Essays on the utilization of health and social services among low-income families in the United States

Abstract

Interventions aimed at supporting low-income families of newborns and increasing access to mental health services in both the prenatal and postpartum periods can have positive effects on child health and development. However, the under-use of US social safety net programs and mental health services during the perinatal period is pervasive. The objective of this dissertation is to describe utilization patterns among low-income families during the perinatal period and evaluate determinants associated with service take-up. The first two papers explore Supplemental Security Income (SSI) coverage among likely eligible low birth weight (LBW) infants born to mothers without postsecondary degrees. The third paper evaluates the effect of the Nurse-Family Partnership home-visiting program on maternal utilization of mental health services among a population of Medicaid-eligible, first-time mothers. Summaries of each paper, including the data sources, methods and results are provided below.

Paper 1 estimates SSI coverage among likely eligible LBW infants born to mothers without postsecondary degrees and compares coverage across county-level measures of socioeconomic disadvantage. SSI is a means-tested, federal income support program that can facilitate healthy child development by alleviating financial hardship, reducing family stress, and increasing adults’ capacity to provide responsive caregiving. Using US birth certificate data linked to SSI recipient count data and American Community Survey data for the years 2012 to 2016, we estimate that less than half of likely eligible LBW infants received SSI nationally over the five-year study period, 0.46 [95% CI: 0.44-0.49]. Despite higher levels of need, families living in economically disadvantaged counties were covered by SSI at lower levels compared to the most advantaged counties.

Paper 2 characterizes and explains geographic variation in SSI coverage among likely eligible LBW infants by linking the data from vital records used in Paper 1 to measures of state safety-net generosity. Differences in state safety net generosity are thought to drive geographic variation in SSI coverage by influencing families’ decisions to apply for SSI benefits. Using multilevel Poisson models, we partition the variance in SSI coverage among states and counties and sequentially add state policy and county socioeconomic variables to calculate the proportion of variance explained at each level. States that offered automatic Medicaid enrollment for SSI recipients had higher SSI coverage than states requiring separate Medicaid applications. Of the total geographic variation in SSI coverage, we found most (62%) was attributable to counties, while 38% was attributable to states. Counties with higher shares of Hispanic people and higher unemployment rates had lower levels of SSI coverage.

Paper 3 leverages a large-scale randomized controlled trial of the Nurse-Family Partnership (NFP) home visiting program in South Carolina to assess its effect on maternal utilization of mental health services related to perinatal depression. Home-visiting programs have been identified as alternative models of perinatal care that have the potential to increase the detection and treatment of depression experienced during the perinatal period by providing patient-centered care, frequent contact, and care coordination. Using Medicaid claims data, we observed high rates of depression/anxiety during pregnancy through 60 days postpartum and low rates of treatment in this population of first-time mothers. We found no evidence that NFP increased the detection or treatment of perinatal depression compared to the control group for the overall sample. Pre-specified subgroup analyses revealed that NFP increased any medication or therapy use among non-Hispanic White women, but reduced medication use among Black women.