retired Doc's suggestion for medical curriculum,part 11, pathophysiology

Pathophysiology was a major emphasis in the specialty of internal medicine which had its origins in Germany (Inner Medizin) and was transported to the U.S. in the early 1900s.By 1915 the American College of Physicians was founded and by 1936 the American Board of Internal Medicine.In the early years all internists were practioners of general internal medicine although they were not so named.Even when I trained, in the 1960s, all of the IM faculty were thought of as internists though most all had a speciality as well. There was no section of general internal medicine. Pathophysiology was emphasized and in my school days most frighteningly on Saturday mornings when various clinical profs would teach in their area of special expertise and ask questions. Most dreaded was the appearance of the chief of medicine. When he finally arrived for his lecture ( there was no published schedule) some of our fears were validated. He asked which atrium usually fibrillates.One of the students who was at the top of the class, volunteered an answer.He opined it was the left atrium because he figured rheumatic heart disease and mitral valve damage were a fairly common cause of atrial fibrillation. The prof said " No concept of pathophysiology, when one atrium fibrillates , both atria fibrillate , no concept at all" The emphasis on pathophysiology was also evident in the surgery courses and pediatrics. We grew up medically thinking that way. So it is with more than a little concern when I read some ( certainly not all) meta-analysis (MAs) and RCTs which make no mention of disease mechanisms. MAs seem to be more guilty of that omission.
Medicine has a long history of determinism which is basically elucidating and studying disease mechanisms or pathophysiology.This is one reason, perhaps, why physicians over the years have distrusted statistics with its randomness implication. The best thought experiment I have heard to highlight the distinction between the random and the deterministic ways of thinking was brought to my attention by Dr. Steve Goodman from Johns Hopkins in an wonderful Annals of Internal Medicine article. ( Here is another one of those articles that should be on the medical students reading list). Here is my paraphrased version. Mr. Jones is faced with the need for surgery.The particular procedure is generally accepted to pose a 15% risk of death. Let us magically produce 100 clones of Mr. Jones. When they all undergo surgery, what will happen? In the random process model ( stochastic interpretation) , 15 will die but we cannot tell beforehand who they will be. In the deterministic model, either all 100 will live or all 100 will die depending on whether Mr. J. and all his clones have or do not have some biochemical or physiological condition(s) that are in fact what causes the mortality risk of the procedure.
The recent meta-analysis I mentioned in my blog that concluded that beta lactams were the drug of choice in community acquired pneumonia was devoid of any disease mechanism discussion.The authors point out that even in cases of atypical pneumonia (excepting Legionella) that beta lactams, which are not senstive to atypicals, produced results equivalent to those produced by antimicrobials which are known to eliminate those organisms. Now, either we have to rethink our concepts of community acquired pneumonia or something is misleading about the data.
Outcome analysis is, of course, important. However, let's not forget Claude Bernard's admonition [quoted from Goodman's article]"What really should be done, instead of gathering facts empirically (I think if Bernard were writing today he would realize the importance of statistics but not to the exclusion of the following) is to study them more accurately, each in its special determinism". The data alone are not enough, we need to figure out what's going on. As Goodman has emphazied we need to consider prior data and biological plausibility ( or pathophysiology) as well as the latest outcomes research factoids. The tools of outcomes research and aggregate data analysis are seductive and computers have transformed the heavy statistical lifting to low energy key strokes but let's not forget the deterministic partner who brought us to the dance.