Smoking status and mammography among women aged 50-75 in the 2002 behavioral risk factor surveillance system.

BACKGROUND: Since the late 1980's, data show an association between smoking status and mammography. Women smokers reported 12-15% lower rates than non-smokers. This study investigated whether an association persists in a recent national-level database.
METHODS: The sample was women aged 50-75 years from the 2002 Behavioral Risk Factor Surveillance System (N = 52,300). Analyses used two definitions of recent mammography, one based on a one-year interval between exams, the other on a two-year interval. Smoking was classified as current, former, or never. Other covariates included sociodemographic variables, health practices, insurance status and usual source of care.
RESULTS: The rate of mammography for current smokers was 16% lower than for never smokers for the past-year interval, and 14.0% lower for the two-year interval. Multiple logistic regression supported the smoking/mammography association. Other covariates associated with lower mammography on both dependent variables were recent Pap test, health insurance/usual source of care, recent dental visit, seat belt use, marital status and age.
CONCLUSIONS: Reasons for the persistent association between smoking and mammography must be determined. These variables will inform interventions with women who smoke. A key question is whether to intervene on smoking directly or whether mediating variables are sufficient targets of intervention.