Measuring patient satisfaction for audit in general practice.

medical literature into a form easily assimilable by busy practitioners is defining and producing practice guidelines. Many practice guideline programmes exist, both in the United States (for example the NIH Consensus Conference program which has been running since 1977) and in various European countries. But there is little evidence that the diffusion of such guidelines effectively modifies clinical practice.'7 As sociologist Ann Greer explains, mere diffusion of written matter is not sufficient to modify practices ingrained in behavioural, social, and local factors.'8 Some of the complexities of implementing guidelines in practice are described by Richard Grol in this issue (p 184).'9 An approach to resolving uncertainty of medical practice would be to develop and integrate three disciplines-medical decision analysis, clinical epidemiology, and evaluation of the quality of health care-into medicine and thus put scientific principle into everyday clinical practice. Medical decision analysis can be described as "a systematic approach to medical decision making under conditions of uncertainty."20 It is an aid to reasoning in that it separates the logical structure of a decision into its component parts and quantifies the probabilities of the different possible outcomes. The use of decision analysis forces the decision makers to consider explicitly the reasons and consequences of each choice. This allows each decision to be explained and justified to patients, to other health care workers and to the decision makers themselves. Clinical epidemiology-the application of the principles and methods of epidemiology to clinical practice-promotes the practice of "a more systematic approach to gathering and interpreting clinical evidence"2 and provides a framework for answering, with scientifically based argument, questions such as, "How accurate are the diagnostic tests or strategies I've applied? What will the consequences be of having this disease? How will the treatment I've prescribed change its future course and affect the patient?"22 Clinical epidemiology thus drives medical practice away from empiricism and tends to give it the characteristics of a science: it has been described as "the basic science of clinical practice."2' Finally, through systematic evaluation of the quality of clinical care it is possible to assess the quality of the structure of care, the appropriateness and quality of medical procedures, and the desirability of outcomes for the individual patient.23 For this a methodical, rigorous, problem solving cycle which follows scientific principle is applied.24 By using an explicit and structured approach subjective influences are minimised and problems and their nature can be identified …