Every interviewee expressed a desire to be consulted and to use their knowledge to help shape central government health policy-making, ‘as an organisation that also deals with health issues, there ‘should have been some form of involvement for us to input into the documents’ (P11, CSO). Of the 19 interviewees, 11 reported having some involvement in the development or validation of the HSSP II, and only one in the development of the NHP II. Three major themes, and five subthemes, emerged from the data on stakeholder engagement: power (tokenistic involvement; stakeholder hierarchy; mutual distrust), representation (preferred stakeholders; open versus invited engagement); and stakeholder-driven engagement.
Power
The stakeholders felt that being involved in policy decision-making would communicate the policy needs of their members and the communities they serve and help hold those responsible for policy implementation to account:
‘[Name of organisation] is the mouthpiece of what happens on the ground regarding HIV and AIDS in Malawi. They are the organisation that can tell the nation and stakeholders what happens in Malawian communities […] There is need for CSOs to position themselves as key stakeholders in mainstream health so that they are included in such policy and strategy [HSSP II and NHP II] processes. By getting involved at such policy or strategy level, there is an opportunity for the organisations to influence donor priorities, demand services or accountability’ (P10, CSO).
Additionally, the stakeholders said that they can provide an understanding of the social context to the scientific discussions that they viewed as often dominating the policy development process:
‘the focus of the MoH was too biomedical, and often overlooked the social or political elements of health. By embedding the engagement of CSOs and NGOs throughout the policy development process, the wider determinants of health would be considered more routinely as these organisations often prioritise addressing social issues in their own work’ (P10, CSO).
Yet, most of the interviewees described feeling powerless to influence the content of health policy in Malawi, because they were either not consulted at all, not engaged throughout the policy life cycle (i.e. they were only consulted at the implementation not the development stages), insufficient time was allocated for stakeholder consultation, or their contributions were disregarded (P17, P11, P12, P10, P7, P15, P19). There was a perceived hierarchy of stakeholder engagement, with the citizens at the bottom having no influence and donors at the top with the greatest leverage over the government because of their substantial financial contribution to healthcare in Malawi. The interviewees described some methods they had used to help foster a more widespread culture of stakeholder engagement, starting with their own organisations’ structures, activities, and partnerships.
Tokenistic involvement
Multiple interviewees described their involvement in the government’s policy and budget-making processes as tokenistic (P9, P17, P12). They recounted how they were invited to a consultation event but felt that the policy had either already been substantively developed or that they received the documents for review too late to provide comprehensive feedback. As a result, they felt undervalued in the process and that their voices were not heard, ‘the documents [NHP II and HSSP II] do not reflect the voices of patients’ (P3, CSO).
A CSO interviewee recalled attending two meetings (a consultation and a validation meeting) aimed at discussing the draft strategy. They received the working documents close to the meeting date so there was little time to review them with other network members and prepare a detailed response. They reported giving some feedback during the meetings, but most was ignored. They made a complaint to the MoH, but nothing was done. The members of this network organisation ‘felt that they were consulted as an after-thought just as a requirement to tick the box that the process involved consultation and endorsement of disability organisation or constituencies’ (P17, CSO). Similarly, representatives of Christian Health Association of Malawi (CHAM) facilities were invited to the HSSP II launch event but were not consulted at all during the policy development process. Thus, their involvement was purely to be seen to endorse the policy (P4, government-funded institution).
Three interviews reflected on their experience of the government’s budget consultation meetings (a public consultation event held annually by the Minister of Finance). The events ‘are cosmetic and aimed to coerce stakeholders to rubberstamping an already made budget. Nothing changes in the budget despite people submitting views during the consultations’ (P12, CSO). One NGO expressed the view that ‘meaningful participation is difficult because the Ministry of Finance only announces the budget when they are already at an advanced state of the budget process. At this point it is nearly impossible to make any significant or substantial changes’ (P9, NGO). Another NGO was frustrated by the late point at which they were consulted:
‘nothing can substantially change at this stage of the budget process. [Name of organisation] has not been involved in the budget processes in the past, but we have learnt that if there will be any of our involvement in future, we will need to engage much earlier in the process’ (P9, NGO).
Local government representatives and healthcare providers were equally frustrated with the policy development process (P4, P18, P15, P19). A member of a DHO stated that ‘maybe we were only consulted 20% [of the time] and by the time we were consulted, the process was already at an advanced stage and there was little room to make changes on the issues and content' (P18, local government). They added that:
‘MoH consultations are more of a window dressing to show the nation that they have consulted with different stakeholders before adopting a national document, but there isn’t much room for external stakeholders (outside the ministry) to influence the process. It’s all about economics […] for me, it’s a waste of time’ (P18, local government).
