Health Committees as Vehicles for Democratic Governance in Health Systems: Lessons from Selected Health Unit Management Committees in Uganda


 Introduction: For over forty years, community participation has been a central component of a well-functioning health system. Despite its important role there are many difficulties in defining and understanding community participation as part of governance. Through a case study of selected health unit management committees in Uganda, this paper demonstrates that these committees can be structures for community participation and vehicles for democratic governance in health systems that advance health equity. Guided by the theoretical underpinnings of deliberative democracy the paper evaluates the performance of health unit management committees as a mechanism for citizen participation in health systems. Methods: This paper uses a qualitative, case-study methodology. Through an in-depth look at the health unit management committees of Kiboga and Kyankwanzi in Uganda, the study considered these as examples of structures for democratic community participation in health system. The study undertook literature review on the theories of deliberative democracy and human rights principles, and this provided the theoretical underpinnings of the study. Findings: Our findings underscore that community participation in health systems through health unit management committees ought to be grounded in the principles of deliberative democracy. The core of deliberative democracy is considered to be authentic deliberation and consensus decision-making, which can happen in both direct and representative democracies, giving rise to the notions of populist and elitist deliberative democracy, respectively. As such, a balance needs to be struck between the competitive notions of democracy and the public health requirements of inclusive and direct participation of communities in decision making processes on matters that affect their health. Conclusions: Community participation in the health sector in Uganda hinges on health unit management committees at the lower service provision points. These HUMCs are also perceived as vehicles to strengthen health governance through realizing the right to health of the communities. However, these have been established without attention to investing in capacity building needed to enable them to exercise community voice in the health system.


