Increasing Use of Research Findings in Improving Evidence-Based Health Policy at the National Level

In February 2016, the Minister of Health decided to increase the use of research findings in improving the quality of the national health policy and planning. The Ministry of Health has instructed the National Institute of Health Research and Development or NIHRD to play a stronger role of monitoring and evaluating all health programs, because “their opinion and research findings should be the basis for changes in national health policies and planning”. Compared to the past, the Ministry of Health has increased the research budget for evidence-based research tremendously. However, there is a gap between the information needs of program and policy-makers and the information offered by researchers. A close dialogue is needed between the users (program managers, policy makers and planners) and the suppliers (researchers and evaluators) to ensure that the evidence-based supplied by research is useful for programs, planning and health policy.


Introduction
Indonesian Minister of Health has positioned National Institute of Health Research and Development, as the "locomotive" in leading and guarding the implementation of the national health program and policies. The institute is expected to generate the evidence-based research that is relevant, reliable, and with high quality knowledge and information for the stakeholders and users (program managers and health policy planners). 1 In addition, the institute should provide accurate explanations for decision makers and program planners to facilitate rapid policy changes, covering a wider health related and re-emerging health development issues. To achieve this, in 2015 2 This article discusses whether this challenging task is deliverable by the institute and whether evidence-based information is needed by the program managers to improve services.

Previous Approaches and Results
The National Institute of Health Research and Development thought that they had implemented evidence-based research when they used Client Oriented Research Activity (CORA) in implementing two types of national level surveys the 2007, 2010, and 2013 National Basic Health Research and the 2011 Health Facilities Research. 3 CORA was considered as the proper approach to implement evidence-based health research because it encompassed multi-sectoral cooperation from the beginning (i.e., involvement of many directorate generals at the Ministry of Health, professional organizations, health experts, Indonesian public health experts, Indonesian doctor association, dentist association, midwife association, nurse association, provincial and district hospitals, Indonesia hospital association, Indonesia vertical hospital association, and many other organizations and universities). CORA was designed to accommodate users (decision makers and policy makers) at national, provincial, and district levels, also cover most of public health needs, yield applicable policy options, develop new medicines, vaccines, and overcome the barriers in communicating results to the wider community. 3

External Review in 2017
In mid-October 2017, a review of more than 12 proposals submitted by National Institute of Health Research and Development researchers for 2018 funding was carried out by an external reviewer. The reviewer concluded that almost all proposals were not directed to support a particular program measure(s). When asked, the answers revealed that none of the principal investigators (researcher) held discussions with the Ministry of Health program managers or relevant policy makers during proposal writing. Their research objectives were not formulated based on any program-based problems discussed with any program managers. In other words, there was no real collaboration between program and research staff at the Indonesian Ministry of Health. Thus it is no surprise that, although the National Institute of Health Research and Development produced 1,319 studies between 2011 and 2015, very few findings were used by program managers, according to its public relation staff. Most research reports were just kept on the library shelves. 1

Trainings from the Knowledge Sector Initiative
Between 2013 and 2017, under joint funding from Indonesian and Australian Governments, the Knowledge Sector Initiative (KSI) project managed core grants for organizational development, quality research, and research communication. The focus was the attainment of 2015-2019 Development Plan (Rencana Strategi) objectives. The KSI provided technical assistance on improving research capacity including activities known as knowledge demand and use, knowledge intermediation (communicating evidence-informed-advocacy and reporting), and knowledge sector to lessen systemic barriers and foster private sector participation in research. At national level, KSI support is directed to accelerate the mobilization and systematization of data. In research communication, the National Institute of Health Research and Development researchers received trainings in development of policy briefs, support for peer review of journal articles and research proposal writing, and the use of multimedia. By 2015, the institute has successfully produced 24 policy briefs covering selected research results, complete with public-friendly infographics and illustrations. 1