A local government representative noted that they were only consulted by the government during national health crises which required action at the local level (P19, local government).
There was the perception that Malawi is lagging behind other sub-Saharan African countries in the degree of stakeholder engagement. For example:
‘the situation in Malawi in terms of involvement of persons with disability in policy or budget is different from the situation in Uganda and Kenya, where persons with disability are fully engaged as a priority (not as an after-thought like the case in Malawi). Their input is given attention before finalising the budget’ (P17, CSO).
Even when an NGO took what they felt to be important evidence related to policy to the MoH, there was a lack of interest in what they had to say, ‘[name of organisation] has done some studies on access to health, but their recommendations have not been implemented by government. The recommendations have not been treated with the urgency and importance which it deserves’ (P9, NGO).
Then, where the policy documents did reflect an organisation’s priorities (the affordability, availability and accessibility of health services), there was insufficient detail for the interviewees to be confident that there would be a health improvement (P6, P17). For example, where the specific issue of HIV/AIDS management is mentioned in the NHP II and HSSP II, ‘it is only in passing and without much detail as to how the strategy or plan will manage HIV and AIDS and coordination around it’ (P6, government-funded institution). Likewise, ‘the documents make some reference to disability, but not enough’ (P17, CSO).
Stakeholder hierarchy
The interviewees’ experiences suggest that not all stakeholders were equal in the exercise of stakeholder engagement. They found a clear hierarchy determined by stakeholder group and knowledge, with donors being the most influential in government policy-making and CSOs, with little power attempting indirect influence via donors.
As donors provide a large proportion of the funding for healthcare in Malawi, they are perceived as having a lot of leverage in policy decision-making, and specifically in driving the development of the NHP II and HSSP II. They were also identified as valuing stakeholder engagement to a greater extent than the government and made concerted efforts to identify stakeholders and engage them in both the identification of health needs and healthcare implementation (P12, CSO):
‘The Health Sector Strategic Plan and the National Health Policy, they were donor driven. They received funding from donors and the donors were influential in the process. They (donors) were active during the consultation and validation meetings, the processes were funded by the donors. The funder influenced the agenda. Follow the money, the money won’t lie’ (P5, NGO).
Thus, the CSOs viewed engagement with donors both as a way to influence their initiatives to meet the needs of the communities they represent and to indirectly influence centralised policy-making. For example, one CSO (P12) is involved in a project with Oxfam because the donor identified them via a stakeholder mapping exercise and invited them to collaborate in further mapping exercises to identify the main needs of citizens and healthcare facilities. The reports from these exercises will be used to evaluate the project’s success and inform the donor’s advocacy agenda (P12).
Having links to donors could limit the opportunities for local level stakeholders to influence policy (although sometimes their desired outcome could still be achieved if a donor independently advocated for it). For example, a CSO had been advocating for HIV self-testing due to the low usage of HIV testing centres, but ‘there was no progress on the government side until PEPFAR took the matter up by setting up future-funding conditions. They said future funding was on condition of progress on self-testing’ (P10, CSO). Another CSO had not been invited to any consultations on the health budget. As they were solely funded by PEPFAR, they believed that the government did not consider them eligible for a say in national budgetary policy-making, ‘[we] are not invited to budget consultations and have not participated in budget reviews. This is mainly because our funding primarily comes from donor agencies such as PEPFAR. We do not receive any funding from government or the national budget’ (P13, CSO). A local government interviewee voiced the same opinion, ‘because they are not targeted for any portion of the national budget, they are not involved in any processes around its development or allocation towards health’ (P16, local government).
A hierarchy of engagement was also observed in government advisory committees. A CSO interviewee reported being part of the technical working groups for the National AIDS Commission in the development of national HIV policy and the HIV and AIDS Act. On the HIV issues, they felt that their ‘issues were taken aboard’ (P13, CSO). However, their influence was still limited as management of the national HIV/AIDs response, and thus most of the decision-making power and influence, remains with the government and donors, who ‘meet regularly to discuss progress in implementation of the HIV and AIDS response, funding mechanisms and alignment with the national policies and strategies’ (P6, government-funded institution).
