Medicaid Expansion Improved Perinatal Insurance Continuity For Low-Income Women

Continuity of insurance coverage is a prerequisite for continuous access to high-quality health care before, during, and after pregnancy for low-income women.¹ Insurance churn, or moving between different insurance plans or between insurance and uninsurance, is common during the perinatal period because of changes in employment, income, marital status, and Medicaid eligibility that commonly accompany pregnancy and childbirth. Before the Affordable Care Act (ACA) was implemented, an analysis of data from the Medicaid Expenditure Panel Survey, 2005–13, found that half of women who were uninsured nine months before delivery had acquired Medicaid coverage by the month of delivery, but 55 percent of women with Medicaid or Children’s Health Insurance Program coverage at delivery experienced a coverage gap within six months postpartum.² A study of data from the 2009 Pregnancy Risk Surveillance and Monitoring System (PRAMS) found that 30 percent of women who had a live birth experienced changes in health insurance coverage from the month before pregnancy to the time of delivery.³

By expanding Medicaid coverage to all low-income adults, regardless of their parental or pregnancy status, the ACA Medicaid expansions had the potential to significantly reduce perinatal insurance churn. Studies have found that the expansions were associated with reduced uninsurance and improved access to care among reproductive-age women and increased Medicaid coverage in the month before conception.^4,5 A descriptive analysis of perinatal insurance after the ACA was implemented found lower rates of uninsurance both preconception and postpartum in Medicaid expansion states compared with nonexpansion states.⁶ However, no study has examined the policy’s impact on the stability of insurance from before to after childbirth. Thus, the objective of this study was to estimate the association between the ACA Medicaid expansions and perinatal insurance churn among low-income women.

Study Data And Methods

Data Source

We used 2012–17 survey data from PRAMS. The Centers for Disease Control and Prevention (CDC) Division of Reproductive Health administers PRAMS in collaboration with state and city health departments.⁷ From birth certificate data, participating states select a representative sample of all women who delivered a live-born infant. The survey collects data via a standardized mail and telephone survey of recently postpartum women, including demographic characteristics, insurance status, health care use, and health outcomes before, during, and after pregnancy.

Sample

We included low-income women, defined as those with household incomes less than 138 percent of the federal poverty level, the income eligibility threshold for low-income adults under the ACA Medicaid expansion. This income eligibility threshold ranged from $20,879 for a family of two in 2012 to $22,411 for a family of two in 2017.⁸ The PRAMS household income variable is categorical, with $5,000–$15,000 increments below $50,000. To conservatively estimate poverty status, we assumed that a woman’s income was the maximum value of the specified income band and then converted this income to a percentage of the federal poverty level based on year, state, and household size.

We restricted the sample to women who resided in states with continuous participation in PRAMS during the study period to ensure that changes in the composition of states in the sample did not bias our estimates of changes in the outcomes over time. This resulted in the inclusion of fourteen jurisdictions that expanded Medicaid under the ACA during the study period (Alaska, Colorado, Delaware, Illinois, Massachusetts, Maryland, Michigan, New Jersey, New Mexico, New York City, Pennsylvania, Vermont, Washington, and West Virginia) and six that did not (Maine, Missouri, Oklahoma, Utah, Wisconsin, and Wyoming). We further limited the sample to respondents with complete insurance information, which was required for the construction of our primary outcomes (92.5 percent of the total sample).

Variables

We used PRAMS insurance status variables at three time points: preconception, measured as insurance held in the month before conception; delivery, measured as the primary payer for the childbirth episode; and postpartum, measured as insurance held at the time of the postpartum survey. Nearly all women (97.2 percent) completed the survey three or more months after childbirth, which is after pregnancy-related Medicaid eligibility ends. Preconception and postpartum insurance variables are self-reported in PRAMS. For delivery insurance, the standard PRAMS variable is the primary payer for delivery recorded on the birth certificate by the delivery site (for example, a hospital). This variable is used by all states (and was complete for all women in sixteen of the twenty jurisdictions and 96.7 percent of women in our sample). For the remaining 3.3 percent of women without payment information from the birth certificate, we used self-reported delivery insurance (partially collected in four states: Colorado, Illinois, Maine, New Jersey). The concordance between self-reported and birth certificate–based delivery insurance was moderately high: For the 11 percent of women in our sample with both birth certificate and self-reported responses for delivery, 85.6 percent had a concordant response. Seventy-three percent of discordant responses were between Medicaid and private insurance.