Literature on Evidence-Based Research that Includes both Policy and Practice
Black, 4 draws a distinction between providing an evidence-based for policy and for practice. Evidencebased policy refers to providing evidence for policies which by the nature are more complex, and takes place over a longer time. Evidence-base for practice refers to providing evidence for specific health practices which are more sharply defined, and usually takes place over shorter period of time. Black points out that, for these reasons, the evidence-base for practice usually has been more successful in influencing specific practices compared to the evidence-base for policy. This is because policies are often decided on the basis of influences from other sectors (e.g. education, finance, etc) and the broader political environment. 4 Other than in the National Institute of Health Research and Development, there are many health researchers in the academic community. However, academic community researchers do not show a great interest towards the achievements of particular public health policies and programs. In general, academic community researchers take an independently stance and seem to have only limited engagement in the policy debates. Consequently, they also are not heavily involved in providing evidence for either policy or practice. This is due at least in part, to the rules of their institutions that research is only for providing a new way of conceptualizing previous or existing theoretical frameworks, mapping the decision-making landscape in a local or regional area, or challenging some public health conventional assumptions.
Below, results of a quick review of selected publications in the Kesmas: National Public Health Journal (2013-2017) show how research topics seem to be chosen independent of any Ministry of Health program targets or goals. A group of researchers looked at education, age at the last childbirth, ideal family size, modern versus traditional contraceptive method, survival of preceding birth, and infant mortality records, in relation to birth interval. 5 Another group of researchers found out that constructing healthy rooms in the house could protect children from tuberculosis, even when they were exposed to adult with tuberculosis who live in the same house for a long period, made it no longer a risk factor. 6 Evaluation of breastfeeding counseling implementation in Jambi, Sumatra, found that there was no national guideline on proper breastfeeding counseling in 2014, so the researcher used WHO guidelines. 7 Another group expressed that the central government should allow more authorities to local governments' innovative policy that are felt more effective in reducing maternal and newborn deaths. 8 A framework showing relations between severe pre-eclampsia syndrome and maternal death, was tested using a case-control study design. 9 The implication of this is that evidence from research findings must be relevant to the problems faced by policy makers if they are to use that for changing policy. On the other hand, researchers are to be mindful that research has little direct influence on service and governance policy currently and in the near future in Indonesia. The relation between research and policy depends on the policy makers. Right now, research evidence is more influential in central policy than local policy, where the use of research depends on the degree of con-sensus on the policy goal. It is used if it supports the consensus and is used selectively if there is a lack of consensus. Most researchers are politically naive. They have a poor understanding of how policy is made and have unrealistic expectations about what research can achieve.

Conclusion
It is a considerable challenge to change the researchers' attitudes, including within the Ministry of Health's National Institute of Health Research and Development. The researchers may not change because they know that they will be disappointed when program managers and policy makers ignore their findings. They need to acquire a more sophisticated understanding of the policy process, that sometimes sensible decisions may not reflect scientific rationality, and that political context is important too, particularly with policies related to services and governance.
On the other hand, policy makers need to be more involved in the conceptualization and conduct of research. Researchers need greater access to information on the priorities of program managers and policy makers, who in turn need to organize and communicate their needs better. A closer relation between the two groups needs to be sustained during the research and beyond if the work is to have any impact on planning and policy in the future.
Researchers at National Institute of Health Research and Development have been enriched in three aspects: additional research funding, reorganization of the four units, and in the provision of technical and knowledge building. These are to facilitate new partnerships between researchers and program managers/policy makers. However, the researchers at the institute will continue to fail in supplying evidence-based information if they do not have the knowledge about program success measures or targets in their proposed research objectives.
The current situation between researcher and program managers/policy makers can be concluded as first, both sides (research and program staff) are not comfortable to work together because they do not understand how to apply the concept of 'evidence-based research'. Second, the critical issue is to ensure research related to practice is well formulated and carried out so that it has higher probability to be implemented. Operations research provides a way forward for achieving this. Third, operations research is distinguished by its focus on programs and their improvement, and will inevitably direct researchers' attention to the related program manager(s) and performance. When properly understood and implemented, operations research serves as the interface between the researcher and the program manager.

Recommendation
The Ministry of Health should explore ways to pro-vide incentives for research staff to carry out research on topics agreed with operational (programmatic) units, including researchers from academic community all over the country. To improve evidence-based health policy at the national level, evidence-based health practice is started by understanding the concept of 'evidence-based.' UNICEF's concept of 'evidence-based' identifies two roles in the process: users of evidence or the stakeholders and policy-makers, and the suppliers of evidence or the researchers. 10 If the evidence (information) is program related, the users can then use the evidence to recognize a policy issue, which may until then have been hidden from the general public, and from policy-makers. Once this is revealed, the users (civil servants, non-government organizations, development agencies or the media) can address this issue; to analyze the identified policy issue(s) to understand the extent and nature of the problem, and then use it as the basis for making policy recommendations; to forecast the future, to see whether a target or policy measure in the short-run will be successful in the long-run as well (or, to assess whether targets are likely to be met); to monitor policy implementation, whether key outputs and outcomes are relevant and associated with the targets (objectives of the policies), or whether these key indicators are going off-track (and thus need a change of policy); and to evaluate policy impact, to see whether each policy produces implicit and explicit impact that are measurables. 10 Beside understanding the concept of 'evidence-based', another way to improve evidence-based health policy at the national level is by understanding how to apply operations research and select operations research variables to be more specific. The operations research principles include the goal of operations research that is to provide program managers with information they can use to make decisions to improve their programs' operations. This goal can help managers decide between alternative courses of action, identify and take advantage of opportunities, and find solutions to service-delivery problems that limit program effectiveness and efficiency.
The next principle is to phrase these criteria in the language of experimental design: operations research positions independent variables as factors that can be manipulated by managers, i.e., type of training, frequency of supervision and prices. Then dependent variables are indicators of program success, such as program outputs (e.g. number of clinic visits, contraceptives distributed), outcomes (e.g. client knowledge, contraceptive continuation rates, prevalence rates), and individual or popula-tion impacts (e.g. fulfillment of individual fertility desires, prevention of unwanted pregnancies, maternal morbidity), or cost-effectiveness of program operations. Operations research does make a clear distinction between independent variables that cannot be changed quickly and those that can be changed by programs in the short term. 11