In response to: G.D. Sweeney
Dr. Sweeney has put his finger on an important "missing link" in the great chain of medical evidence: the lack of adequate teaching. For although a few controlled studies1 have demonstrated the effectiveness of teaching rigorous approaches to clinical reasoning (including Bayesian analysis), the science of clinical decision-making still receives far less recognition in medical school and postgraduate curricula than it deserves. Unfortunately, enthusiasm for such teaching is dampened by the difficulty of applying our current formal decision analytic techniques at the bedside (an experienced consulting service at Tufts-New England Medical Center in Boston found that most full decision analyses of individual clinical decisions required about 2 weeks of effort by a clinical fellow [personal observation]). It may be more realistic to think of decision science as an imaging technique for visualizing medical reasoning2 and employ it as we do other basic sciences such as anatomy, rather than as a clinical science such as surgery.
Our lack of understanding of the role that basic biological knowledge actually plays, and the role it should play, in clinical work is little short of scandalous. The small number of studies that have examined this question empirically have concluded that the use of basic biological knowledge by experienced clinicians is minimal and indirect.3 Sweeney is therefore quite justified, I believe, in questioning whether basic science, at least as we now teach it, is a necessary part of medical education.3
Frank Davidoff, MD