Introduction
For over forty years, community participation has been a central component of a well-functioning health system because of its capacity to make health policy re ect the needs of the people it is there to serve. (1) First highlighted in the Alma-Ata declaration for Primary Health Care (PHC) of 1978, and last revisited in the 2018 Astana declaration, it should refer exclusively to the active engagement of individuals, families, and communities in the process of making decisions around health policy. This promotes good health system governance (2) because it contributes to making the process of developing, implementing, and evaluating health policy center around people's needs and priorities. (3) In addition, it is a central component of the right to the highest attainable standard of health (4) and a precursor of other social, cultural, and economic rights. (5) In the context of the right to the highest attainable standard of health, (4) individuals and groups are entitled to active and informed participation with government in health-related decisions that affect them. This includes participation in identifying and de ning overall health strategy, agenda-setting, decisionmaking, prioritisation, implementation and accountability.(6) Health policy that is responsive to the health needs of particular groups, for example, women, children, adolescents, indigenous and ethnic groups, the elderly, and sexual minorities, cannot be achieved without their active and informed participation.(6) Community participation in health has moved beyond being a public health issue, and is increasingly becoming a democratic governance and human rights subject.
Using Habermas' theory of deliberative democracy, we study how Health Unit Management Committees (HUMCs) in Uganda can serve as vehicles for democratic health system governance through the study of governance processes from Kikoolimbo and Nyamiringa Health Unit Management Committees in Uganda. Deliberative democracy, an intrinsic component of democracy, emphasizes the right and obligation of every citizen to participate directly in the business of governing. (7) Achieving this requires the existence of a free citizenry that has the moral power to form, revise, and pursue a rational conception of the good; that is able to self-authenticate sources of valid claims; and is capable of taking responsibility for their ends.
Despite the central role that participation has in human rights and within the health system, there are many di culties in de ning and understanding community participation. Habermas's model of deliberative democracy combines a procedural account of democratic legitimacy with deliberative politics.(8) He underscores the importance of specifying the ways in which the normative idea of popular sovereignty should be maintained and feasibly implemented in recognisable ways under conditions of modern social complexity. In providing further guidance on the concept of deliberative democracy, Habermas proposes that four key elements are critical. These include communicative action; the power of speech; public sphere; and the principle of legitimacy. (9) Communicative action requires a consensus that is built on informed reasoning and participation. (9) The power of speech principle is built on the importance of developing the private autonomy of individuals, a precursor of the type of public autonomy that can be applied in voting or choosing public representatives. The public sphere, on the other hand, calls for a discursive space where individuals and groups meet and discuss matters of mutual interest and, if possible, reach a common judgment. (10) The element of legitimacy is based on three understandings: that people who are asked to obey the authority feel they have a voice when they speak up; that the law is predictable; and that the authority is fair and does not treat people differently based on groupings. (11) Su ce to note that rule of law, transparency, responsiveness, consensus orientation and accountability are all at the heart of good governance and are all in uenced by community participation processes.
Uganda's health services delivery system is pegged to the country's decentralized system. The lowest level of health service delivery is provided by the Community Health Extension workers (CHEWs) who close the gap between communities and the formal Health system. The CHEWs play an important role in referral, health promotion and disease prevention at community level and their effective involvement leads to better health outcomes such as reduction in maternal mortality. (12) At the district level, there are health centers II, III and IV. Health center II provides a rst level of interaction between the formal health sector and communities. These provide outpatient and community outreach services. Health center III provides basic preventive, promotive and curative care and support supervision of the community and Health Centre IIs under their jurisdiction. There are provisions for laboratory services for diagnosis and management of maternity cases at these facilities.(13) Health center IV provides preventive, promotive, outpatient, curative, maternity including Comprehensive Emergency Obstetric and Neonatal Care, in-Comprehensive Emergency Obstetric and Neonatal Care, patient, laboratory, ultrasound, emergency, blood transfusion and mortuary services. (14) HUMCs are structures that are created at the lowest levels of public health facilities run by local government to ensure that communities are empowered to play their role, take responsibility for their own health and ensure that they actively participate in the design, planning and management of health services. The composition, roles, responsibilities, and management of the proceedings of the HUMCs are set out under the guidelines developed by the Ministry of health.(15) While HUMCs are structures at the lower health facilities, the hospitals have a similar structure that is called the Hospital Management Boards. (16) The purpose of this paper is to demonstrate that the health unit management committees can be structures for community participation and vehicles for democratic governance in health systems. Guided by the theoretical underpinnings of deliberative democracy and human rights the paper evaluates the performance of health unit management committees as a mechanism for citizen participation in Uganda's health system. Through this paper we explore the experiences on the actual practice and how these structures act as the voice for communities in relation to the health services. The paper presents ndings from eld work which looked at opportunities for best practice in utilising community participation as a vehicle for realising health rights and building the agency of community structures to articulate more strongly claims for health rights, with a view to proposing models for best practice on democratic governance in health systems.

Methodology
This paper uses a qualitative, case-study methodology. (17) Through an in-depth look at the health unit management committees of Kiboga and Kyankwanzi in Uganda, the study considered these as examples of structures for democratic community participation in health system. The intention was to obtain an indepth appreciation of health unit management committees as both mechanisms for citizen participation and as a vehicle for realising health rights and building the agency for democratic governance in health systems. The study undertook literature review on the theories of deliberative democracy and human rights principles, and this provided the theoretical underpinnings of the study.
The empirical ndings presented in this paper are drawn from case-studies of an action research project titled 'Health System Governance: Community Participation as a Key Strategy for Realizing the Right to Health', funded by the International Development Research Center.(18) The project explored the potential of community participation in Uganda and South Africa to address inequities in health and promote health governance systems that give voice to the poorest and most marginalized populations. It was conducted between 2012 and 2015 and explored the experiences and perceptions of community stakeholders with regard to HUMCs as an institutional structure for their participation in health governance. (19) In particular, the project explored the mandate and status of these health unit management committees, as well as community and stakeholder perspectives on, and commitment to, either participation or facilitating community participation in health governance through health unit management committees. (19) 3. A purposive sample of fteen (15) key informants were selected to participate in semi-structured interviews using a guide developed to explore the aspects in the community that affect participation and entitlement to health care, as well as the challenges that affect community participation in Uganda's decentralized health system. They provided insights that could not be obtained with other methods, including information about how incidents, local happenings, or conditions at the district level help in understanding problems around the implementation of community participation initiatives. (20) Table 1 presents an overview of the informants. Each interview was carried out in a venue that provided privacy and was accessible for the interviewees. The interviews were conducted by MM with assistance from a research team trained especially for this eldwork. During the interviews, all processes were explained in Luganda, the local language by MM. The interviewees were assured of con dentiality and that they had a right to withdraw from the study at any time if they felt uncomfortable without any explanation or consequences. MM also sought permission to tape record each interview. Each one lasted between 45-90 minutes. Field notes were taken as backup and interview transcripts were given back to the interviewees for veri cation. All key informant interviews were tape recorded, and later transcribed verbatim by an experienced research assistant trained in the basics of qualitative methodology.

