CN Mobile Logo

Search form


Breast Cancer Screening and Socioeconomic Status: 35 Metropolitan Areas, 2000 and 2002

Breast Cancer Screening and Socioeconomic Status: 35 Metropolitan Areas, 2000 and 2002

Studies have suggested that women with low incomes residing in metropolitan areas might be less likely to be screened for breast cancer than more affluent women residing in the same areas.[1,2] However, few studies have examined the associations between breast cancer screening and both individual and area-based measures of socioeconomic status among women in metropolitan areas.[3,4] To examine these associations, the Centers for Disease Control (CDC) analyzed the percentage of women who had a mammogram by using individual data (ie, household income and education level) from the 2000 and 2002 Behavioral Risk Factor Surveillance System (BRFSS) surveys and area-based data (ie, percentages classified as living in poverty [annual family income below the federally defined poverty line] or at a low education level [less than a high school education]) from the 2000 US Census.

This report summarizes the results of those analyses, which suggested that, among women in 35 metropolitan statistical areas (MSAs),* those with annual household incomes < $15,000 were less likely to have had a mammogram than more affluent women (especially in areas where a greater proportion of women were affluent) and those without a high school education were less likely to have had a mammogram than women with more education (especially in areas where a greater proportion of women had higher education levels). Studies are needed to determine how to increase the percentage of women having mammograms among women in low-income and low-education -populations.

The Behavioral Risk Factor Surveillance System is a state-based, random digit-dialed telephone survey of the noninstitutionalized, US civilian population aged ≥ 18 years.[5] During 2000 and 2002, BRFSS interviews were conducted with 251,269 women. Data from 2000 and 2002 were used to provide the best match between individual-level information and MSA data from the 2000 US Census. Weights were used to adjust for differences in probability of selection, nonresponse, and noncoverage. The Council of American Survey Research Organizations estimated median response rates among reporting states in 2000 and 2002 were 48.9% and 58.3%, respectively.[5] Questions were asked regarding general health status, demographic and socioeconomic characteristics, and breast cancer screenings. Female respondents were asked, "Have you ever had a mammogram?" Those who said "yes" were then asked, "How long has it been since you had your last mammogram?"

MSA Classification

Self-reported county of residence was used to classify respondents as residents of MSAs using Office of Management and Budget definitions for MSAs.[6] To reduce the heterogeneity of the MSAs and ensure a sufficient number of respondents in each, only BRFSS respondents who resided in MSAs with populations of ≥ 1.5 million in 2000 were included in this analysis. The 35 MSAs included in this analysis ranged in population from 1,500,741 to 18,323,002. Analy ses were limited to 38,117 women aged ≥ 40 years with no missing information about recent mammography. Area-based data regarding SES (ie, percentage of residents in an MSA classified as living in poverty or having a low education level) were obtained from the 2000 US Census and categorized using previously described cutpoints.[7] Percentage of residents living below the poverty level was based on the 1999 federal definition of a poverty area. Rates for having a mammogram during the preceding 2 years were calculated with combined data from 2000 and 2002.

In examining the bivariate associations between screening and both demographic and health factors, the levels of statistical significance were obtained using Pearson's chi-square tests; 95% confidence intervals (CIs) and P values were calculated. Multivariate analyses of the associations between individual-level and area-based data and breast cancer screening were conducted using logistic regression analyses that employed the following variables: year (2000 vs 2002), age, race, ethnicity, marital status, health insurance, and physician checkup during the preceding year.


By clicking Accept, you agree to become a member of the UBM Medica Community.