Gender differences in the practice of adult primary care physicians

Objective: This study investigates how physician gender affects reactions to six model patients. Methods: Telephone interviews with 3205 internists and family or general physicians were completed between 1996 and 1997 for the Community Tracking Study. Physicians responded to six vignettes describing model patients with presentations designed to have multiple appropriate treatment plans: a 50-year-old man with a 1-month history of exertional chest pain who may need a referral to a cardiologist, a 60-year-old man with a normal digital rectal examination (DRE) who may benefit from a prostate-specific antigen (PSA) test, a 40-year-old married woman with vaginal itching and discharge who may benefit from an office visit, a 60-year-old man with symptoms of benign prostatic hypertrophy (BPH) who may benefit from a urological consultation, a 35-year-old man with back pain and a new left footdrop who may benefit from an MRI, and a 50-year-old man with elevated cholesterol and no other cardiac risk factors who may benefit from cholesterol-lowering agents. Results: Female physicians are significantly more likely than males to refer a patient with BPH to a urologist (37.5% vs. 24.9%, p<0.001). Male physicians are significantly more likely to recommend that a woman with vaginal itching and discharge have an office visit (52.7% vs. 40.6%, p<0.001). Male physicians recommend cholesterol-lowering agents slightly more often than women physicians (39.4% vs. 36.4%, p<0.03) and recommend a PSA test more often than female physicians (73.1% vs. 64.4%, p<0.001). Neither physician attributes, practice characteristics, referral patterns, nor geographical traits account for the disparity between male and female physicians in the treatment of BPH. Approximately 40% of the gender gap in treatment of vaginal itching and discharge can be attributed to physician attributes. The association between provider gender and the decision to prescribe cholesterol-lowering agents and the association between provider gender and recommending a PSA test are explained by physician attributes. Conclusions: Gender differences in treatment seem to appear most strongly for genital-specific conditions. The results may suggest that physicians use fewer resources to treat the genital-specific conditions of patients who share their sex.