Nelson KM, VHA Puget Sound Healthcare System; Schwartz G, VHA Office of Analytics and Business Intelligence; Hernandez SE, VHA Puget Sound Healthcare System; Simonetti JA, VHA Puget Sound Healthcare System; Curtis I, VHA Puget Sound Healthcare System; Fihn SD, VHA Office of Analytics and Business Intelligence;
Objectives:
There is limited evidence about the association of neighborhood environment and outcomes among Veterans. We sought to determine the association of neighborhood factors with all-cause mortality rates among Veterans.
Methods:
We used data from the VA Corporate Data Warehouse (CDW) to obtain individual data on vital status, clinical co-morbidity and demographic characteristics. Census Tract level socio-economic status (SES) data was obtained from the US Census Bureau and linked to individuals by residence. Census tract level SES was characterized by a previously validated index based on 6 components: 1) median household income and percentage of: 2) adults > 25 y old with less than a high school education; 3) unemployed males; 4) households with income below the poverty line; 5) households receiving public assistance; 6) households with children that are headed by a female. The SES index was calculated as the sum of component z-scores standardized to range from 0 to 1 with increasing SES. Logistic regression was used to model the association of the tract level SES index with all-cause mortality in 2012 while controlling for individual level co-morbidity and other demographic characteristics.
Results:
We analyzed data on all Veterans alive on January 1, 2011 for which vital status, demographic and SES data were available (n = 4,814,631). Tract level SES had a large effect on the likelihood of mortality after controlling for individual co-morbidity and demographics. Individuals living in tracts with an SES Index in the 1st decile (lowest SES) were 30 percent (OR 1.3, 95 percent CI: 1.26, 1.34) more likely to die in 2012 than individuals living in tracts in the 10th decile (highest SES). The likelihood of mortality decreased with increasing decile of SES status. Mortality for individuals living in the 9th decile was 4 percent more likely (95 percent CI: 1.012, 1.066) compared to the highest decile.
Implications:
Neighborhood socioeconomic environment is strongly associated with all-cause mortality among Veterans independent of individual-level comorbidities and socioeconomic characteristics.
Impacts:
Census-based measures can be used to identify disadvantaged areas and may be useful to target community-based interventions. Strategies to improve population health will ultimately need to incorporate information about neighborhood characteristics to be successful.