To estimate associations between the length of state-level eviction moratoria enacted in March and April 2020 in the United States and perinatal outcomes. We used data from natality files, 2020-2021 to identify individuals with Medicaid or no insurance who conceived in March-May 2020. The exposure was the number of months exposed to a moratorium (0 (referent, no state-level moratoria), 1-2, 3-4, 5 or more). Outcomes included preterm birth (PTB, < 37 weeks gestation), very preterm birth (VPTB, < 32 weeks gestation), low birthweight (LBW, < 2500 grams), very low birthweight (VLBW, < 1500 grams), primary cesarean, or maternal morbidity. We estimated risk ratios (RRs) using log-binomial regression, including individual, county, and state-level confounders. We conducted several sensitivity analyses to rule out residual state-level confounding including a negative control analysis of 2019 conceptions and difference-in-difference analysis. We included 375,821 births. Following adjustment, having a moratorium in place for 5 or more months was associated with slightly reduced risk of PTB (RR: 0.95, 95% CI: 0.88, 1.02), VPTB (RR: 0.90, 95% CI: 0.8-1.01), LBW (RR: 0.95, 95% CI: 0.9-1.01), and VLBW (RR: 0.91, 95% CI: 0.81-1.02) compared to states without a moratorium. There was no association with cesarean or maternal morbidity. Sensitivity analyses showed that all or most of the observed associations may be explained by residual state-level confounding. State-level eviction moratoria were associated with improved birth outcomes, yet it is likely that all or most of the observed association is due to other policy actions or characteristics of enacting states.
Read full abstract