Health Services/Policy
The Impact of NY Paid Family Leave on Postpartum A1C after Gestational Diabetes: A controlled interrupted time series analysis Natalie Boychuk* Natalie A. Boychuk Boychuk Boychuk Department of Epidemiology, Columbia University Mailman School of Public Health
Paid family leave (PFL) is linked with improved infant health, breastfeeding, and postpartum healthcare use, but potential benefits for postpartum physical health are underexplored. Hemoglobin A1C may signal future cardiometabolic risk and is mandatorily reported in New York City (NYC), providing a marker of postpartum health. We examined the impact of New York PFL on A1C 4-12 months postpartum in privately insured women with gestational diabetes using a controlled interrupted time series of linked NYC birth, hospital discharge, and A1C registry data, 2014-2019 (n=3,371). Deliveries were classified as pre- (January 2014-December 2018) or post- (January 2018-March 2019) PFL. We estimated policy effects in women who were employed during pregnancy, with women who were unemployed as a control series. We used log binomial, linear, and quantile segmented regression to assess level/slope changes in 1) risk of prediabetic postpartum A1C (≥5.7%), 2) A1C z-score, and 3) median A1C, adjusting for race, education, parity, and nativity. Almost 80% (78.5%) of women were employed. Median A1C was 5.5 (IQR: 0.5). When A1C was binarized, no immediate post- vs. pre-PFL differences or sustained changes by employment were present (post vs. pre, treatment vs. controls, adjusted risk ratio (aRR): 1.00, 95% CI: 0.85, 1.17), post-PFL slope treatment vs. controls aRR: 1.00, 95% CI: 0.96, 1.05). In models of A1C z-score, we found an adjusted post vs. pre difference between treatment and control groups of -1.01 standard deviations (95% CI: -1.96, -0.07); however, a positive post-policy difference in slope by treatment was present (β: 0.28, 95% CI: 0.03, 0.52). In quantile regression, we similarly found an immediate -0.09 percentage point reduction in median A1C post- vs. pre-PFL in treatment vs. controls (95% CI: -0.25, 0.02) but no sustained changes by employment. Results suggest that PFL may have improved postpartum A1C among privately insured women immediately after policy implementation.