Abstract B13: Widening socioeconomic disparities in cervical cancer mortality among women in 26 states: 1993–2007

Abstract

Abstract Background: Despite declines in cervical cancer mortality in the U.S. due to widespread Pap testing, mortality disparities persist. Previous studies were limited to area-level determinants of socioeconomic status (SES). We examined recent temporal trends in cervical cancer mortality by individual levels of educational attainment as a proxy for SES. Methods: Data from the National Vital Statistics System were used to calculate age-standardized cervical cancer mortality rates for women aged 25–64 years (1993–2007) by individual levels of education: <12 years (high-school diploma or less), 13–15 years (some college), and 16+ years (college degree or more) and race/ethnicity (non-Hispanic white, non-Hispanic black, and Hispanic). The following 26 states were included in the analysis as they consistently reported education status: Alabama, Alaska, Arizona, Arkansas, Colorado, Hawaii, Illinois, Indiana, Iowa, Kentucky, Louisiana, Maine, Maryland, Massachusetts, Minnesota, Mississippi, Missouri, Nevada, North Carolina, North Dakota, Pennsylvania, Tennessee, Vermont, Virginia, West Virginia, and Wisconsin. Current Population Survey data were used as population denominators and rates presented are per 100,000 population. Log-linear regression was used to assess changes in mortality rates during 1993–2007. Rate ratios (RRs) and corresponding 95% confidence intervals (CIs) were used to assess education disparities for those with <12 years of education versus those with 16+ years of education during 1993–1995 and 2005–2007. The fraction of avoidable cervical cancer deaths in 2007 (had all women experienced the death rates of the reference group, i.e., the most educated non-Hispanic white women) was calculated by applying mortality rates from the reference group to the population of women aged 25–64 years in the study. Results: A total 6,613 cervical cancer deaths occurred during 1993–2007. Mortality rates declined from 4.18 during 1993–1995 to 3.01 during 2005–2007 (−2.9% per year, P<0.05). Declines were greatest for women with 16+ years of education, regardless of race/ethnicity (−3.9% per year, P<0.05). As a result, the education disparity as measured by the RR, widened between 1993–1995 and 2005–2007, from 3.07 (95%CI 2.40–3.93) to 4.41 (95%CI 3.51–5.55) among non-Hispanic whites, and from 3.77 (95%CI 2.03–7.00) to 5.55 (95%CI 3.07–10.03) among non-Hispanic blacks. For Hispanics the education disparity RR during 2005–2007 was 2.54 (95%CI 0.76–8.48). If the entire population experienced the same mortality as non-Hispanic white women with the highest levels of education, about 70% cervical cancer deaths could have been avoided in 2007. Conclusions: Overall cervical cancer mortality declined among women aged 25–64 during 1993–2007; however, the rate of decline was much slower for women with lower levels of educational attainment, leading to widening of SES disparities in cervical cancer mortality rates. We note that many cervical cancer deaths could be avoided by addressing SES disparities, and that additional research is needed to assess individual determinants of these disparities in cervical cancer mortality as well as interventions to reduce them. Citation Information: Cancer Prev Res 2011;4(10 Suppl):B13.