Focus Group Discussions
A total of six focus group discussion (FDG) were carried out. Their size varied from ve to twenty-ve participants and they were conducted using a semi-structured guide. Table 2 presents an overview of each FDG carried out. The participants included members from Health Unit Management Committees, community members, district political leaders, and technical staff from the health system. Discrete groups were created, and careful consideration was paid to the composition of each group. This included taking into account the number of groups needed to be convened to cover an issue adequately, and which combination of individuals in each focus group worked best. (21) One of the advantages of focus group discussions is that they do not discriminate against people who cannot read or write and they can encourage participation from people reluctant to be interviewed on their own or who feel they have nothing to say. (22) Indeed these group discussions stimulated new ideas among people which could have been missed in the individual interviews.

Six men and 3 women
As with the KIIs, MM identi ed quiet and accessible venues and contacted the participants to remind them of the meeting the day before the community dialogue would take place. The researcher made it clear that participants had the right to withdraw from the discussions at any time without any explanations or consequences if they felt uncomfortable and also sought permission to tape record the discussions.
Discussions lasted up to one and half hours. Discussions in Kampala were conducted in English, but those in rural communities were conducted in Luganda, one of the main local languages in Uganda. This is so because it is a rural district, and chances are that some participants may have found it di cult to communicate in English. To avoid patronizing, the researcher rst asked the participants the mode of language they would be comfortable with prior to starting the discussions in order to make sure that all participants were given opportunities to share their views.

Community Dialogues
As part of the research process, two community dialogues attended by over 100 people each were also held separately for communities around the two health centres. The dialogues were used to discuss the functionality of health unit management committees and to collect community views on their roles and expectations as bene ciaries of the committees. These community dialogues lasted for ve hours each.
The dialogues were recorded and transcribed.

Data analysis
All relevant documents and transcripts were read until the researcher was familiar with the emerging issues. A thematic content analysis was carried out. The rst step was to review the relevant policies and guidelines, carefully reading of the existing literature and studying transcripts from the eld interviews. The data from the literature and interviews was helpful in developing a coding structure for the interviews. This included speci c codes and categories linked to the concepts of democracy and community participation through HUMCs. This allowed for theoretical guidance. The second step involved careful reading of transcripts from the KII, FDG, and community dialogue, after which coding using both emerging and a priori codes was carried out. These were then clustered into emerging and a priori categories. Finally, themes were identi ed and presented here as individual ndings, and illustrative quotes were chosen to accompany theme narration, which include the analyses from our informants and from the document review. Issues of validity and conformability were considered during the data analysis by checking and re-checking with participants for them to con rm or disapprove our interpretations.