We followed methods previously used by the CDC to hierarchically characterize insurance coverage at each time point into one of three categories: Medicaid, private, or uninsured.³ The Medicaid category included women who reported enrollment in Medicaid or a state-named Medicaid program. The private insurance category included women who reported private coverage alone or in combination with Medicaid and women who reported TRICARE or other military insurance. The uninsured category included women who reported no insurance. Consistent with the US census,⁹ other national surveys,¹⁰ and previous analyses of PRAMS data,³ women who reported only Indian Health Service coverage were classified as uninsured. The only exception was Alaska, where the Indian Health Service response option on PRAMS included other state-specific programs and thus was classified as Medicaid.³

We then generated four measures of perinatal insurance churn between these three time points: continuous insurance, defined as insured with no change in insurance status; Medicaid–private churn, defined as any switching between Medicaid and private coverage; insured–uninsured churn, defined as any switching between any type of insurance (Medicaid or private) and uninsurance; and continuous uninsurance, defined as uninsured with no change in insurance status.

Study Design

We used a quasi-experimental difference-in-difference design to assess changes in perinatal insurance churn before and after Medicaid expansion in states that expanded Medicaid relative to those that did not. The prepolicy period was defined as the years before the year of Medicaid expansion implementation (2012–13 for most states). We chose 2012 as the start of the study period, as it was the first year when postpartum insurance status was available in the PRAMS data. The transition period was defined as the first year of Medicaid expansion implementation (2014). The postpolicy period was defined as the years after the transition period (2015–17 for most states). Two states implemented the ACA Medicaid expansion after January 1, 2014: Pennsylvania (January 1, 2015) and Alaska (September 1, 2015). For Pennsylvania and Alaska, we considered the transition period to be 2015 and 2016, respectively.

Statistical Analysis

For each outcome, we estimated linear probability models including state and year fixed effects, as well as interaction terms for state expansion status and whether a birth occurred in the transition or postpolicy period. The primary coefficient of interest was the interaction between state expansion status and the postpolicy period, representing the pre-post change in the outcome in expansion states relative to nonexpansion states. Adjusted models also included the following individual-level covariates: timing of postpartum survey, maternal age, race/ethnicity, language of survey completion, education level, marital status, parity, and comorbid maternal conditions. Standard errors were clustered at the state level to account for the state-level implementation of the policy.

We conducted stratified subgroup analyses by parity (multiparous versus nulliparous) and maternal chronic conditions (self-reported prepregnancy diagnoses of hypertension, diabetes, obesity, or depression). We did not run subgroup analyses by race/ethnicity because of small sample sizes by race/ethnicity, particularly in nonexpansion states. To test whether the impact of the policy varied by subgroups, we included an additional interaction term in our primary difference-in-differences model specification.

To better understand changes in specific patterns of insurance churn, we also conducted supplementary analyses applying the main difference-in-differences model to more detailed churn outcomes; for example, although our main analysis examined continuous insurance, in supplementary analyses we also examined continuous private insurance and continuous Medicaid separately.

Further, we conducted a number of assumption checks, including prepolicy parallel trends tests and an event study analysis. Finally, we conducted the following sensitivity analyses: first, excluding expansion (Delaware, Massachusetts, New York City, Vermont) and nonexpansion (Wisconsin) states with generous Medicaid for low-income adults or parents before or at the time of the ACA expansion; second, excluding the one-year policy transition period; third, using self-reported prenatal care insurance rather than delivery insurance in the main continuity measures; and finally, estimating standard errors using the wild cluster bootstrap method.

We set a 95 percent confidence level a priori and conducted all analyses using Stata, version 15.1. Observations with missing covariate data were included using a missing indicator variable for each covariate. Design features and survey weights provided by the CDC were applied with Stata’s survey commands to account for the complex survey design. Additional details on the statistical methods are in the online appendix.¹¹ The University of Michigan Institutional Review Board deemed this study of deidentified survey data exempt from review.

