Using geocoding with CHIS data to understand Asian cancer disparities

Background: Few data sources exist that offer robust samples of Asian Americans. Those that do exist may not interview in multiple Asian languages. There is a consequent paucity of research on Asian health topics that require representative samples from often monolingual populations on complex topics, such as the role of culture in breast cancer screening.

Objectives: Using data from the California Health Interview Survey (CHIS), a large telephone survey of California adults, we demonstrated how geo-coded data can be used to examine cultural and structural barriers to cancer screening at the community clinic level.

Methods: We analyzed the 2003 California Health Interview Survey. The survey provides a robust sample of Asian Americans generally and oversamples some Asian American groups. The survey is also conducted in four Asian languages. Further, geo-coded data is available that allowed us to link data on culturally concordant clinics to survey respondents. Bivariate and multivariate analyses were conducted to compare the relative impact of navigation and acculturation on breast and cervical cancer screening, controlling for sociodemographics among both Asian Americans and non-Hispanic whites.

Results: Collectively, health care navigation factors played a stronger role than acculturation factors for clinical breast exam among Filipina and Koreans, mammography among Chinese, and Pap screening among Chinese and Koreans. Specifically, types of insurance and usual source of care were associated with differential rates of screening among most groups studied.

Conclusions. Our findings suggest that navigation factors more directly intervene on health care use, warranting future exploration in surveys and studies.