Objective: We assessed the contribution of variation in risk factor prevalence to population-density and county-level variation in breast cancer mortality rates.
Methods: In 1995 we collected risk factor information in a telephone interview of a random digit dialed sample of: (1) 1241 women from counties in the upper and lower tertiles of population density as of 1970 in the Northeast and South of the United States (Design A); (2) 2492 women from counties in the upper and lower tertiles of 1970-1979 breast cancer mortality rates in the four populations from Design A, and; (3) 276 women in Nassau County in New York State. We calculated 1990-94 mortality ratios (MRs) adjusted for breast cancer risk factors.
Results: The high/low population-density fully-adjusted MRs in women > or = 55 years were 1.01 (95% CI 0.9-1.2) and 1.00 (95% CI 0.8-1.2). The fully-adjusted MRs for high versus low mortality counties ranged from 0.95 (95% CI 0.8-1.2) to 1.29 (95% CI 1.0-1.6) in women > or = 55 years.
Conclusions: Differences in risk factor prevalence explained higher rates in high-density versus low-density areas in older women. Modest elevations in the adjusted high/low breast cancer MRs among older women in certain groups of counties may reflect unidentified risk factors but more likely are due to chance.