Poster #66 - Changes in Income Distributions and Racial Disparities in Low Birth Weight

• In Event: 2-163 - Poster Session 08
  In Poster Session: PS 08 Section - Diversity, Equity & Social Justice

Integrative Statement

Low birth weight (LBW ; <2500 g.) is approximately twice as common for black/African American infants relative to whites (Martin, Hamilton, Osterman, Driscoll, & Drake, 2018), contributing to black-white disparities in infant mortality and even adulthood inequities in health and well-being (Braveman & Barclay, 2009). However, geographic differences in the magnitude of the racial gap in LBW are substantial. One likely determinant is the distribution of household income—in terms of average levels and black-white income differences—due to the close link between community financial resources and many health predictors (e.g., educational levels, household stress, health care resources) (Link & Phelan, 1995). Addressing gaps in extant literature, the current study uses longitudinal data to examine changes in county-level median household income and racial income differences as predictors of the black-white LBW gap.

Figure 1, below, provides a timeline of assessments for study variables. National birth records for approximately 24.8 million singleton births to non-Hispanic black or white mothers were utilized to code LBW and maternal risk factors. Maternal risks included sociodemographic (education, nonmarital childbearing, and age) and health factors (smoking during pregnancy, insufficient weight gain, and inadequate prenatal care). LBW and maternal risks were aggregated to the county as black-white absolute differences in prevalence estimates and pooled for three-year periods, spanning from 1992-2014. Median household income was derived from the Small Area Income and Poverty Estimates program (Bell et al., 2007). The racial income gap was computed as the absolute difference in median household income between black and white households—measured via Decennial Censuses and American Community Surveys; weighted averages were used for years when income estimates were not available. Black density (i.e., percent of county residents categorized as non-Hispanic black) and population change from prior period (as percentage) were included as covariates. A total of 732 counties, with an average of 3.8 observations, met inclusion criteria: 20 LBW cases for each racial group; data available for substantive predictors; and at least two periods of data. Hypotheses were tested using county by period fixed effects models.

Model results are shown in Table 1. Findings indicated that a $5,000 increase in county median income was associated with a reduction in the black-white LBW gap of 2.8 births per 1000. This within county effect was equivalent to 1 SD unit change for median income and .12 SD units for the racial LBW gap. Contrary to hypotheses, a decrease in the racial income gap of $5,000 was associated with an increase in the black-white disparity in LBW of 1.3 births per 1000. Adjustments for maternal sociodemographic and health risks did not substantively change estimates for median income or racial income differences (attenuating estimates by ≤ 25%).

Thus, widespread increases to household income may be an important path for reducing racial disparities in LBW. Similar to prior research on the differential health benefits of socioeconomic resources for black and white Americans (Fuller-Rowell, Curtis, Doan, & Coe, 2015), findings point to the complex nature of racial income differences as determinants of racial health disparities.

Authors

• David S. Curtis, University of Utah
  Presenting Author

• Thomas E. Fuller-Rowell, Auburn University
  Non-Presenting Author