Strengths And Limitations

The strengths of this study include a large sample of births across multiple states with detailed longitudinal information on insurance status from before to after pregnancy. Furthermore, the difference-in-differences approach exploited natural variation in Medicaid expansion adoption while controlling for fixed differences between expansion and nonexpansion states. However, the results should be interpreted with knowledge of the following limitations. First, the difference-in-differences design assumes that the outcomes in expansion and nonexpansion states would have changed in the same direction and magnitude if not for the policy intervention (that is, the parallel trends assumption). Graphical inspection of the prepolicy trends from 2012 to 2013 did not provide visual evidence of diverging trends. We also did not find statistical differences in the sociodemographic and clinical characteristics of the two groups from the prepolicy period to the postpolicy period, which suggests that the composition of the treatment and control groups was not changing differentially over time. Further, statistical tests of the year-to-year difference in the outcomes in the prepolicy period did not reveal statistically significant differences (see appendix exhibits A2–A4).¹¹ However, our ability to further investigate the plausibility of this assumption was limited by the availability of only two years of prepolicy data, which is insufficient to establish a stable trend. Second, PRAMS does not contain the detail necessary to examine transitions in continuity across private insurance plans (private–private discontinuity) or Medicaid programs (Medicaid–Medicaid discontinuity). Thus, estimates of insurance disruptions are likely conservative. Third, our sample included data from nineteen states and one city. Although this set of jurisdictions represents a broad range of geographies and policy contexts, the results might not generalize to specific states not included in the survey. Finally, respondents self-report insurance status at the preconception and postpartum points, which may be subject to recall and reporting bias. However, we would expect any reporting bias to be constant within individuals over the perinatal period (and nondifferential across states), which would not affect measures of insurance changes.

Study Results

The study sample included 47,617 women, representing 2.1 million low-income women in nineteen states and New York City. Of this total, 16,363 women resided in nonexpansion states and 31,254 resided in expansion states. Exhibit 1 describes the sample characteristics. In the prepolicy period, expansion states in the sample had a higher proportion of women age thirty-five and older (11.4 percent compared with 7.3 percent in nonexpansion states), a lower proportion of non-Hispanic White women (37.0 percent compared with 61.2 percent in nonexpansion states), and a higher proportion of non–English speakers (18.6 percent compared with 6.7 percent in nonexpansion states).

Exhibits 2 and 3 show the unadjusted trends in insured–uninsured churn and Medicaid–private churn by state Medicaid expansion status (trends for continuous insurance and continuous uninsurance are in appendix exhibit A1).¹¹ For each outcome, expansion and nonexpansion states had similar levels before expansion implementation in 2014 and relatively stable estimates from 2012 to 2013. The proportion of continuously insured women in the prepolicy period was 44.2 percent in expansion states and 41.4 percent in nonexpansion states (exhibit 4). A small proportion of women were continuously uninsured: 1.6 percent and 1.5 percent in the prepolicy period in expansion and nonexpansion states, respectively.

SOURCE Authors’ analysis of 2012–17 Pregnancy Risk Assessment Monitoring System data from nineteen states and New York City. NOTES N = 47,617. Sample includes low-income women (<138 percent of the federal poverty level) with complete insurance information. Unadjusted survey weighted proportions are presented. Bars represent 95% confidence intervals. The transition period is the first year of Medicaid expansion implementation.

SOURCE Authors’ analysis of 2012–17 Pregnancy Risk Assessment Monitoring System data from nineteen states and New York City. NOTES N = 47,617. Sample includes low-income women (<138 percent of the federal poverty level) with complete insurance information. Unadjusted survey weighted proportions are presented. Bars represent 95% confidence intervals. The transition period is the first year of Medicaid expansion implementation.

In adjusted models, Medicaid expansion was associated with a 10.1-percentage-point decline (95% CI: −16.0, −4.3) in insured–uninsured churn in expansion states relative to nonexpansion states (exhibit 4), representing a 28 percent decrease from the prepolicy baseline of 36.2 percent in expansion states. This decline was driven by a 5.8-percentage-point increase in continuous insurance (95% CI: 0.01, 11.6; 13 percent increase from baseline in expansion states) and a 4.2-percentage-point increase in Medicaid–private churn (95% CI: 1.4, 7.1; 23 percent increase from baseline in expansion states). We did not find evidence of differential changes in the rate of continuous uninsurance in expansion and nonexpansion states.

