Diabetes is a greater risk factor for ischemic heart disease (IHD) in women than in men (1,2). In the U.S., IHD-related mortality has declined among men with and without diabetes (3). Among women, a decrease in IHDrelated mortality has been observed only for those without diabetes (3). This difference may be attributable to biological (4) and behavioral factors (5) or possibly differences in the quality of health care received (3). We investigated whether there were differences between men and women regarding the quality of health care related to IHD prevention in a population-based cohort of patients with diabetes, aged 20 -80 years and sampled from 10 managed care health plans and 68 provider groups in the U.S. (6).
RESEARCH DESIGN ANDMETHODS -Participants were surveyed using a standardized computerassisted telephone interview or selfadministered written instrument. Of contacted eligible people, 91% responded to the survey. We examined data from the participants for whom medical records were available to document diabetes care. Interrater reliability () for the main quality measures derived from medical record data ranged from 0.85 to 0.92. The quality of diabetes care related to IHD prevention received by patients during a 12-month period was measured by the frequency of selected process of care measurements: 1) current use of aspirin, lipid-lowering medications, and antihypertensive medications (documented by review of the medical records); 2) received recommendation to take aspirin to lower cardiovascular disease (CVD) risk (among those not using aspirin; ascertained by self-report); and 3) received lipid profile testing (among those not using lipid-lowering medications), urine microalbumin/protein testing (among those not using antihypertensive medications), and HbA 1c testing (documented by medical record review). History of CVD was defined according to selfreported myocardial infarction, stroke, coronary artery bypass, or angioplasty. Hierarchical logistic regression models with random intercepts for health plan, to account for the multilevel study design (health plan, provider group, and patient levels), were used to estimate the predicted probability of receiving each process of care measure after adjusting for the proportion of men in the health plans. We then calculated the risk difference (and 95% CI) between men and women regarding these predicted probabilities.RESULTS -There were 1,302 women and 1,564 men with a CVD history and 3,385 women and 2,506 men without a CVD history. Because women had a history of CVD (27.8%) less often than men (38.4%), analyses were stratified by CVD history. Among patients with a CVD history and patients without a CVD history, as compared with men, women were more likely to be aged Ͼ65 years (55.5 vs. 51.6% and 36.5 vs. 31.1%, respectively), to be from U.S. minority racial/ethnic groups (56.6 vs. 50.9% and 61.9 vs. 57.7%), to report less than high school education (33.4 vs. 25.0% and 24.4 vs. 17.6%), to have a diabetes duration Ն10 years (59.9 vs. 52.2% and 44.8 vs. 38...