ABSTRACT
The study question determines the relevant design. The policy question, the intended decision makers and their objective functions determine the relevant data and evidence.
These two homegrown truths are all too often overlooked. Many discussions about the merits of randomised controlled trials (RCTs) disregard these simple observations and are often accompanied by confusion about the exact role of scientific evidence in the policy-making process. This is even more the case at a time when ‘evidence-based’ this or that is the obvious, tempting and almost politically correct catchword. Clearly, this arises most frequently in the context of evidence-based medicine (Sackett et al. 1996; Sackett and Rosenberg 1995), which is the focus of this book, and extensions to evidence-based policy (Ham et al. 1995; Klein 2000; Innvaer et al. 2002). This last is exemplified by the WHO in its World Health Report 2000 on improving the performance of health systems, which raises doubts about what, according to the WHO, constitutes evidence (WHO 2000). The WHO also maintains a special section on its Web site about evidence and information for policy, and has a global programme on evidence for health policy. WHO is co-sponsoring the European Observatory on Health Care Systems, which supports and promotes evidence-based health policy making through comprehensive and rigorous analysis of the dynamics of health care systems in Europe. More generally, in the United Kingdom the government is trying to create a culture of ‘evidence-based policy’ that will apply to all policy areas, including public health and health care (Strategic Policy Making Team 1999). It is likely, indeed, that governments worldwide subscribe to the use of evidence in policy making, including the pharmaceutical area. Logically, all this requires a clear definition of what constitutes evidence – who could be against evidence? – and what policy and policy making are, and how evidence enters such processes.
