborative governance are presented in Table 3. Two thirds of the participants were male (39, 65%), with almost half (27, 45%) aged between 30 to 40 years. Fifty percent (30, 50%) had a secondary level of education.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Demographic profile of the participants.

https://doi.org/10.1371/journal.pone.0248914.t003

iv.) HFC meetings and agenda

According to the Projects Implementation and Management Act, a maximum of eight meetings should be held in a financial year, with five members constituting a quorum; an allowance of Ksh. 500 (US $5) per head to cover lunch was set aside for those who attended. The policy document stipulates that the County Executive Committee (CEC) member for health needs to grant permission for extra sitting days to discuss issues related to the day-to-day functioning of the health facility, because of financial implications. In the study period, health facility committees met between seven and 12 times in a financial year, hence slightly more often than stipulated. The required quorum was not reached in seven instances across the facilities, but meetings proceeded. Only one facility publicly posted the year’s schedule of health facility committee meetings, ensuring that they were open to public involvement and awareness of meetings. Fig 2 shows the frequency of agenda items at these meetings.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Frequency with which the issues were discussed at HFC meetings.

Source: Research data, 2017.

https://doi.org/10.1371/journal.pone.0248914.g002

The most common items were financing, including budgeting, payment of contractors, cash withdrawals and change of signatories. Discussions on projects included follow up of ongoing construction at the facility, such as building a maternity wing or a new ward, renovating rooms, fencing, painting and repairs. Other agenda items included the purchase of land adjacent to facilities, purchase of a water tank, quotations for and purchase of an ambulance, and the sale of maize grown on common land adjacent to the facility. At one facility committee meeting, there was an extended discussion on the formulation of Authority to Incur Expenditure (AIE). Observation of the meetings, review of documents, and interviews confirmed this focus of items.

The majority of respondents revealed that there was a recurrent shortage of drugs at all facilities, reflecting community frustration with constant stock outs.

Days like Monday or Tuesday, there are a lot of people. You may find that there are no drugs. From the county, the drugs arrive very late. The patient may have to go and buy prescribed drugs yet not all of them may afford. Sometimes we go and borrow drugs in the neighbouring facilities. We try and maintain our facility though it reaches a breaking point. If it is an injection, you have to go and buy before you are injected. It is cumbersome (Female respondent, 30–40 committee member).

The procurement process is long and tedious, like the jargon, but we hope to get used to this by way of training and exposure of the committee members. We have cases of misuse of funds by some of the committee members since they are used to the old ways of management of resources including the deliberation that is done by a few people who are not members of the community (male respondent, 30–40 years).

Discussion

The study examined the implementation of collaborative governance in primary health care facilities in Uasin Gishu County, Western Kenya, under the policy of devolved governance of 2013. Although there are documented structures to support collaborative governance functioning, the operationalization, functioning processes, roles, and attributes of the office bearers were unclear. This is not uncommon. Other scholars have also noted that community health committees are limited by lack of clearly defined roles and responsibilities [15]. In The Philippines, for example, similar challenges of collaborative governance on local health boards have been documented [36], including conflicts between authorities and other stakeholders in relation to their respective roles, and boards were not always valued by municipal health officers. Some board members felt that they lacked influence to make health plans. This frustrated efforts of collaborative governance and intimidated community members.

In the study setting, the involvement of community members as consumers and providers in decisions about service were central to the provision of health care. This is consistent with the recommendation of the World Health Organization that community members have a right and duty to participate in planning and the implementation of their own health services [37]. This remains a focus for the attainment of universal health coverage. The legal documents required representation on the committee from women, youths and persons with disabilities, reflecting the 2010 Constitution of Kenya (Art. 27), which specifies that vulnerable, previously discriminated and marginalized groups be included in governance in all public sector committees [28, 38–40]. However, at times several rounds of nominations to health facility committees were necessary to ensure such representation. Decision makers found involvement of members of marginalized groups to be minimal, as we have reported elsewhere [23]. This suggests that more effort needs to be made to enhance the inclusion of such groups. Further, at the time of this study, HFC membership had been extended to a new group of members—administrators representing the county government—whose roles and responsibilities were not clearly outlined.

