Social
Highway displacement and mortality: impacts of the “Tucson Limited Access Highway” Rae Anne Martinez* Rae Anne Martinez Martinez Martinez University of Colorado Denver
In the post WWII era, the US increasingly invested in highways as an essential piece of infrastructure. Often, these highways were built through cities, leading to displacement of residents whose homes were demolished and potentially disrupting social and material resources needed for health. Here, we examine the construction of the “Tucson Limited Access Highway” in AZ to conduct an individual-level investigation of the relationship between displacement and mortality (age at death), as well as to investigate the change in neighborhood quality following displacement. Using full count 1940 census and other historical sources, we identified individuals displaced by highway construction and created a propensity score-matched comparison population. For PS-matching, logistic regression was used to estimate the odds of displacement based on Hispanic surname, race, education, age, nativity, individual income/wages, and home ownership. Matches were made using the nearest neighbor method without replacement until a sample of 3:1 (unexposed:exposed) was generated (N=1960). We then used extant linkage files from the IPUMS Multigenerational Longitudinal Panel to link individuals from the full-count 1940 census to the Social Security Administration Numident file. However, individuals who are Hispanic, immigrant, and women (key characteristics of our study population) are less likely to have been linked between data sources. Thus, we developed a standardized, two-step process for collecting mortality data for those not linked, leveraging publicly available death certificates and gravesite/cemetery data. At present, we have completed 95% of the mortality data collection. Once done, we will conduct descriptive analysis to assess differences in demographic characteristics between groups and use linear regression to estimate age at death by displacement status. For displaced individuals, we will also descriptively examine change in neighborhood characteristics and distance of move.
