This paper reports findings from interviews with both senior policy makers and researchers in NSW. While the samples were small, this reflects the size of the relevant research and policy communities in NSW, and the response rates were good.
Policy makers and researchers recognise the potential of research to contribute to policy and are making significant attempts to integrate research into the policy process. Most policy makers reported having needed data and reviews in the past 12 months, having commissioned research or reviews during this period, and having used evidence to contribute to the content of policy. The rates of use of evidence by policy makers appear to be somewhat higher than those reported in previous Australian surveys (eg [19
]). Similarly, over the past 12 months, most researchers felt that their work had had a policy impact; in particular, they felt that their research had contributed to raising awareness of issues and to setting policy agendas.
However, policy makers and researchers agreed that much more could be done to increase the use of research in policy. Reports of current practice indicated that only around half of the researchers thought their research had been used to get issues on the policy agenda or select preferred policy options in the past two years. Although policy makers drew on research findings to contribute to the content of policy, it was not often used to set agendas or to evaluate policy.
This paper identifies four potential strategies for increasing the use of research in policy.
First, making research findings more accessible is likely to be helpful. Policy makers reported that they often found it difficult to access brief summaries and systematic reviews. Many respondents also indicated that research conducted in NSW was often not presented in a useful way to inform policy and program issues. Similar results have been reported by others [19
]. The difficulty in accessing reviews was reported despite the fact that NSW Health employees have access to research reviews on the web through the Clinical Information Access Project (CIAP), encompassing a wide range of research databases such as Medline and Cochrane. In contrast, participants reported that they could easily access local data and that these data were very useful. Decision makers at NSW Health have access to data resources including the Health Outcomes Information Statistical Toolkit (HOIST) which houses a wide range of relevant population-based data collections (eg census data and morbidity and mortality data). This system allows manipulation, analysis and reporting of data at the state and regional level. There appears to be a high level of awareness and use of this system and other sources of local health data within the Area Health Services and Department of Health.
Researchers reported a high level of effort in disseminating their research to policy makers. Thirty nine percent of respondents had regularly developed targeted strategies for communicating their findings to non-academic audiences. While peer review papers and conference papers remain the standard methods of dissemination, there is certainly evidence of a second tier dissemination strategy aimed at policy makers, primarily through research reports and presentations.
However, despite these efforts by researchers, policy makers still found it difficult to access research findings. It seems likely that new approaches are required that more closely target the specific needs of policy users [25
]. There is evidence that decision makers appreciate very brief summaries of research findings, preferably with a more detailed structured abstract and along with clear statements of implications for practice [27
]. The Canadian Health Service Research Foundation's (CHRSF) 1:3:25 format (one page of main messages, a three-page executive summary, and a 25-page report), for example, is widely recommended [29
Second, increasing the opportunities for interaction and exchange between policy makers and researchers is key to promoting the use of research evidence in policy. This was identified by both policy makers and researchers in our samples, consistent with the findings of two systematic reviews [3
]. Our respondents reported a moderate current level of interaction, indicating that partnership and collaboration is feasible. For example, around half of the researchers reported being involved in policy development committees, being used as a sounding board or being funded by government for policy-relevant research. Likewise, around half of the policy makers indicated that they were involved in some research activities. This was in broad agreement with researchers' views about their involvement of policy makers.
Opportunities for researchers and policy makers to meet informally and mechanisms to help policy makers and researchers to identify individuals relevant to their work are likely to be important in promoting exchange. Policy makers in our sample reported that they often wanted to seek advice from researchers, but sometimes could not find the expertise that they needed, and that they tended to use existing contacts. Researchers perceived that input from policy makers into their research would be of value but were often not sure how best to identify appropriate individuals.
