Abstract 2807: The relationship between age and efficiency of cervical cancer screening using population-based surveillance

Abstract

Abstract Data for in situ cervical carcinomas are not routinely collected by U.S. tumor registries. Some experts suggest that comparing preinvasive and invasive cervical cancers (ICC) rates may represent our best population-based appraisal of screening strategies. From 1991-1997, a total of 5,921 women diagnosed with in situ or ICC were reported to the Markey Cancer Registry in Lexington, Kentucky, a population-based registry. Maximum-likelihood Poisson regression analyses were used to estimate the incidence rate of in situ and ICC (per 100,000), summarized over age-and race-specific groups for each of seven years, 1991 to 1997. U.S. Census age and race data for 1990 were used to estimate denominators for two race groups: African American (AA) and all others (white). Incidence rates and rate ratios are reported. A total of 4,290 and 1,631 were diagnosed with in situ and ICC in the study period. While more than 80% of cases among women <30 years of age were diagnosed as in situ lesions, only about half of cases diagnosed among 50-54 year-olds were identified at a preinvasive stage. For women >65 years, only 42% (+4%) of tumors were diagnosed as in situ lesions. Specifically, among 20-39 year olds, AA were less likely than white women to be diagnosed with in situ carcinoma (p<0.05), while after age 60, they were more likely than whites to be similarly diagnosed (p<0.05). White women 20-49 years were 1.4-23.8 times more likely to be diagnosed with in situ carcinoma than ICC; however, AA ages 25-34 were 1.4-1.6 times more likely to show in situ disease. Whites >55 were less likely to be diagnosed with in situ disease than ICC. While year of diagnosis was significant in the analyses, relationships between age and diagnostic stage persisted across each of 7 years examined. These data suggest that Pap test most efficiently identifies treatable, high-grade precancerous lesions among women younger than 50 years. HPV infection prevalence varies by age, but not race, while ICC disproportionately affects women of color, especially as they age. Several factors may explain these findings. Screening may be performed less often overall or more inconsistently among older women and among AA. Also, if screening in the population decreases with advancing age, symptomatic older women with higher likelihood of malignancy will be over-represented in the data. Directly tailoring screening programs to recruit and retain non- and under-screened women and women with medical record evidence of poorly triaged Pap test abnormalities may improve detection of cervical cancers at a preinvasive stage, where there is a very high likelihood of survival. We believe that these findings are consistent with those of others, and together they suggest our best public health strategy may be to actively recruit middle-age and older women into cervical cancer screening programs who have undergone Pap test screening fewer than three times in a ten year period. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 2807.