In our adjusted analysis of more detailed insurance churn patterns (see appendix exhibits A10 and A11),¹¹ we found a 7.8-percentage-point increase in continuous Medicaid coverage in expansion states relative to nonexpansion states (95% CI: 2.2, 13.4). We also confirmed that the majority of the reduction in insured–uninsured churn was a result of reduced uninsured–Medicaid–Medicaid churn (−5.4 percentage points; 95% CI: −8.4, −2.4). Further, we found that the majority of the increase in Medicaid–private churn in expansion states was a result of increased private–Medicaid–private churn (2.4 percentage points, 95% CI: −0.6, 5.5). We did not find evidence of an association between Medicaid expansion and the rate of continuous uninsurance, continuous private insurance, or private–uninsured churn.

In subgroup analyses, we did not find evidence of significant differences in the impact of Medicaid expansion by parity or chronic conditions (appendix exhibit A12).¹¹ In each of the sensitivity analyses conducted (appendix exhibits A5–A9),¹¹ our results were largely similar in both practical and statistical significance. Of note, the difference-in-difference estimates were larger when we excluded jurisdictions with more generous Medicaid eligibility criteria for low-income adults or parents before expansion (Delaware, Massachusetts, New York City, Vermont, and Wisconsin). In addition, when we used self-reported insurance for prenatal care, rather than primary payment for delivery recorded by the delivery institution, the increase in Medicaid–private churn associated with Medicaid expansion was small and not statistically significant. This could reflect the fact that women are switching to Medicaid later in pregnancy and thus are continuing to report private insurance as the payer for prenatal care, despite eventually enrolling in Medicaid.

Applying our results to total birth counts in the fourteen expansion jurisdictions in our sample, our findings suggest that the ACA Medicaid expansions resulted in an estimated 22,000 more low-income women with continuous perinatal insurance per year (1.9 percent of all births in those states), 38,000 fewer low-income women experiencing insured–uninsured churn (3.3 percent of all births), and 16,000 more women experiencing private–Medicaid churn (1.4 percent of all births) (data not shown).

Discussion

Using multistate data from the period 2012–17, we found that Medicaid expansion significantly improved perinatal insurance continuity for low-income women. Recent clinical guidance from the American College of Obstetrics and Gynecology emphasizes the need for continuous access to health care not only during pregnancy but also in the preconception period, when many pregnancy risk factors, such as chronic conditions, can be managed and optimized before pregnancy, as well as during the postpartum period, when many women experience morbidity and significant physical and mental health needs.^12,13 With maternal mortality rising in the United States, there is renewed attention to the 31 percent of maternal deaths that occur during pregnancy, as well as the 51.7 percent of maternal deaths that occur during the postpartum year.¹⁴ Improvements in the stability of perinatal insurance for low-income women could have important implications for the quality and continuity of perinatal care and, ultimately, maternal and infant health outcomes.

In the general adult and pediatric populations, insurance churn is associated with disruptions in physician care, increased emergency department use, lower medication adherence, and worsened self-reported quality of care.^15–18 Although research is limited among obstetric populations, uninsurance in the month before conception has been associated with lower rates of timely prenatal care use, higher rates of preterm delivery and low birthweight, and higher risk for postpartum visit nonattendance.^19,20 Research on earlier Medicaid expansions has found associated improvements in the receipt of early prenatal care and improved birth outcomes,^21,22 as well as better access to care, improved mental health, and reduced psychological distress among low-income parents.^23,24 However, the limited evidence so far on the ACA-related Medicaid expansions and infant outcomes, such as preterm birth and low birthweight, indicates no overall effects, although one study found a reduction in the Black–White disparity in preterm birth.^25,26 Research on the impact of policies such as Medicaid expansion on maternal and infant health is constrained by the lack of large-scale population-level data on longer-term infant outcomes and child development and, importantly, on maternal health outcomes. In light of rising maternal morbidity and mortality rates in the US, there is an urgent need for more systematic and broad surveillance of maternal health outcomes both during and after pregnancy, including the patterns of insurance coverage that may influence these outcomes.

