Background: Colorectal cancer screening (CRCS) rates vary across subgroups with Hispanic Americans and uninsured populations having among the lowest levels. Recent reports also suggest that screening disparities among Hispanics persist, with a slower decline in death rates among Hispanics compared with non-Hispanic whites.

Objective: To increase understanding of the barriers to CRCS in low income Hispanics, this cross-sectional study examines cultural (language and nativity), structural (financial and availability of health care), and sociodemographic factors associated with CRCS among Hispanics residing on the Texas-Mexico border.

Methods: Five hundred and forty-four people meeting eligibility criteria (50 years or older, self-identified as Hispanic or Latino, and no history of cancer) were randomly recruited from colonias (unincorporated, un-zoned, semi-rural communities) in three counties along the Texas-Mexico border. Univariate and multivariate analyses using 2 tests were used to examine associations of the dependent variable, CRCS (FOBT, colonoscopy, sigmoidoscopy and barium enema), and independent variables (i.e., gender, age, birth place combined with years in the U.S., primary language spoken, household income, highest education, marital status, health insurance coverage, travel time to health care provider, and having a place for usual health care). Odds ratios and 95% confidence ratios are reported.

Results: Overall, 34% of participants reported receiving any kind of CRCS. Seventy-three percent of the study participants were women and 27% were men. Mean age of participants was 63 years. Almost all (98%) were Mexican American: 77% were born in Mexico, 71% of whom have lived in the US over 20 years. Roughly 14% reported speaking English well. Only 9% were high school graduates, 72% had an annual household income of less than $10,000, and almost half had no form of health insurance coverage. The majority (73%) reported a usual place to receive health care and 89% traveled less than one hour to this provider. Factors significantly associated with any CRCS in univariate analyses included a combination variable representing nativity (birth place combined with years in the U.S.; P=0.029), health insurance (P<0.000), and usual place of care (P=0.0001). In the final multivariate model these factors were not significant.

Conclusion: Structural factors such as nativity, health insurance and having access to a usual source of health care, were significant independent risk factors for not being screened for CRC in a minority, low income Hispanic population. However, these factors were no longer significant in adjusted multivariate models. Further research is needed to understand the mediating structural and cultural barriers impacting colorectal cancer screening in this population. Findings from this research have the potential to inform policy aimed at decreasing structural barriers to improve CRCS in this population.

Second AACR International Conference on the Science of Cancer Health Disparities— Feb 3–6, 2009; Carefree, AZ