Researchers' role in policy decision-making: purveyors of evidence, purveyors of ideas?

and responses from Lomas (Lomas 2005a) and Lavis (Lavis 2005) under the umbrella title Perspectives on Evidence, Synthesis and Decision-Making. Together with Lomas' s essay in our inaugural issue (Lomas 2005b), they provide sometimes complementary , sometimes competing, but invariably thoughtful perspectives on how (and to what extent) research and other evidence can be synthesized, presented and received to inform health policy and management decisions. At the risk of overkill, I now dip my oar in these already busy waters to consider the role of researchers in the decision-making process. In their commentary, Pope and colleagues make a useful distinction between reviews or syntheses that provide knowledge support and those that offer decision support. The decision support mode engages researchers with policy makers, managers, stakeholders and " experts " in a consideration of policy options. Beyond summarizing and clarifying the relevant evidence, what are the appropriate roles of researchers in this inevitably messy process? They might reasonably be expected to advocate for the research evidence, challenging policies that ignore or defy evidence pointing to a high probability of public benefit or harm. But should the line be drawn there or is it appropriate for researchers to join the policy fray as advocates for ideas, values or poli-cies? Aren't researchers' ideas as good as anyone else' s – maybe better if they've been immersed in the policy area under consideration, sometimes for an entire career. On the other hand, researchers' ideas and arguments reflect not just their expertise but their entire life experience and may contain varying degrees of personal or professional self-interest. Wearing the mantle of objectivity, researchers are well positioned to inject their values into policy discussions in the guise of evidence. By becoming advocates for ideas, researchers may debase their currency as purveyors of evidence. Clearly, these are questions that individual researchers, research synthesis teams and those who establish the rules of engagement for a decision support synthesis process will need to address and resolve.