Abstract B17: Racial/ethnic variation in colon cancer surveillance among individuals with a family history of colon cancer

Abstract

Abstract Background: Family history of colon cancer is associated with increased participation in colon cancer surveillance. The consistency of this association has not been systematically explored within racial/ethnic groups. Objective: To explore the association between family history of colon cancer and colon cancer surveillance in a large population with considerable racial/ethnic diversity. Methods: We utilized data from the 2009 California Health Interview Survey, a health-related telephone questionnaire administered to a two-stage geographically stratified random sample of non-institutionalized California residents. Individuals aged ≥50 years without a personal history of colon cancer were eligible for our analysis. Family history of colon cancer was defined as having reported ≥1 first-degree relative diagnosed with colon cancer. Colon cancer surveillance was defined as self-reported utilization of colonoscopy within the past 10 years. Binary logistic regression was used to estimate odds ratios (ORs) and 95% confidence limits (CL) for the association between family history of colon cancer (compared to no family history) and colon cancer surveillance overall (adjusted for age and race/ethnicity) and by race/ethnicity (adjusted for age) while accounting for the complex survey design and weighted to the California population. Results: The unweighted study population consisted of 30,857 individuals (23,106 [75.0%] non-Hispanic Whites, 3,219 [10.4%] Hispanics, 2.508 [8.1%] Asians, 1,173 [3.8%] non-Hispanic Blacks, and 851 [2.8%] Other), of whom 2,582 (8.4%) reported ≥1 first-degree relative diagnosed with colon cancer. Colonoscopy within the past 10 years was reported by 17,216 (56%) of individuals in this sample. Overall, individuals with a family history of colon cancer had 2.6 times the odds of having a colonoscopy within the past 10 years compared to individuals without a family history (OR=2.6, 95% CI: 2.1, 3.1). An overall test of homogeneity revealed statistical interaction between family history of colon cancer and race/ethnicity (P=0.001). Asians and non-Hispanic Whites with a family history of colon cancer had higher odds of having a colonoscopy within the past 10 years (Asians: OR=6.2, 95% CI: 3.3, 11.6; non-Hispanic Whites: OR=2.9, 95% CI: 2.9, 3.4). The odds of having a colonoscopy within the past 10 years among non-Hispanic Blacks and Hispanics with a family history of colon cancer were lower than other racial/ethnic groups (non-Hispanic Blacks: OR=1.6, 95% CI: 0.7, 3.7; Hispanics: OR=1.3, 95% CI: 0.7, 2.4). Conclusions: Colon cancer surveillance in individuals with a first-degree relative with colon cancer varies by race/ethnicity. Although our results reflect the conditions in California (e.g. racial/ethnic distribution and 92% insured) and are potentially sensitive to misclassification of family history of colon cancer, our findings raise awareness of a higher-risk subset of non-Hispanic Blacks and Hispanics for whom surveillance is suboptimal and should be encouraged. Citation Information: Cancer Prev Res 2011;4(10 Suppl):B17.