Context The long-term effect of policies restricting contact between residents and pharmaceutical company representatives (PCRs) during internal medicine training is Unknown. The McMaster University Department of Medicine in Hamilton, Ontario, implemented a policy restricting PCR contact with trainees in 1992, whereas the Department of Medicine at the University of Toronto, Toronto, Ontario, has no such policy. Objective To determine if the presence of a restrictive policy and the frequency of contact with PCRs during internal medicine training predict attitudes and behavior several years after completion of training. Design, Setting, and Participants Retrospective analysis of the attitudes and behavior of 3 cohorts of physicians: University of Toronto trainees, prepolicy McMaster trainees, and postpolicy McMaster trainees. Surveys were mailed to 242 former University of Toronto and 57 former McMaster trainees who completed their internal medicine training between 1990 and 1996, with response rates of 163 (67%) and 42 (74%), respectively. Main Outcome Measures Physician attitude, assessed by a question about the perceived helpfulness of PCR information, and behavior, assessed by whether physicians met with PCRs in the office and the frequency of contacts with PCRs (current contact score, consisting of conversations with PCRs, PCR-sponsored events attended, gifts, honoraria, and consulting fees received). Results In both the unadjusted and multiple regression analyses, postpolicy McMaster trainees were less likely to find information from PCRs beneficial in guiding their practice compared with Toronto and prepolicy McMaster trainees, with unadjusted odds ratios (ORs) of 0.44 (95% confidence interval [CI], 0.20-0.94) and 0.39 (95% CI, 0.13-1.22), respectively. All 3 groups were equally likely to report that they met with PCRs in their off ice in the past year (88%). Postpolicy McMaster trainees had a lower current contact score compared with Toronto (9.3 vs 10.9; P=.04) and prepolicy McMaster trainees (9.3 vs 10.8; P=.18). In multiple regression models, greater frequency of contact with PCRs during training was a predictor of increased perceived benefit of PCR information (OR, 1.29; 95% CI, 1.13-1.47) and was positively correlated with the current contact score (partial r=0.49; P<.001). Number of PCR-sponsored rounds attended during training was not a consistent predictor of attitudes or behavior. Conclusions Policies restricting PCR access to internal medicine trainees and the amount of contact during residency appear to affect future attitudes and behavior of physicians.
