Influence of longitudinal primary care training on medical students' specialty choices

At the University of South Florida College of Medicine, a program designed to give students four years of primary care training began in 1983. As of 1992, six classes that included program participants had graduated. The present study examined the effect of the program on the participants' choices of specialty by comparing their choices with those of other graduates.

Each year program volunteers were solicited from a class size of about 96 freshmen. Of the 201 volunteers from the classes of 1987-1992, 93 were randomly selected to participate in the program. The participants received primary care education one half-day per week in a community-based clinical setting. The specialty choices of the 543 graduates from 1987-1992 were determined from the National Resident Matching Program and were divided into primary care (family practice, internal medicine, pediatrics); surgery, obstetrics-gynecology, and psychiatry; and other (high-technology specialties). Comparisons were made (1) between volunteers and nonvolunteers and (2) between volunteers who were participants and those who were not. The Z-test was used, with alpha set at .01.

Significantly more volunteers--with no difference between participants and nonparticipants--matched with primary care specialties, and more volunteers were women. Significantly more nonvolunteers matched with high-technology specialties--again, no difference between participants and nonparticipants.

The program did not seem to influence the students' specialty choices, because students interested in participating--regardless of whether they actually participated--were more likely to match with primary care specialties and were more likely to be women.