Correspondence
Case Records of the Massachusetts General Hospital: A Home-Court Advantage?
N Engl J Med 1996; 334:197-198January 18, 1996
- Article
To the Editor:
Saint et al. (Sept. 28 issue)1 describe a “home-court advantage” for Massachusetts General Hospital (MGH) physicians in their discussions of the clinicopathological conference (CPC) cases. Although this certainly may be true, I would like to point out that a home-court advantage can only be proved if a team tests itself on the road. To use the example given in the article, if the Celtics played the Lakers only in Boston Garden (or in the new FleetCenter, if it imparts the same magic) and almost always won, the Celtics could simply claim to be the better team. To prove the authors' hypothesis, the MGH team would have to go “on the road” and test their oratorical skills and clinical acumen in less familiar surroundings.
As a medical resident, I find the CPCs a constant source of challenge and instruction. But I would like to see how the MGH wizards would fare in some away games.
1 ReferencesWilliam Greene, M.D.
Yale University School of Medicine, New Haven, CT 065101
Saint S ,Go AS ,Frances C ,Tierney LM Jr . Case Records of the Massachusetts General Hospital -- a home-court advantage? N Engl J Med 1995;333:883-884
Full Text | Web of Science | Medline
To the Editor:
Saint et al. reviewed data from the six-year period from 1989 through 1994 and showed that physicians affiliated with MGH performed better than physicians not associated with MGH in making correct diagnoses in the Case Records of the Massachusetts General Hospital. Up until 1991, medical students were also included in these CPCs. How did the students do as compared with the discussants?
I evaluated all 185 of the 312 CPCs published between January 1, 1986, and December 31, 1991, in which the medical students' diagnoses were published. I used the scoring system of Saint et al., awarding one point to the students if they were correct or partly correct in stating the main diagnosis. In five instances, the students listed disorders in the differential diagnosis; their diagnoses were counted as correct if the main diagnosis was considered. One case was excluded in which the medical students were given incorrect information.
The students made correct diagnoses in 132 of the 184 cases (72 percent), which is very close to the success rate of discussants not affiliated with Harvard Medical School (80 of 108 cases, or 74 percent; P = 0.67) and to that of discussants affiliated with Harvard Medical School but not MGH (71 of 93 cases, or 76 percent; P = 0.41). However, the students could not match the very high success rate of physicians from MGH (97 of 102 cases, or 95 percent; P<0.001).
These results confirm that medical students frequently arrived at the correct diagnoses. The fact that students were not involved in all the CPCs appeared to be a function of the vacation schedule of the medical school.1 The astonishingly high success rate of the beginners is notable. The fact that the medical students were rotating through the internal-medicine clerkship at MGH further supports the notion of a “home-court advantage,” which may be as valuable as decades' worth of clinical experience and knowledge.
1 ReferencesChi-yuan Hsu, M.D.
Beth Israel Hospital, Boston, MA 022151
Case Records of the Massachusetts General Hospital (Case 29-1986)
Full Text | Web of Science | Medline
To the Editor:
In an autopsy study of clinical diagnoses at a Harvard-affiliated institution in 1983, Goldman et al. found a consistent discrepancy rate of 22 percent between clinical diagnoses and autopsy results in 1960, 1970, and 1980.1 That rate is remarkably close to the 24 percent of diagnoses missed by physicians not affiliated with Harvard, as reported by Saint and colleagues.
The CPCs no longer include much information from autopsies. Dr. Cabot understood the importance of autopsy correlation, as evidenced by his startling 1912 paper on 3000 autopsies.2 During 1924, the year Dr. Cabot started the CPC (then known as the Ante-mortem and Post-mortem Records), the Journal published 156 cases, of which all the 134 resulting in death included autopsies.3 In 1994, there were 46 CPCs in the Journal. These included only 11 deaths; 3 of the reports (27 percent) described autopsy results (including 1 that had an unexpected finding). Without autopsies, do we really know how good our diagnoses are?
3 ReferencesJames R. Gill, M.D.
Yale–New Haven Hospital, New Haven, CT 065101
Goldman L ,Sayson R ,Robbins S ,Cohn LH ,Bettmann M ,Weisberg M . The value of the autopsy in three medical eras. N Engl J Med 1983;308:1000-1005
Full Text | Web of Science | Medline2
Cabot RC . Diagnostic pitfalls identified during a study of three thousand autopsies. JAMA 1912;59:2295-2298
Web of Science3
Case Records of the Massachusetts General Hospital: ante-mortem and post-mortem records as used in weekly clinicopathological exercises. Boston Med Surg J 1924; Vol. 190 & 191.
Author/Editor Response
The authors reply:
To the Editor: Dr. Greene correctly points out that a home-court advantage can be proved definitively only if a team performs better at home than on the road. It would be interesting to see discussants from MGH away from their home turf. Toward this end, we issue an invitation to any MGH discussant to discuss a case during a CPC here at the University of California, San Francisco.
Dr. Hsu, after comparing the diagnoses of medical students with those of the discussants, concludes that the home-court advantage may also apply to medical students. Although this is possible, another hypothesis is that the diagnostic ability of some third-year medical students at Harvard is similar to that of non-MGH faculty. Indeed, some high-school basketball players are quite capable of making the transition to the National Basketball Association without going through college (e.g., Darryl Dawkins, Moses Malone, and Kevin Garnett). Maybe some medical students (from Harvard and perhaps elsewhere) are similarly capable of making the transition directly to faculty-level positions without all the trouble of a medical residency.
Dr. Gill correctly points out that the autopsy remains the gold standard for assessing diagnostic accuracy. We agree that a higher percentage of autopsies would be desirable.
Finally, accuracy in diagnosis is by no means the sole measure of success, nor should it be. As the movie Hoop Dreams so wonderfully depicted, success often lies in the attempt. With this in mind, all the CPC discussants performed admirably.
Sanjay Saint, M.D.
Alan S. Go, M.D.
Lawrence M. Tierney, Jr., M.D.
University of California, San Francisco, School of Medicine, San Francisco, CA 94121
