Previous reports suggest that use of preventive measures, such as screening mammography (SM), differs by ethnicity. It is unclear, however, if this is determined directly by ethnicity or indirectly by related socioeconomic factors. We studied self-reported data from 18,245 women aged 40-49 who participated in the Behavioral Risk Factor Surveillance System telephone survey in 1992 and 1993. Of these, 11,509 (63%) reported having obtained mammography within the preceding 2 years for screening purposes only. Using reports of other preventive healthcare behaviors, education level, socioeconomic status, and healthcare access problems as independent variables, bivariate associations were assessed, and a logistic regression model was developed. Models for each ethnic group were developed, with consistent results. Women who engaged in other preventive health measures, such as Pap smear (odds ratio [OR] 8.99, 95% confidence interval [CI] = 7.6-10.7), cholesterol measurement (OR 2.64, 95% CI = 2.3-3.0), and seatbelt use, were more likely to obtain SM. Women with healthcare access or insurance problems (OR 0. 59, 95% CI = 0.5-0.7) and current smokers (OR 0.71, CI = 0.6-0.8) had a lower likelihood of obtaining SM. Ethnicity, alcohol use, marital status, and education level were not significantly associated with women's reports of SM. Although ethnicity apparently does not influence a woman's likelihood of obtaining SM, access to healthcare and insurance and engaging in other healthy behaviors do. Health policy planners should consider the importance of these related factors when developing preventive health programs for women.