Referral Patterns for Coronary Artery Disease Treatment: Gender Bias or Good Clinical Judgment?

Abstract

OBJECTIVE: To determine whether a gender bias exists in referral for coronary bypass graft surgery among patients with catheterization-documented coronary artery disease. DESIGN: Historical cohort study (1969 to 1984). SETTING: A referral medical center. PATIENTS: A total of 5795 patients with catheterization-documented coronary artery disease. MEASUREMENTS: Surgical referral patterns of men and women grouped by risk for cardiac death and by treatment effectiveness. Time trends were evaluated for three periods: 1969 to 1974, 1975 to 1979, and 1980 to 1984. RESULTS: Overall, when no adjustment was made for baseline risk for cardiac death, no statistical difference was found between men and women regarding referral for surgery (46% compared with 44%, respectively). When an adjustment was made for such risk, the male-to-female odds ratio for surgical referral was 1.28 (95% Cl, 1.05 to 1.58) among patients with a low risk for cardiac death. This effect was most evident in the 1980 to 1984 period (odds ratio, 1.73; Cl, 1.29 to 2.31). In the high-risk group, the odds ratio was 0.84 (Cl, 0.68 to 1.04), with little change occurring during the study. Men were more likely to be referred for surgery when surgery offered the least survival benefit relative to medical therapy (odds ratio, 1.29; Cl, 1.08 to 1.54). This effect was most pronounced in the 1980 to 1984 period (odds ratio, 1.63; Cl, 1.27 to 2.10). CONCLUSIONS: Women are less likely than men to be referred for coronary bypass graft surgery among patients with a low risk for cardiac death, in whom surgery offers little or no survival benefit over medical treatment. Women are at least as likely as men to be referred for bypass surgery among more symptomatic and more severely diseased patients, in whom surgery offers the greatest survival benefits. These trends were most prominent in the recent years of the study. Based on surgical survival benefits, these referral patterns may represent more appropriate treatment referral for women than men.