Ethical clearance
Ethical clearance was obtained from the National Council for Science and Technology under permit number SS 2731.
Case Unit Description

Participation in health in Uganda
In Uganda, community participation is seen as a mechanism for empowering and encouraging the active participation of all citizens at all levels in their own governance. The National Health Policy (NHP) states that government will continue to ensure that communities, households and individuals are empowered to play their role and take responsibility for their own health and wellbeing. (23) Community participation has also been incorporated as a democratic principle under the National Objectives and Directive  Nyamiringa Health Centre III is located in Kiboga district in Nyamiringa sub-county. While the ministry documents indicated that the facility has a total of 9 staff (28) on the ground it was established that facility had three health workers and two causal labourers. This is far short of the nineteen approved sta ng levels of a health centre III (28) As a health centre III Nyamiringa Health Centre should have these nineteen staff, led by a senior clinical o cer and it should also have a functioning laboratory.
First, by the time of this research, although Nyamiringa facility had a well-constituted committee, they had not had any meeting either as a committee or even with the health centre in-charge. In fact, since the new facility in-charge had been posted to the facility about ve months ago, the committee and the new in-charge had never met and so did not know each other. The in-charge at this facility highlighted that one of the major challenges was that community members preferred to seek services from the district hospital even for services that could be provided at the facility even when the district hospital is far away from this community. In the community dialogues held at Nyamiringa health facility, the community members on the other hand indicated that health workers are continuously absent and sometimes the facility is locked when they need services.
During the research process, we noted that this facility had a maternity ward that was not functional due to lack of water supply, bathrooms and toilets to support its operation. It was the position of the facility in-charge during an interview that this maternity ward would not be opened until all the required facilities to support its functionality were put in place. The facility also had a pit latrine that had been damaged as a result of heavy rains. However, no mobilisation had been done within the community to address this issue.
We also got to understand that a key characteristic of this community is that it is a mainly pastoralist community. During interviews with the HUMCs members, it was evident that the response to this community's calls to participate in activities at the health facility was low. Most of the families together with their children preferred to attend to their cattle in the elds. During interviews, the committee members indicated that the community also has key days during which major buying and selling of cattle takes place and on such days response to health programmes or even services like immunization every Wednesday is low. We also got to know that this facility does not have any form of accommodation for the health workers and professionals. They instead have to travel a long distance every-day to the health facility. Also, this means that any emergencies that happen late in the night cannot be attended to at the health center, which also does not have an ambulance to transport patients to the main hospital. These are the kinds of challenges that the health committees would engage on with the districts authorities as they affect their functionality.

Kikoolimbo Health Unite Management Committee
Kikoolimbo health centre is located in Kyankwanzi district's Wattuba Sub-county. Just like Nyamiringa, this facility was also in transition to upgrading from a health centre II to the level of health centre III. This meant it would be required to provide a wider range of services. It is also important to note that Kyankwanzi district was curved out of Kiboga district and thus separated administratively from mother district. It is therefore a fairly new district, implying that most of the health system structures are still being put in place. For example; the former level IV in this district is also in a transition to become a district Hospital since Kiboga hospital that was originally serving both areas is now located in another district. This partly explains why this Health Centre has also had to be upgraded to a Health Centre III status.
Since its opening, this Health facility had only one health unit management committee which had served for about ten years. Although it started off with a fully constituted health committee, it had six members left out of the nine required members. It also did not have any record of previous meetings and or minutes recorded during committee meetings, neither was there a record of interaction between the committee and the service providers at the health facility. The health unit management committee needed to be reconstituted for various reasons including the facts that: it was not functional and had not met for months; it was not constituted as required by health unit management committee guidelines and that there was a lack of understanding of the roles by its members.
Another key aspect that was noted during the interviews with the health unit management committees is that the land on which the health centre is constructed was donated by a member of the community who also mobilised key leaders within the community during the process of its construction. This particular member felt that serving as committee chair was his permanent reward for the donation made. As such, no attempt had been made to change leadership of the committee. Other committee members also had held onto their positions on the committee given the prestige that is attached to being a committee member.
This health centre also had challenges relating to the social determinants of health, which though not directly linked to service delivery, can affect delivery of services. These included: the facility was understaffed with only one midwife and one clinician far below the standard of nineteen and further lower than what the government documents indicated as fourteen staff deployed (30) yet it was expected to be delivering a wider range of services especially maternal health services; secondly, the lack of electricity at the facility to support health service delivery especially during the night was a challenge as the facility only had access to a solar panel that cannot last throughout the night. Third is the nonavailability of water at this facility where a nearby borehole was no longer functioning; and lastly is the state of the housing for the health workers and professionals at the facility. The housing is insu cient to accommodate all the facility staff. This results in complaints that health workers arrive late to start work at the facilities, as well as complaints by staff about the conditions of their housing. All these are challenges that would make the health unit management committee inactive.

