| Summary: | Background The field of social epidemiology has grappled with the measurement of macrosocial, upstream factors and the assessment of their impact on health. Economic insecurity is one such determinant of health that has been understudied in the public health literature. Objectives The goal of this dissertation was to develop and validate a novel determinant of health, economic insecurity, and assess its association with mortality, tobacco smoking, and heavy and binge drinking. Methods In aim 1, county-level indicators of economic insecurity were drawn from the US Census and Federal Reserve Bank for seven states in the north-Atlantic and Midwest regions of the US in the year 2000 based on our theoretical framework. We used confirmatory factor analysis to assess our measurement model and evaluate model fit. We validated this model using county-level indicators of socioeconomic status, social disruption, and mortality outcomes. In aim 2, we expanded this measurement model to all counties in all states in the US for the years 2000 and 2010. We then estimated the association between the change in economic insecurity between 2000 and 2010 and the rate of change in mortality due to suicide, drug and alcohol poisoning, and chronic liver disease and in midlife mortality (ages 45-54) from 2001-05 to 2011-15, using linear regression models. Finally, in Aim 3 we test the association between change in economic insecurity and trends in three health behaviors from 2002 to 2012: cigarette smoking, binge drinking, and heavy drinking using ecological longitudinal models. Results The five indicators of percent unemployed, percent not in labor force, percent employed in the service sector, percent of income spent on rent, and percent with subprime credit rating (credit score below 660) produced a measurement model of county economic insecurity with adequate fit. Counties with elevated economic insecurity in either or both 2000 and 2010 had larger increases in suicide, poisoning, and chronic liver disease mortality than counties ...
|
|---|