Joseph Alpert on graduate medical education

Dr. Alpert, Editor-in-Chief of the American Journal of Medicine, has written two articles [1] [2] about books he feels are must reads for anyone involved in graduate medical education. Relatively late in my career I am becoming involved in graduate medical education. I am eager to learn all I can about it and these titles naturally caught my eye. Alpert references a series of books on the subject by Kenneth M. Ludmerer. Drawing from these books he points to a degradation in the quality of graduate medical education over the past few decades. The end result of this decline is that trainees now have insufficient time to evaluate, read or think about individual patients and their disease processes. Dr. Alpert suggests two external pressures responsible for this: 1) regulations on resident work hours and 2) shorter and shorter lengths of stay for hospitalized patients.

Dr. Alpert suggests that the 80 hour work limit might be reasonable if programs had more flexibility in scheduling. It should be pointed out, though, that while the 80 hour limit makes life easier for residents there has been no evidence that it has improved patient safety, which is what it was designed to do. Like a lot of so called systems improvements the work hour regulations were not based on evidence at all. In fact, they were based on a single dramatic anecdote (a case of missed serotonin syndrome back in an era when that entity was not well defined).

Reduced hospital lengths of stay have been driven by economics. Gone are the days when patients would be kept on service a few extra days so they could be presented at a weekly subspecialty conference. Inefficient though that practice was it had its educational appeal.

A related development, not mentioned specifically by Dr. Alpert, is the pressure to make a diagnosis with too much specificity too early. This pressure comes from the coding world. It runs counter to appropriate clinical reasoning and has the potential to lead to diagnostic error.

Alpert's essays provide great food for thought but I partially disagree with one point. In advocating for the teaching of a systematic and logical approach to diagnostic evaluation he says:

The second element required is sufficient time for trainees to question and examine each patient carefully, followed by time for intellectual reflection on precisely which tests should be ordered. Trainees should be told that the current “shot gun” approach to testing should be carefully avoided.

That may be sound as a general clinical rule but what about situations in critically ill patients where multiple diagnostic possibilities exist each of which needs to be diagnosed and treated within mere minutes? Sometimes time is limited because time is muscle (in the case of acute coronary syndrome), brain (in the case of stroke) or mortality (in the case of, for example, sepsis or aortic dissection). In such cases, less well known in the days of my early training, a deliberate focused approach is impractical and may have to give way to a shotgun approach.