Result One: Limited Community Roles and non-representation in HUMCs
The community members in this study reported not being aware of who represents them at the health unit management committees of the health facility nearest to them. This made them feel that neither they nor their interests were represented by the health unit management committees. Community members who participated in the community dialogues held at the two health facilities indicated that they were not aware of the existence of these structures and, in most cases, were not even aware of who were the members serving on these committees. The community dialogue was an opportunity to address this identi ed barrier to participation, and to inform the community of what they should expect from the committee and the key activities within their work plans. This community dialogue acted as a space for participation and empowerment which is a key aspect of the rights-based approach. As one community member stated: 'I don't think that those committee members are meant to represent me and my family at this health facility. But I know that they are big people and the community and have authority. During all the social functions they have a front seat and are given a microphone to greet us. … how can a person I don't vote represent me? I think that other big people in at the district sent them here to oversee their own interest but not our interest'. (Community member, Nyamiringa Health Center).
For each of the target health committees, a capacity assessment process was undertaken. They included a capacity assessment tool developed to help in establishing the health unit management committee's understanding on issues including their selection, mandate and composition, the mechanism for their performance, the reporting and feedback mechanism, and the challenges and recommendations for dealing with these challenges. During the assessment, it was found that there was limited knowledge of the Ministry of Health's guidelines for HUMCs, particularly on the roles and responsibilities of committee members, as well as among community members, local leaders, and other stakeholders. The health unit management committees at Kikoolimbo HCIII and Nyamiringa HCIII were not very active largely due to limited knowledge of what was expected of them. The capacity assessment process was key in informing the groups through the reading and interpretation of the guidelines verbatim in a participatory methodology. This did not only introduce the groups to their purpose, mandate and functionality mechanisms, it also generated debate on how the few times they had met things has been done incorrectly.
'Do you mean that I have authority to appoint members to the HUMCs committee? This has never crossed my mind, for all the years that I have been the LC 3 chairperson of this area. I thought that these committees are occupied by the families of people who donated land to these facilities. If this is true, will you help me interpret these guidelines when I next organise a community meeting? If it's true that I indeed have powers to do these appointments, I should now make it a priority. Our current Committee chairmanship has been occupied by family members of those that donated land on which the health facility was built. The chairperson has been chairing the committee for close to 20 years now. Other members of the committee were sent to us from the political leaders. We have no idea how these get to be sent to our facility. (Local leader in Kiboga district) It was noted that although health unit management committees are a participation structure, there is no mechanism or process within the guidelines that require these committees to report back and account to the members of the communities they represent. This was a critical issue because accountability is a key component of a rights-based approach. However, it was clear that communities were not receiving feedback from their reporting and this may be one of the reasons communities did not support the HUMCs to enable them be functional. Additionally, the Health Unit Management Committees also mentioned that they thought they were answerable to their appointing authority, which is the political leader at the respective sub-county level. This clearly dominated the understanding of the committees and necessitated creation of awareness about their accountability to the community as opposed to the appointing authorities.
In addition, the guidelines on the representation on the committee are not clear. Although there is a requirement for women representation on the committee, there are other critical groups that do not have any special provisions for inclusion, such as disabled persons and youth. As a result, they are poorly represented on the committee. This presents major challenges in advancing their health rights.
'I am a person with a disability, and I make use of this facility all the time and this is why I had to attend this dialogue, which I think is the rst I have seen that is not discussing politics and campaigns. While I see disability representation in many other spaces of political representation, we have not seen a space where disability issues are channeled as part of this facility. Women with disabilities particularly face challenges when they come here to deliver, some shy away and use the traditional birth attendants. Now that you are talking about representation through the HUMCs, the next appointment needs to be mindful of special people like us. (Participant in community dialogue at Kikolimbo health center).
As part of the values of the health unit management committees, they should be able to call for the protection of self-determination as an important ingredient in enabling the community to decide the rules by which they live and by which they are bound.

