Use of evidence in decision making is viewed as a crucial step in attainment of good health outcomes. Evidence on the most cost effective interventions exists, but its use in decision making is still suboptimal. Uptake of evidence in public health policy development, also referred to as knowledge translation (KT), is poorly understood, especially in low income countries (LIC). Although a lot of work has been done on this subject, its applicability in LIC is limited for a number of reasons. My PhD study attempted to address this gap by constructing plausible explanations rooted in contextual realities on what it takes to get evidence into policy. These were detailed in a Middle Range Theory (MRT) on KT in Uganda but can also serve as a reference for other low- and middle countries.
Generating and getting evidence into policy requires time and investment. Learning from work undertaken in Uganda and other published literature, it is clear that there is no simple formula for improving evidence uptake, but a better understanding of facilitating factors and interactions of stakeholders can help researchers and policy actors improve KT. The focus should be on maximizing the facilitating factors on the one hand while mitigating barriers on the other.
In many LIC countries the rapidly changing political context limits the long term horizon of decision makers because they are not certain how long they will stay in office. The need for votes to return to office influences decision making, leaning more towards quick fixes with popular support, as opposed to a strategic long term and sustainable approach. Kingdon argues that a decision is taken when the political, policy and problem streams meet. The political window will provide different opportunities for KT based on the political implications of the decision. In one particular case for Uganda, for example, evidence on the disadvantages of user fees for health was available and had been discussed already without making a decision. Presidential elections in 2001 provided an opportunity. So this case is a clear example of Kingdon’s problem, political and policy streams coming together to form a window of opportunity. In the problem stream, there was a lot of agitation to try to get policy makers to focus on the problem of user fees. The agitation came from the community in the form of complaints. These complaints were presented to politicians, including those campaigning for the presidency at the time. In the political stream, the issue of user fees for health care was included in the election manifesto of the two presidential candidates. Several proposals were floating in the policy stream, like the proposal of civil society organizations to abolish user fees as a means to improve access to health services by the poor. Presidential elections provided thus an opportunity for the three streams to meet and a window presented itself for a policy decision on user fee removal.
When we talk of evidence, we need to be mindful of the fact that there are different types of evidence and that different actors attach different importance to each. There is no consensus as to which type of evidence is deemed most important, and the issue under consideration will influence the nature of evidence that drives decision making. Looking at biomedical policies, like the case of malaria treatment policy change in Uganda in 2006, as compared to a social policy concerning people’s welfare, like the abolition of user fees, we saw very clearly that actors attach different emphasis to different types of evidence in the two cases. In the case of malaria treatment policy change, actors put a lot of emphasis on evidence from efficacy studies while in the case of user fee abolition, politicians listened very carefully to community complaints.
This withstanding, evidence must be of good quality, comprehensive enough to inform different aspects of decision making, and provided in a timely manner. Who generated the evidence is another topical issue as researchers must be viewed as credible and independent. Where this is lacking, or seems to be lacking, however good the evidence is, if there are suspicions regarding the researchers involved, evidence will be seen in the same light, presenting a missed opportunity for KT. The abolition of user fees in Uganda is a case in point where evidence generated by the World Bank was judged as `misleading the policy makers`. This stemmed from the fact that the World Bank was supporting retention of user fees of which the introduction had been supported by them as a way to guarantee loan sustainability. The evidence they generated at the time was in line with their prior position.
Inadequate attention has been paid to ensuring effective dissemination. How do we reach decision makers and influential stakeholders? Targeted dissemination and face to face interactions are important but the easier thing to do is sharing a research report and/or holding a dissemination workshop whose impact tends to be limited. Influential stakeholders, who can be mobilized through targeted engagement, are often not fully exploited, for example the politicians, village chiefs and the media. In the case of the malaria treatment policy change in Uganda, effective and targeted dissemination was one of the facilitating factors for KT.
Evidence generation and its uptake in policy development occur amidst stakeholder interaction and this takes us to another important point: the need for KT partnerships. Ssengooba et al noted that shared platforms for learning and decision making among stakeholders enhanced the uptake of evidence from prevention of mother to child transmission (PMTCT) studies into policy. Similarly, in our study looking at the malaria treatment policy change in Uganda, partnerships bringing together researchers, policy makers, civil society, donors and implementers provided an opportunity to discuss all available evidence and make policy decisions in line with the evidence.
Partnerships need to encompass the whole process right from evidence synthesis up to application. Partnerships must be inclusive; there must be mutual trust and sharing of evidence coupled with open discussion. This provides an opportunity of discussing the different types of evidence in order to come to the most appropriate decision given other contextual issues. In addition, conflicts of interest can be openly handled within such partnerships.
Who owns the KT process in countries? The Paris Declaration and the Accra Agenda for Action emphasize Government ownership of the health development agenda. Hence, the strengthened capacity of the Ministry of Health to lead the KT process is essential. In the case of the malaria treatment policy change in Uganda, the ministry of health, whose capacity had been strengthened by the East African Network on Monitoring Antimalarial Treatment (EANMAT), fully owned the KT process, convened partners and chaired all working groups which enhanced KT.
Contextual issues also merit attention and these include systemic factors encompassing political, economic, and social dimensions which may be at the national or international level. Research recommendations which are contrary to cultural and social norms are unlikely to influence policy. Medical male circumcision in Uganda is a case in point: despite the availability of high quality evidence, influencing policy was not an easy path. Lack of appreciation of the benefits by the intended beneficiaries, and politicians openly contesting the evidence were among the challenges.
This raises the issue of local research capacity whereby local researchers, who know the context well and interact with decision makers more often, may be more influential in KT than international researchers. Given the weak research capacity in low income countries, partnerships in research need to be strengthened through North to South collaborations but also through South to South collaboration by building regional capacity since contextual issues within a given region may not differ significantly. In this regard, regional professional bodies play a role and indeed have been instrumental in some cases like the case of change of malaria treatment policies in East African countries.
To conclude, we noted that KT is a process that has to be institutionalised, planned for, properly funded, requires capacity building, but which also needs to be tailored to the nature of the policy issue. Evidence concerning non-contested technical issues tends to be taken up more easily than evidence pertaining to more controversial policy issues. There is a need to map the actors involved and ensure targeted dissemination noting that different actors define evidence in different ways, hence the need to take this diversity into consideration in the dissemination. Partnerships are important in that they provide an opportunity for actors to come together, review the different pieces of evidence and come to the most appropriate decision in a given context. These however need to be built on mutual trust and must have mechanisms for managing conflicts of interest. Stakeholder involvement from the start, already at the evidence synthesis stage, is vital; it takes more effort and time if the issue is contested (low agreement). However, one needs to pay attention to the nature of the issue under consideration as this will influence stakeholder interactions and KT processes, whether the policy issue is contested or not.