Good judgement or sex bias in the referral of patients for the diagnosis of coronary artery disease? An exploratory study.

objectiveTo examine the sex-specific prevalence of referral to a cardiologist for noninvasive ischemic testing (NIIT) or angiography among outpatients with clinical suspicion of new coronary artery disease (CAD).

designRetrospective records-based cohort study.

settingGroup cardiology referral practice in a tertiary care teaching hospital in Toronto. PATIENTS: Of 1212 new patients referred between Mar. 1, 1991, and Mar. 31, 1993, 339 (232 men and 107 women) had symptoms of CAD and had initiation or modification of antianginal therapy, or were prescribed NIIT or angiography by their cardiologist. OUTCOME MEASURES: Rates of prior NIIT and of subsequent diagnostic procedures for women and men.

resultsWomen were less likely than men to have had prior NIIT (51 [47.7%] v. 162 [69.8%]) (p < 0.001). Women with class I or II angina were less likely than their male counterparts to have had prior NIIT (20/40 [50.0%] v. 84/109 [77.1%]) (p = 0.001); a similar difference was found between women and men with class III or IV angina (5/15 [33.3%] v. 43/60 [71.7%]) (p = 0.006). Men and women with atypical angina were equally likely to have had prior NIIT. Among patients with a positive initial NIIT result a higher proportion of women than of men did not have any further testing done by the cardiologist (53.5% v. 33.7%) (p = 0.02). Cardiologists referred a higher proportion of women (40.2%) than of men (25.4%) for NIIT (p = 0.006) and, for patients with a positive prior NIIT result, were also more likely to refer women (25.7% [9/43]) than men (12.0% [16/133]) for further NIIT (p = 0.04). Women with a positive prior NIIT result were more likely than their male counterparts to have a negative retest result (42.8% v. 18.2%). After adjustment for age, anginal class and overall result of NIIT, women were less likely than men to be referred for angiography (odds ratio 1.4), although the difference was not significant.

conclusionsIn this exploratory study sex-related differences in patterns of use of NIIT by cardiologists seem largely based on differences in testing before referral by family physicians and general internists.