Project 1: Hispanic Neighborhoods and Cancer Risks and Outcomes
Current Status
Project 1 involves secondary analysis of geo-referenced data to study multi-level linkages because neighborhood ethnic and socio-economic characteristics and health risks, health outcomes, and mortality. The project has a specific focus on the Hispanic population, in comparison to the non-Hispanic white and non-Hispanic black population.
Analysis of geocoded mortality data. In the third year of project I, we continued analysis of mortality data for the states of Texas and California that has been geo-coded to the Census tract level. A manuscript using these data was accepted for publication by the American Journal of Public Health in January 2006.This manuscript examines mortality rates in California for immigrant and U.S.-born Hispanics and non-Hispanic Whites, with attention to validating and correcting vital rates given errors in ascertainment of Hispanic ethnicity. We have submitted a second manuscript to this journal that is currently under review. This paper examines cause-specific mortality patterns comparing Hispanic immigrants and natives at younger ages. We believe that together will have an important impact on understanding of Hispanic mortality differentials. Specifically, they 1) validate vital registration data as a source of information about Hispanic mortality patterns; 2) show that lower Hispanic mortality rates are entirely attributable to lower mortality among foreign-born rather than U.S.-born Hispanics; 3) revise understanding of Hispanic mortality at younger ages, showing that immigrant Hispanics have lower mortality from most specific causes at these ages, while U.S.-born Hispanics do not, and 4) identify cause-specific patterns that contribute to mortality differentials.
We have also analyzed these geocoded mortality data with respect to area-contextual correlates of mortality from specific extrinsic causes, including homicide, suicide, substance use, and accidents. This analysis compares African Americans, non-Hispanic Whites, and Hispanics. Two abstracts based on this work have been accepted for presentation at the November 2005 meetings of the American Public Health Association meetings in Boston, and manuscripts are in preparation.
Analysis of geocoded National Health and Nutrition Examination III. The purpose of this analysis is to identify multi-level correlates of neighborhood characteristics on cancer and other health risks and risk biomarkers. We have submitted two manuscripts from this analysis for peer review, examining contextual variation in serum carotenoids and contextual variation in health risk behaviors. We are completing a manuscript that examines contextual variation in social support. Another current analysis examines nutritional and behavioral risks for prostate cancer. We have also completed a manuscript that examines ethnic variation in determinants of objective assessments of physical function among older subjects.
Analysis of patterns of cancer treatment and survival, SEER
We have published several manuscripts that analyze socio-economic and ethnic disparities in cancer treatment or outcomes using the SEER or SEER-Medicare database, including a study of physician-based variation in use of androgen deprivation therapy (forthcoming, Journal of the National Cancer Institute), and determinants of risk of :”androgen deprivation syndrome” (Archives Int Med.) Manuscripts that examine small area socio-economic status differences in melanoma survival and variation in mammography-screening quality are under review.
Other analyses of ethnic disparities
Center researchers have also published a number of papers focusing on mechanisms producing ethnic disparities among Mexican Americans, and other Hispanics. A particular focus of our research has been the comparison of patterns of health in older Mexican Americans in the United States, compared to older Latin Americans in Mexico and Central/South American countries. These papers have used data from the Hispanic Established Population for Epidemiological Studies of the Elderly in comparison to component studies of the Health, Well-being and Aging in Latin America and the Caribbean health surveys. These population-based surveys use related instruments to examine aging in health in 8 Latin American national populations. This work emerges in part through our recent (2005) designation as a collaborating Center of the Pan American Health Organization. Manuscripts published, submitted, or in preparation are listed below.
In Year 2 of the project (June 2005), the project leader co-organized a one-day meeting to discuss Hispanic mortality issues at the University of Maryland Population Research Center. This meeting was attended by representatives from a number of academic institutions, as well as the CDC, NCI, NHLBI, and Census Bureau. Out of this conference we have developed a collaboration with researchers at the CDC mortality statistics division (Elizabeth Arias) and NHLBI (Paul Sorlie, William Schauman) to estimate revised life tables for racial/Hispanic-origin populations based on rates of assessments of the accuracy of report of race and Hispanic origin on the death certificate. A manuscript is in preparation.
An abstract studying census tract correlates of disability for Hispanics using data from Census 2000 will be presented at the International Conference on Urban Health in Amsterdam in October 2006.
We have prepared a manuscript analyzing contextual correlates of SIDS deaths using linked infant-birth death files geocoded to the county level.
