Correspondence

Clinical Problem-Solving

N Engl J Med 1995; 333:1157October 26, 1995

Article

To the Editor:

Your editorial in the June 1 issue of the Journal 1 gives an in-depth discussion of differential diagnosis and how it is being taught. I would like to comment on the concept of probability. Corporate executives use probability to make decisions about product development and marketing. Players of the stock market use probability to choose investments and predict price movement.2 A knowledge of mathematical probabilities will not make a good poker player, but a total disregard for them will make a bad one.3 Probability is also used implicitly by all physicians when they see patients. A five-year-old complaining of chest pain will not be sent to the coronary care unit. The same is not true for a 45-year-old.

I believe that probability analysis can make the process of differential diagnosis easier and that medical students should learn about it. Why? Because it limits the illnesses to be considered by excluding those that are rare. By focusing on the disease entities that are more probable, physicians should always be closer to the true diagnosis. They can be more focused in their thinking. The number of tests to be ordered can be reduced.

This last point brings us to the most pressing issue in medicine today: the cost of care. Fewer tests lead to lower costs. Adequate money management is at the foundation of every successful business, stock trader, and poker player. You cannot continue to play the game if you do not have any money left to bet. This is a lesson that all physicians need to understand.

Daniel Adler, M.D.
106 Prospect St., Ridgewood, NJ 07540

3 References

1. 1

   Kassirer JP. Teaching problem-solving -- how are we doing? N Engl J Med 1995;332:1507-1509
   Full Text | Web of Science | Medline

2. 2

   Fishback D. Applying probability analysis to option trading. Cedar Falls, Iowa: Futures Learning Center, 1994.
   

3. 3

   McDonald J. Strategies in poker, business, and war. New York: W.W. Norton, 1989:22-3.
   

To the Editor:

I enjoyed your editorial on the effectiveness of teaching clinical problem-solving through the series of case vignettes launched in the Journal some two years ago. Although many of my colleagues have voiced the criticisms you summarized (e.g., substandard care was practiced, too many tests were performed, important clinical clues were overlooked), I continue to find these vignettes to be useful educational tools. Over the past year I have presented many of these cases to both medical students and the house staff of the Department of Internal Medicine, using the interactive approach favored by the editor of the Journal 1 — that is, each case is presented in small chunks with intervening spaces for discussion. This practice not only allows learners to play the part of master clinician and to formulate their own differential diagnoses and therapeutic plans, but it also allows them to compare their performances with those of the Journal's experts. What usually results is a very lively forum in which each diagnostic step is placed under a critical microscope and examined for appropriateness, cost effectiveness, and so on. Most participants are quick to seize on any deviation from logical clinical reasoning and in many cases reach a correct diagnosis before the experts in the published case. Thus, we have learned a great deal not only from the cases in which an appropriate diagnosis was reached rapidly in a cost-effective manner by an expert clinician, but also from the controversial cases in which clinical misjudgments and even gross errors delayed or prevented a correct diagnosis. In fact, I believe many of our best discussions grow from a thoughtful consideration of the errors made by both the case physicians and our own student “master clinicians.”

Geoffrey S. Greene, M.D.
University of South Florida College of Medicine, Tampa, FL 33606

1 References

1. 1

   Kassirer JP. Teaching clinical medicine by iterative hypothesis testing: let's preach what we practice. N Engl J Med 1983;309:921-923
   Full Text | Web of Science | Medline