Results Two: Devolution of Power and Implications for HUMCs
The health unit management committees of Nyamiringa and Kikoolimbo operate as part of the overall devolution of powers to local governments in Uganda. The 1997 legislation on Local Government (LG) was enacted to put into effect the provisions of the Constitution that devolved powers previously exercised by the central government to the district local authorities. In line with this decentralisation, health care was also redesigned with a corresponding health unit level for each level of local government or administrative unit. The Act mandates the district chairperson to assign one of the secretaries to be responsible for health and child welfare. The law further creates executive committees at each parish and village administrative unit, which should consist of a number of people including a public health coordinator. The second schedule of the Act super cially highlights health policy as one of the functions and services for which the government is responsible.
On the other hand, district councils are responsible for the medical and health services including hospitals, other than hospitals providing referral and medical training, health centres, dispensaries, subdispensaries and rst-aid posts; maternity and child welfare services. HUMCs are seen, at least in o cial policy, as essential elements for decentralization.
The entire idea of HUMCs is about actualising decentralisation. We at the Ministry can no longer be able to effectively oversee the realisation of health at the lower district levels without support of the local government. Despite a few challenges like nancing, we have showed the will and interest to fully devolve power and service delivery as demonstrated in the policies and political positions created. The HUMCs are just an example of such structures. (Interview with the Commissioner from the Ministry of Health).
For proper functioning of these health unit management committees, the Ministry of Health developed a set of guidelines to be followed for operationalising these committees. The guidelines of health unit management committees are for health centres at levels two, three and four. According to the guidelines, the HUMCs for health centres two and three are appointed and approved by local council three. On the other hand, the health unit management committees for health centre four are nominated by the district health committee with the approval of the district council. During the study, the research team engaged Kyankwanzi district local council III chairperson regarding the health unit management committee that was not well constituted. He informed the team that he did not know that it is at the discretion of the local council III Chairperson to appoint and the council approves the health unit management committee members. By the end of the project, he had exercised his powers under the guidelines and had participated in the orientation process which the research project supported.
It was further revealed that although the guidelines make mention of the role of the Committee in budgeting and planning, during the assessment it was pointed out that this was not practical since most of the resources received at the health facility were already pre-determined at another level unknown to the committee members. As such, even if mechanisms and spaces for participation identi ed during the rst two years encouraged bottom-up planning, some aspects, particularly budgeting, remained a topdown process that undermines this function as contained in the guidelines and therefore undermines the roles of the committees.
'The funds that come to us as HUMCs is under the primary health care budget. This budget already predetermined and no matter how much work we need to do, we don't have much opportunity to change. It possible to observe and hear many cries in our community about some issues impacting in the community health but the limitations that come with the budget make it complicated to support these initiatives. You have seen that our maternity ward was opened without a placenta pit but we can´t do much as a committee. We need to have a more transparent and non-xed way of engaging with the budget cycle. The funds needed should cover both our processes and the needs of the community. It's very disempowering for the committees to have assumed authority that is not supported by funds.' (Interview with a member of the HUMCs Committee at Nyamiringa)

