Does the number of trainees in a postgraduate training program influence the pass rates on certifying examinations?

We tested the hypothesis that small training programs (3 or fewer residents) lack the "critical mass" needed for an optimal learning experience, and thus graduates of small programs will have a lower pass rate on the Royal College of Physicians and Surgeons of Canada (RCPSC) certifying exams that graduates of large (10 or more residents) training programs. Pass rates on the RCPSC certifying exams (written and oral) were compared to the training program size for each of 6 years from 1984/85 to 1989/90 within 10 of the 43 RCPSC (sub)specialties selected by meeting predefined program size requirements. These 10 specialties met the size variation requirements needed to test the hypothesis: neurology, cardiology, emergency medicine, community medicine, neurosurgery, urology, plastic surgery, dermatology, anatomical pathology, and respiratory medicine. Of these, 3 specialties had a significantly lower written exam pass rate for candidates trained in small compared to large programs. The same 3 specialties (neurology, neurosurgery, and community medicine) had a higher proportion of International Medical Graduates (IMGs) in small training programs. The significantly lower pass rate of IMGs, compared to Canadian/USA graduates, accounted for a portion of the correlation of small program size with lower pass rates in these 3 specialties. By pooling the results from the 10 specialties evaluated, candidates from small (3 or fewer residents) training programs have slightly lower pass rates (11%) on written certification examinations compared to candidates from large (10 or more residents) training programs. This small but statistically significant effect on the pooled results was due to averaging of a more marked program size effect from 3 of the 10 specialties.(ABSTRACT TRUNCATED AT 250 WORDS)