Policy Devolution and Child Health: Evidence from the United States

Abstract

The decentralization of policymaking authority in the United States results in spatial variation in conditions relevant to population health. There is considerable evidence of a divergence in health outcomes between states since the 1970s; internal differences in social policymaking have been characterized as a contributor to this health geography. Documented inequalities predominantly relate to adult health outcomes, such as chronic disease and mortality. Recent research shows that there are larger disparities in longevity between states when individuals are classified by their state of birth rather than state of residence at death. These findings are consistent with life course theories which emphasize the long-term effects of epidemiological exposures during developmental windows, such as before birth, during infancy, and childhood. Since children experience different sources of harm than adults, there is a need to expand consideration of the effects of decentralized policymaking in terms of child health exposures and outcomes. This dissertation provides evidence of the effects of policy devolution on child health in domains of immigration enforcement, environmental regulation, and treatment of water. In the first study I use a comparative interrupted time series design to estimate changes in use of the pediatric emergency department (ED) during the month the Arizona state legislature passed the punitive immigration policy Senate Bill 1070 and data from the Arizona Medicaid enrollment archive to calculate changes in children’s access to public health insurance during this period. Results suggest that Hispanic children in Arizona had reduced access to emergency healthcare and public health insurance following the state’s concurrent passage of that exclusionary policy and freeze on enrollment in the state’s Children’s Health Insurance Program. In the second study I use ED discharge data, information on concentrated hog feeding operations, and spatial inverse probability of treatment weighting to estimate the impact of the contemporary Iowa hog-raising industry on incidence of pediatric acute gastrointestinal illnesses (AGIs). Results indicate that children in rural communities near large hog populations use the ED for AGIs more frequently, with largest differences among children during warm-weather months when hog effluent is applied to agricultural fields, in areas with karst terrain, where private wells are common, and in communities downhill of hog-raising operations. In the third study I provide evidence for the distributional benefits of community water fluoridation (CWF). I use data from the 2019 Oregon Healthy Teens Survey and the causal forest algorithm to estimate the causal effect of CWF in subgroups. Results suggest there are only small average benefits from CWF among respondents in terms of reduced prevalence of cavities, but children with self-identified functional challenges and who experience household- and school-based adversities are less likely to miss school due to toothache when municipalities provide this intervention. Policy devolution creates space for states to pursue different social, economic, and health agendas. Some states tolerate greater harm to children than others. These decisions are structural determinants of child health and origins of many geographic disparities in adult health.