#60 Black and white womenʹs employment status and mortality

PURPOSE: Studies of the impact of womenʹs employment on health have mostly been limited to white women. Since white and black women have different labor force experiences, we studied the association between employment status and mortality in a biracial cohort of middle-age women. METHODS: Participants were black and white women from the Atherosclerosis Risk in Communities (ARIC) Study who reported being a homemaker (n = 1958) or employed (n = 5403) at the baseline exam (1987-89). The outcome was defined as death during a mean follow-up of 10.2 years (n = 457, 6.2%). Proportional hazards regression analysis was used to assess the association between employment status and mortality, with homemakers serving as the referent. Race specific models were adjusted for socio-demographic variables (age, education, income, race) and selected risk factors (SBP, hypertensive medication, diabetes status, physical activity, BMI, HDL and LDL serum cholesterol, smoking status, alcohol use, history of cancer), both including and excluding deaths occurring within the first two years. RESULTS: Mortality was higher among homemakers than among those employed and the absolute magnitude of this difference was greater among black (7.4% employed vs. 15.4% homemakers) than white (3.4% employed vs. 5.6% homemakers) women. After adjusting for socio-demographic factors, the hazard ratio (HR) of death for employed women was 0.53 (95% confidence interval (CI) = 0.39,0.73) for black women and 0.58 (95% CI = 0.44,0.76) for white women. After additional adjustment for risk factors, this association was modestly attenuated (HR = 0.61, 95%CI = 0.43,0.86 for black women and HR = 0.59, 95% CI = 0.45,0.79 for white women). Estimates did not substantially change when analyses were repeated excluding women who died during the first two years. CONCLUSION: In ARIC, employed women were less likely to die than homemakers. Associations were similar across race groups and were not explained by selected biological and behavioral risk factors.