Despite the considerable leverage of donors over the government, an NGO noted that in health policy development the power to determine the content ultimately remains with the MoH, ‘the Ministry of Health is too bureaucratic and only operates through its technical working groups and related structures. The Principal Secretary sometimes heads the technical working groups, thus maintaining authority and upper-hand in the processes’ (P9, NGO). The government was also perceived as constraining influence from local level stakeholders by holding most of the policy development and consultation events in the MoH buildings in Lilongwe:
‘involvement during the processes for developing the NHP II and HSSP II mostly happened at the MoH headquarters level. There were times when the District Health Office would be involved. Teams from the Ministry headquarters would go to the districts with a questionnaire to ask questions related to policy and strategy. Consultations to finalise the two documents mostly happened at the Ministry headquarters’ (P15, local government).
Hierarchical dynamics were also observed in health decision-making at the district level. Observing the greater value placed on the contributions of those with professional expertise to health decision-making diminished the confidence of CSOs in the processes of engagement:
‘although [name of organisation] members are found all over the country, the secretariat only engages with district authorities minimally due to limited capacity and lack of authority and leverage to influence the District Health Officer who are viewed (by district-level stakeholders) as more senior than the members of [name of organisation]. Most community-based organisation’s leaders do not have the same level of professional expertise as the district health officers (who are doctors) and as such, they feel inferior when engaging with them’ (P10, CSO).
It is worth noting that whilst local level stakeholders want to influence central policy, they are not always open to top-down influence over their own activities. For example, a local government representative reported that:
‘some of the stakeholders are cooperative and listen to directions and advice from the District Health Office in terms of overall district priorities, plans and strategies, but some do not and do their own activities based on their own preferences and priorities […] Most of them [CSOs] do not follow the essential health package and district plans’ (P18, local government).
Likewise, donors exert influence on the government at the central level, including in the extent of stakeholder engagement, but the government may not have much influence on donors at the district policy or implementation level, ‘we are able to collaborate and coordinate with donors at the central level. But at the district level, they are doing their own thing’ (P6, government-funded institution).
Mutual distrust
There was evidence of mutual distrust between stakeholders which may influence their willingness to engage and be engaged in consultation processes and their effectiveness in them.
The development of the NHP II and HSSP II were considered ‘a MoH thing. It’s also very political’ (P20), with the documents developed to appeal to donors rather than to affect and enforce the changes needed to improve the health and wellbeing of the people of Malawi (P5, P6, P20). The lack of trust between the government and stakeholders may have inhibited meaningful engagement, ‘there is lack of trust between CSOs and government and this affects collaboration’ (P14, CSO).
Concerns were raised about the motives of politicians in policy-making, ‘government decisions suffer political interference from politicians who push government’s decisions towards their interests and preferences’ (P14, CSO). Suspicion of intentions as inhibiting meaningful engagement was also felt to impact government perceptions of CSOs, in particular, ‘government officials are less able to collaborate with civil society because they are always suspicious of recommendations of CSOs that they aim to undermine government’ (P14, CSO). The stakeholders thought perhaps the MoH did not involve NGO/CSOs in policy development more because they are perceived as trying to push their own agendas, ‘collaboration among civil society is challenging because stakeholders also want to get a share of the budget of any initiative’ (P12, CSO). This concern may be warranted as, for example, one NGO reported following the policy development and budgeting processes closely to ensure their own project was included in the HSSP II and received funding (P5, NGO).
Representation
There were concerns about appropriate representation of different groups in health policy, with issues of government-preferred stakeholders and invited versus open consultation.
Preferred stakeholders
The government was perceived as having a very small group of preferred stakeholders who they engaged repeatedly in the policy development process. This practice benefitted CHAM, the largest non-government provider of healthcare in Malawi, with whom the government routinely consulted on health policies and who would be directly affected by them:
‘when the government is developing policy or strategy documents, they invite CHAM secretariat to the consultation of validation meetings. Prior to attending government consultations, CHAM may call for meetings of its membership for their input e.g. CHAM has previously convened its members to discuss Human Resources for Health policy to seek their input. This is because issues of human resources have been very topical among CHAM members. Many CHAM members were invited to this discussion at a consultative meeting probably because it involved welfare of staff at CHAM member facilities.’ (P4, government-funded institution).
However, the practice was seen as excluding other stakeholders, such as those outside health, ‘there is some narrow thinking that health is cross cutting only with nutrition, agriculture and the environment. The health sector only restricts engagement to those working in the health sector when dealing with health matters’ (P14, CSO). The government was also perceived as having a preferred CSO partner, the Malawian Health Equity Network (MHEN), which has:
‘regular interface with government for consultations through biannual and adhoc meetings. They have a good rapport with government. In addition, MHEN Director holds many ongoing confidential bilateral discussions and engagement with government officials to discuss issues as they arise and offer views or advice before decisions are taken’ (P1, CSO).
