Factors Associated With Postpartum Uninsurance Among Medicaid-Paid Births.

Importance: To reduce postpartum uninsurance and improve postpartum health, the American Rescue Plan included an option for states to extend pregnancy-related Medicaid from 60 days to 1 year after childbirth. Recent estimates of postpartum uninsurance among Medicaid-paid births would provide information on who would benefit from state adoption of this extension. Objective: To estimate rates of postpartum uninsurance among individuals with Medicaid-paid births. Design Setting and Participants: This cross-sectional study used survey data from the 2015 to 2018 Pregnancy Risk Assessment Monitoring System (PRAMS), a representative sample of births in 43 states and New York City, New York, and included PRAMS participants for whom Medicaid was the primary payer for childbirth. The mean weighted PRAMS response rate was 60.9% for the sites and years included in this study. The data were analyzed from December 2020 to January 2021. Interventions or Exposures: Self-reported postpartum uninsurance measured at the time of the PRAMS survey (mean [interquartile range], 4.2 (3.0-5.0) months after birth). Main Outcomes and Measures: Survey-weighted rates of postpartum uninsurance and 95% CIs overall, by state, by state Medicaid expansion status, and by maternal sociodemographic characteristics. Adjusted odds ratios of the association between maternal characteristics and postpartum uninsurance. Results: We identified 63 370 respondents with a Medicaid-paid birth. Of these, 22 016 (41.1%) were non-Hispanic White individuals, 17 442 (22.0%) non-Hispanic Black individuals, 6808 (13.3%) Spanish-speaking Hispanic individuals, 7000 (13.7%) English-speaking Hispanic individuals, 2410 (4.0%) Asian/Pacific Islander individuals, and 3894 (1.3%) Indigenous individuals. Of these, 41.7% were no longer insured by Medicaid postpartum and 22.0% were uninsured postpartum. The postpartum uninsurance rate was 3 times higher in Medicaid nonexpansion states (36.8%) compared with expansion states (12.8%). Postpartum uninsurance varied significantly across states, ranging from 1.7% in Massachusetts to 56.7% in Texas. There were substantial racial and ethnic disparities in postpartum uninsurance: 54.9% of Hispanic, Spanish-speaking individuals (adjusted odds ratio [aOR], 6.2; 95% CI, 5.5-7.0) and 42.8% of Indigenous respondents reported postpartum uninsurance (aOR, 4.3; 95% CI, 3.8-4.9) compared with 15.2% of non-Hispanic White respondents. The odds of postpartum uninsurance was higher among unmarried people (aOR, 1.3 compared with married people; 95% CI, 1.2-1.4), those 35 years and older (aOR, 1.5 compared with those younger than 20 years; 95% CI, 1.2-1.9) and those with lower levels of education (aOR, 0.8 for more than high school compared with less than high school; 95% CI, 0.7-0.9). Conclusions and Relevance: In this cross-sectional survey study of 43 states, a high proportion of people with Medicaid-paid births were uninsured in the postpartum period, particularly those living in Medicaid nonexpansion states. The study findings suggest that state extensions of pregnancy-related Medicaid eligibility through the first year postpartum could disproportionately benefit Hispanic and Indigenous people, unmarried people, those with lower education levels, and those living in Medicaid nonexpansion states.