TO THE EDITOR: The prospect of an oversupply of physicians has led some experts to propose a reduction of the total number of graduating residents to 110% or 120% of the number of graduates of U.S. medical schools. However, not everybody agrees with the prediction of an oversupply of physicians. There is also some evidence that medicine has attempted to address this issue through various measures, including limiting the number of international medical school graduates (IMGs) entering residency programs. Nasir (
+1) demonstrated a pattern of dissimilarity in the way family practice programs responded to requests for residency applications. In this study, 50% of responding programs replied to an IMG applicant, whereas 90% replied to a U.S. medical school graduate asking for an application for residency. We conducted a similar study (
+2), in which residency programs in psychiatry responded significantly more frequently to a U.S. medical school graduate asking for an application than they did to an IMG.
We then investigated whether the results of our study among psychiatry programs are applicable to other disciplines; that is, whether other disciplines respond more frequently to an application request from a U.S. medical school graduate than to one from an IMG. During November 1998, we mailed two letters requesting an application for a residency position to 105 randomly selected family medicine (approximately 22% of the programs), 103 internal medicine (24%), 105 neurology (86%), 95 obstetrics/gynecology (36%), 101 pediatrics (46%), and 101 general surgery training programs (38%). The letters differed in only two respects: the names of the writers (American or Indian) and the medical school from which the writers graduated (United States or Indian). Example of the letter:
I am a graduate of the University of …. I am very much interested in a residency training position in [specialty]. Kindly send me an application for your residency training program.
Return addresses were both from southeastern Michigan.
The study was approved by both institutions' Internal Review Boards.
We received 167 responses for the IMG and 256 responses for the U.S. medical graduates (27% vs. 42%; chi-square=28.66; P<0.0001). All disciplines except obstetrics/gynecology had higher point response rates for the U.S. graduates. However, the difference was significant only for family medicine (57% vs. 32%; chi-square=13.26; P<0.0001); pediatrics (43% vs. 17%; chi-square=16.03; P<0.0001); and internal medicine (41% vs. 24%; chi-square=6.39; P=0.01). The response rates for the U.S. graduates were marginally higher in neurology (42% vs. 30%; chi-square=3.51; P=0.06) and surgery (40% vs. 29%; chi-square=2.66; P=0.1). In obstetrics/gynecology, the rates were 28% for the U.S. graduates and 34% for the IMGs (chi-square=0.62; NS).
The detailed analysis of positive vs. negative, as well as further quality of responses, are both beyond the scope of this letter. The quality of responses was mostly similar. Some programs specified the visa status they required from IMG applicants. In both family medicine and obstetrics/gynecology, most of the responses to both requests suggested using the electronic application system (ERAS). In two disciplines, neurology and pediatrics, we noted, occasionally, a qualitatively different response (for example, a letter to the IMG stated that the interview process was finished, whereas the U.S. graduate received the application; or the IMG was informed that he had missed the deadline, whereas the U.S. medical graduate received the application).
Our study had several limitations:
Nevertheless, the results of this study, in combination with the results of the studies among family medicine and psychiatry programs, demonstrate different response rates for the application for residency training positions, mostly in favor of U.S. medical school graduates. Results of these studies suggest that medicine is trying to limit the influx of IMGs at the first level—the request for an application for a residency position. As we pointed out before (
+2), we can only guess about the reason for the higher rejection rate for IMGs, but discrimination seems to be a plausible explanation.
The authors thank Cynthia L. Arfken, Ph.D., for statistical analysis and expertise.