There was acknowledgement of the suitability (and ease) of using MHEN as the ‘go-to’ CSO for stakeholder engagement: ‘in terms of civil society alliances, MHEN is the most active in convening civil society in the area of health’ (P12, CSO). There was support for its inclusion in a range of MoH committees and, therefore, its ability to influence policy-making across the health sector:
‘MHEN is a member of the government committees on health, including the Health Financing Committee, Community Health Technical Working Group, Human Resources for Health Technical Working Group. Through these ongoing engagement and participation in working groups, MHEN is able to represent its network and bring the voice of the people to the national decision-making processes related to health’ (P1, CSO).
Yet, there was a perception that its involvement meant the exclusion of other CSOs, and that it could not be considered representative of all CSOs:
‘The challenge with Ministry of Health’s engagement with NGO stakeholders is that they assume that MHEN is the representative of all health NGOs, but not all NGOs doing work in the area of health are members of MHEN. The organisation’s view is that MHEN can’t replace grassroots voices in the engagement with the MoH. MHEN does not have capacity to represent all voices, simply impossible’ (P3, CSO).
Further, MHEN was criticised for not adequately representing all the CSOs in its own network (P3, CSO). Instead of all or a representative group of MHEN network members attending policy development activities to represent the CSO and citizens’ voice, the secretariat would attend, ‘the secretariat represents the voices of its member organisations, and through that process they speak on behalf of the people in Malawi’ (P1, CSO). The use of an individual to represent the views of a diverse group of CSOs was viewed as problematic by some organisations (P10 and P3). An alternative approach, which was proposed as a way to increase the CSO influence and representativeness, was for a member of each CSO in a network CSO to attend the consultation event when they wanted to influence government policy:
‘as a network, not participating in the budget processes is a missed opportunity because it is not only the secretariat that must attend, but the members. If members attended, for example, it could mean more voices on HIV on the budget process. The network has potential to influence the health budget using its leverage of having too many stakeholders in its network. If members attended, there would be a [name of organisation] member at every platform available’ (P10, CSO).
For general health policy stakeholder engagement, there may be limitations with using a single organisation, albeit one that operated as a network, to represent diverse NGO and CSO stakeholders. The practice may work better for policies in more specific disease areas. For example, one CSO was used to represent all CSOs with an interest in HIV/AIDs in the development of the HIV and AIDs ACT and the National Strategic Plan for HIV. The role of representation also appears to have been more formalised than in the development of the NHP II and HSSP II:
‘the role included coordinating civil society input during consultation, review and implementation processes. They sought input from civil society and represented such voices during consultations. They were more involved in HIV-specific processes because it is where their expertise and relevance fit best’ (P10, CSO).
Open versus invited engagement
The issue of preferred representation extends to the level of public consultation, where the government typically holds consultation events with open and invited participation (i.e. the act of including named stakeholders in the open invitation). The government issued an open invitation to public consultation events on the HSSP II and issues them annually for the annual health budget. For example:
‘the budget publicises the budget through a media advert in which they call for interested parties to attend budget consultative meetings held in all districts across the country, and in the advert they mention the stakeholders who they want to attend the consultations. However, the budget consultations are open to anyone who is able to attend’ (P9, NGO).
However, by specifying stakeholders who they would like to attend, unnamed stakeholders may feel unwelcome or undervalued in the consultation process, ‘their argument is that it’s an open invitation they put in newspapers so that anyone can attend. I wouldn’t say I went. I attended once but I wasn’t like, we were invited’ (P2, NGO). This NGO representative thought that by holding open invitation events the government could avoid criticism of the representativeness of the stakeholders involved as they would, therefore, be self-selecting. However, to ensure appropriate representation, another interviewee felt strongly that specific stakeholders should be invited to get involved and involved more comprehensively in the policy development process. They felt this was the only way to ensure inclusion of the voices of patients in the HSSP II as public consultation exercises made a show of stakeholder engagement but did not influence the substance of the policy:
‘The organisation was not involved in any of the processes. They were not invited to any of the consultation or validation conferences. As a result, they even wrote a letter to the Ministry of Health protesting the lack of representation of patients’ groups and voices in the processes leading to the adoption of the two documents’ (P3, CSO).
Furthermore, whilst open events mean that anyone can attend, they do not ensure representation of an appropriately wide range of stakeholders. In particular people from rural communities and vulnerable groups are likely to remain excluded (P3, P12, P14). To elicit a response from a greater number and more varied NGO and CSO stakeholders, using the example of the budget, a CSO recommended that the government also ‘circulate a tool with leading questions to seek views on what stakeholders want to see for a budget to be a citizens’ budget’ (P12, CSO). This could overcome any issues of CSOs not having the resources to attend consultation events.
