From randomised trials to rational practice.

Abstract

From the age of Enlightenment onwards, philosophical thinking has become increasingly influenced by empiricism: observations lead to theories, but experiments are needed to put the reasoning to the test. However, it was not until the middle of the 20th century that well-designed experiments were at last introduced in medical treatment, in the form of randomised controlled clinical trials. This design is now standard in medicine, but in everyday practice a multitude of management decisions must still be taken without good evidence. There are several reasons for this: there may not be a trial at all or only a single trial; trial results may be equivocal; patients may be different from those enrolled in trials; new procedures require practice, or a trial may not be feasible. 'Logical reasoning', with all its fallacies, is still required - not only to fill the gaps in empirical knowledge but also to interpret existing evidence and to plan new trials. In fact, the generation of new knowledge is a continuous, cyclical process in which newly gained insights in pathophysiology give rise to new therapeutic experiments, the results of which generate fresh hypotheses, and so on. Compassion, curiosity and doubt are the essential forces that keep the cycle moving. Conversely, the progress is slowed down by present-day legalism, which distorts investigator accountability and patient autonomy.