At county level, HFCs were established under the Projects Management Act 2014, but the act provided contradictory information on the technical skills and requirements for individuals to be committee members. This included the specification that an O level was the minimum education qualification, but also that nomination was open to anyone without secondary education with good social repute and past experience in managing community institutions. The capacity of members of committees to actively participate in the deliberation of issues requires literacy, since members need to be able to read supporting documentation, and hence the effectiveness of committees can be compromised if individuals cannot make informed decisions, or comprehend and critically analyse health care and financial issues. In a study conducted in Kilifi, Kenya, Goodman and colleagues indicated that 34.1% of committee members had not attended school, while 34.2% and 41.3% could not read Kiswahili and English respectively [38]. The authors argued that the high levels of illiteracy among HFC members hampered their decision making abilities [38]. In this present study, in contrast, only one member of all HFCs had primary level schooling as the highest educational qualification, and all others had a least some secondary schooling; in theory therefore committee members were able to participate in an informed manner.

The study findings show that, at face value, the election processes were conducted in an open and transparent manner that enabled community participation. Localized word of mouth and posters were used for communication. However, although these were the most common modes of communication, they appear to have been inadequate and ineffective to encourage community members to participate since turnout was always less than half of the expected population. This suggests lack of interest in HFC and/or the ineffectiveness of communication, and concurs with a study conducted by Odini [41, 42] in Western Kenya, in which people in rural areas appeared to lack interest in information aimed at fostering their participation in community service provision projects [40, 41].

As described in (Table 4), the executive members of all HFCs (chairperson, secretary and treasurer) were trained on how to run committees and on the Public Finance Management Act, and these trainings empowered them. In Kilifi, Goodman and colleagues documented that members who were more educated and trained on health facility management exhibited more authority and tended to intimidate the less informed members of the committees [38]. Lodenstein and colleagues, in exploring the role of health facility committees in Benin, Democratic Republic of Congo and Guinea [31], noted that the operations of HFCs were individualized and not systematic, with minimal or no community consultation. This led to the exclusion of marginalized voices, inability to provide feedback to communities, and difficulties working with formal administrative structures to articulate health facility issues [31]. In the study reported in this article, there were differences in practices of power and an imbalance of participation among committee members. Facility in-charges, for example, held a lot of power, although not all exercised this effectively. Since executive members were trained on the Public Finance Management Act, they possessed technical information on running facilities and displayed greater power in decision making than other committee members. Choonara [37], studying the importance of on-the-job training and leadership development in South Africa, also found that training helped develop leadership and financial management skills in primary health facility teams. Ansell and Gash [43] have argued that power imbalances are a common problem among stakeholders in collaborative governance, and that measures should be put in place to empower those who are weaker and poorly represented [29, 43–45]. As stated earlier, leadership is critical in facilitating broad and active participation, ensuring broad influence and control of group dynamics [43].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Codes, categories and themes of election process and challenges of implementing collaborative governance.

https://doi.org/10.1371/journal.pone.0248914.t004

The United Nations’ SDG Goal 17 encourages the engagement of partners. The policy documents which support the health care framework envision attaining improved health care through stakeholder collaboration [46, 51, 52]. Stakeholder participation in decision-making and the implementation of health care services was emphasized by respondents as important in collaborative governance at multiple levels of the health system [33, 47–52]. While stakeholders played a positive role in primary health care provision, there were no clear mechanisms for controlling or regulating their entry, operations and exit [50]. In Uasin Gishu County, stakeholder engagement was the direct responsibility of the Chief Executive Officer at the county level [10], and therefore this officer should have been able to put structures in place for effective collaborative process.

Conclusion

The facility committee members contributed to discussions in relation to construction projects as opposed to the day-to-day functioning of the facility, and meetings were most frequent when project funds were discussed. Committee members with political connections and retired government workers had most influence and power. Lack of clarification of roles and responsibilities, conflict of interests, elaborate structures for stakeholder involvement, and variable competence of health facility committee members were the major challenges to enable collaborative governance. There was weakness in the inclusion of members of marginalized groups, and lack of clarity around processes involving stakeholders, the duplication of projects, and other difficulties associated with the provision, management and sustainability of activities. This suggests that structures for stakeholder involvement and health facility committee functioning need to be strengthened to support the effective implementation of collaborative governance.

Supporting information