There is a fine balance that needs to be maintained between research and improvement in safety and quality in health care -when do we need more research and when can we just get on with it? The moral imperative to improve care may have been a distractor, preventing adequate attention to research. Three research areas are proposed as current priorities for patient safety: getting evidence into practice, measurement of safety, and the evaluation of complex interventions. A focus on these areas should ensure that research becomes more central to the process of WHAT PLACE DOES RESEARCH play in safety and quality and when should we just "get on with it"? The discourse of modern government is crisp, if not blunt: the National Health and Hospitals Reform Commission had among its terms of reference the requirement to "reduce inefficiencies generated by cost-shifting, blame-shifting and buck passing". 1 (p. 357) The development of performance indicators continues apace, and the research community is not always seen as essential to the process of making health care safer.The research-policy divide is well documented and has led to the modern development of new specialties such as knowledge transfer or translation, 2 and brokering and exchange. 3 However, policy imperatives mean policy makers are impatient with research timeframes and wary of recommendations that may make compromise difficult. 4 Human interaction has been described as "the engine that drives research into practice", 5 yet the inclusion of clinicians, managers and policy makers (or research users) into research funding programs is rare in Australia and only partial in Canada. 5 In the case of quality and safety, however, there are further problems. Traditional quantitative research methods may be inadequate tools (particularly for areas such as clinical handover and other health care communication), and in addition, the need to improve safety and quality has been associated with a moral imperative which has been held to obviate the need for research and has been associated with denigration of its value. Berwick, in a recent piece advocating safety and quality research, used rhetoric displaying the customary antagonism of safety and quality exponents towards research: "Health care researchers What is known about the topic? It is unsure how much safety and quality has improved over the last two decades. There is a deficiency of high quality research to support the design of safety improvement.
What does this paper add?The lack of formal study means that neither clinicians nor policy makers understand the system in which they work sufficiently to reliably get evidence into practice. Inadequate measurement in safety prevents appropriate priority setting. Meaningful evaluations must include the study of failure. The use of a theoretical basis for both intervention and evaluation increases the likelihood of transferable learning. What are the implications for practitioners? Safety and quality advocates are requested to accompany their exhortations for cha...