Association Between the New York State Coronavirus Disease 2019 (COVID-19) Lockdown and Rates of Preterm Birth and...

To estimate the association between the New York State coronavirus disease 2019 (COVID-19) lockdown and preterm birth rates among individuals with Medicaid insurance overall and by onset of delivery, and to investigate contemporaneous changes in stillbirth rates.

We used New York State Medicaid claims to conduct a quasi-experimental study of 2018-2020 hospital deliveries of liveborn neonates. We estimated the effect of the state's lockdown policy (New York State on PAUSE) on preterm birth by comparing changes in preterm birth rates before (January 1-March 21, 2020) and during the lockdown (March 22-June 8, 2020) with changes before and during the same dates in the 2 years earlier using difference-in-difference models. We stratified models into spontaneous and medically indicated preterm births. As a secondary outcome, we estimated the effect of the lockdown on stillbirth rates per 100 live births.

There were 38,272 hospital deliveries to individuals with Medicaid insurance from January 1 to June 8, 2020, and 81,054 on the same dates in 2018 and 2019. From January to March 21, 2020, the preterm birth rate was 10.0%; roughly half were spontaneous. Using difference-in-difference models, New York State on PAUSE was associated with a nearly 1 percentage point decline (-0.93, 95% CI, -1.64 to -0.21) in the preterm birth rate. The overall difference was larger when the time period was limited to the first half of the lockdown (-1.07, 95% CI, -1.95 to -0.20). Rates of both spontaneous and medically indicated preterm birth decreased, but only the change in the rate of medically indicated preterm births was statistically significant (-0.58, 95% CI, -1.08 to -0.08). New York State on PAUSE was not associated with any change in stillbirth rates.

The New York State COVID-19 lockdown was associated with a modest decline in preterm birth rates, particularly medically indicated preterm birth; there was no change in stillbirth rates. This association was particularly strong in the early part of the pandemic and was slightly more weighted toward medically indicated preterm births.