Conversely, issues may also arise with invitation only events. In one example, the interviewee’s organisation was not represented in a HSSP II consultation meeting, which they attributed to confusion over the composition of the devolved local government:
‘we were not fully involved perhaps because they assumed [location of local government institution] was represented in the processes. They assumed that we fall under the District Health Office and thought that by including the district health office, we were represented’ (P16, local government).
Stakeholder-driven engagement
The interviewees believed there to be an absence of an established culture of stakeholder engagement in Malawi, acknowledging that the benefits to stakeholder engagement in policy-making could only be realised via significant changes in thinking and practices by both government and stakeholders: ‘the challenge with Malawian civil society is that they are only reactive in their advocacy. They only react to problems late when a crisis has already occurred, not actively identifying issues and then doing something about it’ (P12, CSO). Examples were given of how individual stakeholders have attempted to overcome perceived barriers to stakeholder engagement to try and influence the government and increase the representation of local level organisations in government health decision-making.
All the interviewees voiced frustration with poor communication from the government around health policy. For example, ‘the major challenge is that government is too bureaucratic and that affects how government is able to communicate with stakeholders and how it provides access to government information. Currently it is difficult’ (P14, CSO). One organisation had addressed the dual issues of poor government-driven communication and not feeling engaged in government decision-making by placing members of the MoH on their boards rather than waiting for the MoH to consult them. This provided an opportunity for the MoH to hear about the challenges faced by these facilities and their needs and the needs of the communities to which they provide healthcare (P4, government-funded institution). This organisation has also introduced stakeholder engagement measures into their own decision-making processes ‘to enhance meaningful engagement with its members so that they participate more in decision-making’ (P4, government-funded institution). They now have:
‘several committees through which the secretariat engages with the CHAM members’ facilities. Under that arrangement, CHAM facilities are able to interact between each other, and also get involved in CHAM’s decision-making bodies through input into the committee resolutions which feed into CHAM’s decisions. CHAM also hold regional meetings where it meets and engages with its members at the regional level for information sharing and consultations’ (P4, government-funded institution).
Despite concerns over the representativeness of MHEN, the network organisation was reported as encouraging other stakeholders (who were not members of MHEN) to attend consultation events (P20 and P3). There were further examples of CSOs encouraging and modelling stakeholder engagement, such as:
‘[name of organisation] has held meetings to develop standards of care for victims and the last meeting that they had was held in [location] to come up with minimum standards of care for victims of gender-based violence and roles of stakeholders in areas such as police support and healthcare’ (P9, NGO).
These practices were also demonstrated and funded by donors. For example, one CSO referred to a two year project funded by Oxfam which aimed to increase the responsiveness of primary healthcare providers to citizens by performing stakeholder analyses to identify health service delivery issues that are of concern to the community. They will then approach the district health office and health service providers ‘to seek corrective action for improvement of services’ (P12, CSO).
Another way of increasing stakeholder engagement was through raising policy literacy, as a lack of understanding about the purpose and process of health policy and budgets was considered a major barrier to influencing health decision-making. It was hoped that building knowledge would lead to advocacy:
‘Although persons with disability have strong voices; they have little knowledge about health and advocacy processes which plays a role in the lack of involvement in health matters. They simply have average knowledge about the relationship between disability and health policy, budgets and broader issues. As such, their advocacy is weak and could be strengthened’ (P17, CSO).
Other CSOs noted that although they were encouraging citizens and CSOs to advocate for a voice in healthcare decision-making, the government either did not provide the opportunities for engagement or for meaningful engagement (P3 and P12). This is despite the development of a patients’ welfare charter. As one interviewee ruefully stated, ‘the idea of a patients’ welfare association is to give people their rights to be involved in decisions about their health. The Ministry of Health previously developed a patients’ welfare charter, but it was never implemented’ (P3, CSO). It was hoped that the implementation of the charter might boost citizen engagement in health advocacy, which is especially important in a distributed healthcare system where the needs of citizens in rural villages vary from those in urban settings, ‘the importance of protecting patients’ rights is even more critical in the villages where patients view public health services as a favour from government, as opposed to viewing such services as obligations enshrined in the constitution’ (P3, CSO). Accordingly, citizens continue to feel at the bottom of the hierarchy of influence and therefore voiceless in health policy decision-making.