Result Three: Legitimacy through a clear legal framework for HUMCs
Providing a legislative authority as a basis for community participation through the HUMCs was considered important in both our document review and by our participants. Despite the existence of guidelines and policies meant to ensure fair and equal participation, and providing advice on what each role entails, the participants from Nyamiringa and Kikoolimbo report that the HUMCs have no speci c operational legal framework to effect their work. They, however, seek guidance from the local government and the Ministry of Health Guidelines for Health Unit Management Committees.
HUMCs are an important part of the health system, unfortunately they lack a substantive law which is a fundamental gap, and this goes to the core of there existence and operations. The guidelines provided by the Ministry of Health are very brief, unclear and not re ecting the reality on ground. These guidelines are founded on politics and less on democratic values. They assume that the appointing authority will have the mind and understanding of the health system. They reduce the community members to spectators and yet they are the bene ciaries. (Interview with a key informant from an national NGO).
The Local Government Act (31) seeks, among others, aspects to amend, consolidate and streamline the existing law on local governments in line with the Constitution and to give effect to the decentralisation and devolution of functions, powers and services. This Act provides for decentralisation at all levels of local governments to ensure good governance and democratic participation in, and control of, decision making by the people. Despite these promising provision, the Act falls short of recognising health unit management committees as a community participatory structure on matters of health. This is a missed opportunity because, according to the committee members, there are many con icts of interests that limit accountability pathways and hinder a community's ability to complain about service provision.
'The other problem with the existing guidelines is the requirement that the in charge of the health facility should be the secretary to the HUMCs. We have found this a serious governance problem in the operations of the HUMCs. At many health facilities, the in-charges don't have an understanding of their roles as secretaries and their unique position as overall supervisors of the health facilities also makes it complicated for them to participate in the meeting, take note, treat patients as they come during the meetings and then keep the records of the meetings. Because of these complexities, HUMCs end up without minutes and follow up on discussed matters is complicated. The in-charges are less bothered because this function does not usually form part of their assessment'. (Interview with a key informant working as a program advisor to an international organisation) The guidelines are also not clear on a procedure that happens after Committee meeting minutes are written. When asked what happens after meetings and what is done with the meeting recommendations, different committee members also had different answers and mentioned different duty bearers to whom they thought meeting follow-ups were taken. For example, some mentioned the chief administrative o cer, and others the District Health O cer (DHO). The absence of particulars in the guidelines on how meeting deliberations should be followed up makes it complex for the committees and hence affects their capacities to engage beyond the meetings in some cases.

