The proposed research will use data from eight cross-sectional surveys of adults with mortality follow-up linked to information about the place of residence of the survey respondent to assess the impact of the residential clustering of immigrants and co-ethnics on the empirically identified mortality and morbidity advantage of the foreign born. More specifically, this research will address the issue of whether living in a residential enclave, which is defined as a contiguous set of Census block-groups with statistically identified densities of recent immigrants of particular ethnic groups with language constraints, affects the morbidity and mortality prospects of immigrants. The secondary research questions are whether the neighborhood economic conditions and city level income inequality and residential segregation attenuate or eliminate any favorable effects of living in such an enclave. Data from three independent sources will be used to conduct these analyses in three steps. Individual data on health risk behavior, demographic characteristics, and health outcomes comes from adult respondents to the 1987-1994 National Health Interview Survey (NHIS), an annual nationally representative cross-sectional survey of the non-institutional population of the United States. These data will also be linked to the National Health Interview Survey-Multiple Cause of Death (NHIS-MCD) files for the same eight year period, which contains death certificate information for survey respondents from the NHIS to provide data for the mortality portion of the analysis. These matched records are then linked via geocodes to the STF3-A data from the 1990 Census of Population at the block-group, census tract, and Metropolitan Statistical Area. This data will be used to characterize the residential environment of the survey respondents. Additional tabulated data from the Census Bureau on residential segregation and income inequality in Metropolitan Statistical Areas will be used to augment the data available from the Census STF3-A. The analysis will consist of three steps. First, spatial models will be developed to identity immigrant residential enclaves from the available information at the block-group level. Second, individual mortality and morbidity models will be developed for immigrants to establish the individual level pathways to differential health among immigrant populations. These models will be specified with and without selected health behaviors to assess the intervening role of health behaviors in health outcomes. Finally, the individuals will be placed in their residential contexts and the direct effects of living in a residential enclave and the modifying effects of the neighborhood and city-level economic and residential segregation conditions will be assessed using appropriate multilevel techniques.
