Health Disparities
The association between hospital and obstetric unit closures on postpartum acute care in rural North Carolina Mekhala Dissanayake* Mekhala Dissanayake Mollie Wood John Jackson Chantel Martin Michele Jonsson-Funk Rachel Urrutia
Background: Hospital and obstetric (OB) unit closures are concentrated in the rural Southern United States, often where marginalized racial/ethnic groups reside and access to maternal healthcare is strained. Objective: Estimate the effect of hospital/OB closures on postpartum acute care (PPAC) among rural North Carolina (NC) Medicaid beneficiaries by county racial composition. Data: NC birth certificate/Medicaid claims, births from 01/01/2014-12/31/2019. Methods: We categorized rural counties as lower (LNHW, <80% White) or higher (HNHW, ≥80% White) Non-Hispanic White, reflecting the racial/geographic stratifications of rural NC. Our outcome was PPAC: any maternal emergency department or inpatient hospital admission up to 6 weeks after birth. We used comparative interrupted time series analyses of births (with a 1-year baseline) to estimate the effect of living in closure counties (any closure) vs. control counties (with hospital, no closure) on the monthly risk of PPAC over 1-year follow up. We stratified analyses by county racial composition and maternal race/ethnicity (Non-Hispanic White (NHW) and Non-Hispanic Black (NHB)). Results: In LNHW counties there was 1 eligible OB/2 hospital closures (N=16,973 births). Immediately after closure, overall effect estimates of PPAC risk were near zero (-0.7 percentage point, 95% confidence interval (CI): -8.6, 7.2). We observed similar risk for NHW births (1.6, 95% CI: -9.2, 12.3), but a 6.1 percentage point decrease (95% CI: -13.3, 1.1) for NHB births. In HNHW counties there was 1 eligible hospital/4 OB closures (N=5,098). Estimates were near zero overall ( -1.0, 95% CI: -4.1, 2.1) and for NHWs (-0.04, 95% CI: -3.5, 3.4), but for NHBs we observed a large but imprecise increased risk (15.2, 95% CI: -39.7, 70.2). Conclusion: The risk of PPAC after closure changed the most among NHB births, though imprecisely estimated due to small sample size. If valid, the effect may result from differential access to healthcare resources.