Discussion
The ndings as presented in this paper underscore that community participation in health systems through health unit management committees ought to be grounded in the principles of deliberative democracy. As noted in earlier research, deliberative democracy in its essence is not a new concept. It is perhaps as old as democracy itself. The core of deliberative democracy is considered to be authentic deliberation and consensus decision-making, which can happen in both direct and representative democracies, giving rise to the notions of populist and elitist deliberative democracy, respectively.
As such, a balance needs to be struck between the competitive notions of democracy and the public health requirements of inclusive and direct participation of communities in decision making processes on matters that affect their health. Through this approach, individuals and their communities collectively enjoy meaningful participation in making of decisions that affect their health and access to health services. Achieving this level of community participation can be through a number of ways, including the formulation of policies and legislative frameworks that are grounded in the tenets of deliberative democracy. The policy frameworks can develop a transparent and interactive process by which the community's speci c roles are clearly de ned, each representative is perceived as a valid representative of the community, capacity is built for engagement with the health system, and the health system is responsive to community concerns Self-determination is an important component of community participation and of an equitable and responsive health system. Community participation through the health unit management committees must be premised on the recognition of each person as a valid speaking partner with a unique and valuable knowledge to contribute. As such, respect for the inherent dignity of persons and selfdetermination must inform all participatory processes and strategies of the committees and each person's expertise, experience and input must be valued. Local ownership and community context should inform the committee decisions. This calls for efforts to examine the history and diversity of the community as important elements that will shape effective and e cient community participation through health committees. Respecting local knowledge, ability of the communities and accepting their potential is key for the work of the committees. Overall, participation should go beyond mere consultation into requiring investment through providing information, building community capacity and creating public mobilization and awareness. In achieving this, the notion of deliberative democracy should be put in practice through operationalising the functionality of health committees with the tenets of the power of speech, public sphere, communicate action and legitimacy given prominence.
As emphasised in earlier literature, the membership of the health unit management committees should be by way of popular vote, which should be organised through elections that are based on universal and equal suffrage and should be held by secret ballot. This would, guarantee the free expression of the will of the communities. There must be mechanism for ensure that the voice of the disadvantaged groups such as persons with disabilities and women are brought to the discursive spaces. During the operation of the committees, judicial guidance ought to be taken into account to advance the principle of participatory democracy as a long-standing, deeply entrenched and constantly evolving tenet of society which has been subsumed in the constitutional order.
Community participation through health unit management committees has the potential to thrive through the democratic value of devolution of power, which has been constitutionally rolled out as decentralisation in Uganda. This should however be anchored in the understanding that decentralisation moves beyond the form underscored under the legal and institutional frameworks. There must be practical commitment to democratic values where the political leadership supports effective empowerment of communities to enable them effectively to participate in decision-making processes that affect their health. There should be efforts towards reducing central government control which would wrench powers from the community. The committees should be able to map and deal with the power dynamics including de ning resource distribution mechanisms that could include external factors and actors. The working of the committees should provide opportunity for conducting a stakeholder analysis and a knowledge and capacity assessment in the communities as part of the decentralization system.
There is clearly a challenge with politically appointed community representatives as opposed to those popularly elected by the communities. Even then, it was clear in the ndings that for community participation to happen meaningfully through the health unit management committees there needs to be political will and buy-in by different duty bearers at the local government level. The political leaders have, through the decentralized system of governance, been assigned key roles relating to selection of the committees and can also be engaged in advocacy to address key challenges within their sub-counties. The political leaders can contribute towards effective participation through creating an enabling environment, including support, nancial resources and training.
In undertaking their functions as health unit management committees there are a number of challenges that affect both their functions and ensuring that they act as a mechanism for community participation in health decision-making at the health facilities. For instance, through the devolution of power, the architecture of the health unit management committees fused the roles of health unit management committees with those of local government political leadership. Since the political leaders choose the HUMCs' members, and some of the politicians end up on these committees a con ict of interest is immediately created. This leads to two major challenges, rst the health unit management committees lose their identity of being community participation mechanisms, but rather become another political space. Secondly the health unit management committees become accountable to the political leadership appointing it and not community. The best practice would be that the Health unit management committee members are directly elected by the communities to ensure that accountability and reporting to the community is not lost.

Conclusion
It can be concluded from this paper that the normative framework of deliberative democracy is crucial in setting both structural and procedural parameters for community participation through the health unit management committees as vehicles for Democratic Governance in Health Systems. This framework should be based on both Habermas' perspectives on deliberative democracy, and also on the evidence from facts and experiences as drawn from health Nyamiringa and Kikoolimbo health unit management committees. The framework should, however, not be presented as creating, limiting or undermining the standards of community participation under international human rights law obligations, principles of democracy and/or principles of health governance. A normative framework grounded in theories of deliberative democracy, human rights and health governance is critical in clarifying the architecture of community participation in decentralised health systems through health committees.
Community participation in the health sector in Uganda hinges on health unit management committees at the lower service provision points. These HUMCs are also perceived as vehicles to strengthen health governance through realizing the right to health of the communities. However, these have been established without attention to investing in capacity building needed to enable them to exercise community voice in the health system. This undermines their role in providing oversight and accountability in the health system. This paper emphasizes the need to strike a balance between the competitive notions of democracy and the public health requirements of inclusive and direct participation of communities through a human rights approach. This can be achieved through numerous ways including formulation of legal and/or policy frameworks that are based on empirical evidence of practice in a given country context.
While the intention of this paper is not to make a point that health unit management committees are a magic bullet for effective community participation in health system, the paper emphasises the importance of building health unit management committees as structures of citizen participation in health systems grounded in the principles of deliberative democracy and human rights.

Declarations
Ethical Approval and Consent to participate Ethical clearance was obtained from the National Council for Science and Technology under permit number SS 2731.