A greater intensity of interaction and exchange is achieved by actively involving policy makers in conceptualising, designing, and implementing research [31
]. Experiences with 'research partnership' models in the UK suggest that the involvement of decision makers both focuses the research on users' needs and encourages a better understanding on the part of researchers of the decision making environment [32
]. In Australia, funding schemes such as Australian Research Council linkage grants and the National Health and Medical Research Council's new Partnership Awards require the formal involvement of an industry partner, and are important mechanisms for encouraging collaboration between researchers and policy makers. Internationally, the CHRSF's Research, Exchange and Impact for System Support program [33
] and United Kingdom's National Institute for Health Research Service Delivery and Organisation Programme [34
] might provide models for the future in Australia.
Third, there are clearly some structural barriers to increasing the use of research in policy that could be addressed. Both policy makers and researchers felt that enhancing policy makers' understanding of research is important; likewise, the need to improve research infrastructure and funding was regarded as important in generating relevant evidence. Policy makers felt that organisational reinforcement for evidence-informed policy could be improved. Although researchers agreed that there was a high need to increase the use of research by policy makers, more than one-third of the respondents in the current sample did not regard these activities as being a high personal priority. This is in part the result of a perception among researchers that their efforts to impact on policy were not valued by Universities or by funding agencies. This view is probably well founded; for example, in Canada, Phaneuf et al. [35
] surveyed academic promotion committee members and found that they regarded peer review publications as substantially more important in their decision making than work with policy. Although it was not directly examined in this study, it is evident that concerns about intellectual property, independence and the right to publish are also structural barriers to a greater engagement of researchers with policy agencies [36
There seems little doubt that it will be necessary to address these structural barriers to increase the use of evidence in policy. In terms of increasing the receptivity of policy makers to research, the two main approaches that have been described are the use of tools to assess organisational capacity to acquire and apply research evidence (for example a self-assessment tool developed by the CHRSF [37
]) and continuing education programs (for example the CHSRF-sponsored Executive Training for Research Application (EXTRA) program [38
]). A relevant Australian example is the Australasian Cochrane Centre (ACC) Policy Liaison Initiative, through which the ACC provides support and training to an Evidence-Based Policy Network within the Australian Government Department of Health and Ageing [39
With regard to incentives for researchers to engage in research transfer activities, there is a need to develop a measure of the impact of research on policy. A reliable measurement tool would enable these activities to be included in consideration of applications for promotion or in assessment of research track record for funding applications. For example, the Netherlands Council for Medical Sciences has developed a methodology and indicators for evaluating the societal impact of applied health research [40
]; this is considerably more sophisticated than the generic measures for research impact proposed as part of the (now abandoned) Australian Research Quality Framework.
Finally, there was a view among policy makers that there is a lack of relevant research that could inform policy. Almost half the sample believed that the health research being conducted in NSW was not relevant, or had variable relevance, to health policy. To increase the relevance of research, policy makers need be able to clearly identify and communicate gaps in knowledge and policy priorities for research to researchers. A greater understanding of the policy context by researchers could increase relevance by focusing the research on more useful questions, collecting information critical for policy decisions (for example on costs) and improving the description of the research results and their implications. Research partnerships may improve the relevance of research and therefore its translation to policy [41
The development of a national system for health data linkage through the National Collaborative Research Infrastructure Strategy (NCRIS) Population Health Research Network presents particular opportunities for new policy-relevant health research in Australia. Linked person-based data for entire populations provides powerful information about the outcomes of health systems, and how these are shaped both by environmental factors, patient factors and service configuration. However, to provide the information health policy makers need, this enhanced capacity to describe and monitor system outcomes must be accompanied by new multidisciplinary research to develop health service interventions and test these in real-life service settings.
We intend to repeat our policy maker and researcher interviews in 2010. The 2010 sample of policy makers is likely to be almost entirely new, given the rapidity of change within the policy environment. Nonetheless, we would hope to demonstrate increased use of research evidence in health policy in NSW, as a result of the activities of the Sax Institute and initiatives such as the NHMRC Partnerships Program and the NCRIS Population Health Research Network, and reflecting a general, worldwide interest in promoting the efficient transfer of research evidence into policy.