Policy Implications

National rates of perinatal insurance churn would be significantly reduced if all states adopted the Medicaid expansion.

This analysis suggests that national rates of perinatal insurance churn would be significantly reduced if all states adopted the ACA-related Medicaid expansion, which may have important implications for maternal health equity. Medicaid expansion across all states could disproportionally improve the stability of perinatal insurance for non-Hispanic Black and Hispanic pregnant women, who are more likely than non-Hispanic White women to experience perinatal insurance discontinuity, reside in Medicaid nonexpansion states, and have incomes less than 138 percent of the federal poverty level.^27,28

Short of full expansion of Medicaid to all low-income adults, states could also pursue expansions of pregnancy-related Medicaid through the first year postpartum, a policy proposal that is currently being considered on the federal level and by several states.^29,30 This policy could also produce benefits in expansion states, where the gap between income eligibility thresholds for pregnant women (state median: 203 percent of the federal poverty level) and parents/low-income adults (state median: 138 percent of the federal poverty level) continues to result in women churning on and off of pregnancy-related Medicaid coverage.^6,31 Although a postpartum extension would not improve continuity of coverage from preconception to delivery, it would be an important step toward reducing loss of Medicaid coverage at sixty days after delivery, particularly given that many women continue to experience pregnancy-related morbidity beyond sixty days. Federal lawmakers could also address the ACA “coverage gap,” which currently means that women with household incomes less than 100 percent of the federal poverty level in ACA Medicaid nonexpansion states cannot access either Medicaid or subsidies to purchase Marketplace coverage before and after pregnancy-related Medicaid eligibility ends (which are only available to people with household incomes of 100–400 percent of the federal poverty level).

We found that nearly half of the decrease in Medicaid–uninsured churn was offset by Medicaid–private churn in expansion states (and the private–Medicaid–private churn pattern in particular). Studies in the general low-income adult population have found that churning between insurance types, even without a gap in insurance, is associated with disruptions in physician care and reductions in patient-reported quality of care that are similar in magnitude to churning on and off insurance.¹⁵ We are not aware of any research on the impact of Medicaid–private churn in obstetric populations; however, this evidence would be an important contribution to the literature and would help inform how policies to reduce this pattern of churn would affect health care use and outcomes.

Despite the positive impact of the ACA Medicaid expansions, rates of discontinuous perinatal insurance are remarkably high.

Despite the positive impact of the ACA Medicaid expansions, rates of discontinuous perinatal insurance are remarkably high in the US. In our sample in the postpolicy period (2015–17), 47.5 percent of low-income women in expansion states and 57.5 percent in nonexpansion states reported noncontinuous perinatal insurance. Another analysis of women at all income levels in forty-one states after the ACA from 2015 to 2017 found that 33.9 percent of women experienced a health insurance disruption from preconception to postpartum.⁶ A number of recently proposed and implemented federal and state policies could further exacerbate perinatal insurance instability in both expansion and nonexpansion states. First, recent changes to ACA implementation, such as reductions in insurance navigator funding, are forecast to reduce overall levels of coverage, particularly among nonelderly adults who purchase private insurance on the ACA Marketplaces.³² Second, a number of states have approved or pending Medicaid waivers to implement premiums (eleven states as of July 2020), and new eligibility restrictions such as work requirements (fifteen states).³³ Although these waivers generally exclude pregnant women (from conception to sixty days postpartum), they would likely reduce Medicaid enrollment in the preconception and extended postpartum periods. Third, the Department of Homeland Security recently implemented a “public charge” rule that affects the ability of legal immigrants who enroll in Medicaid to adjust visa status, reenter the US, or obtain permanent residency.³⁴ The policy is likely to reduce coverage rates among Hispanic women, contributing to higher levels of continuous uninsurance and Medicaid–uninsured churn.

Conclusion

The ACA Medicaid expansions resulted in substantial improvements in perinatal insurance continuity for low-income women. A number of policy options are available to federal and state policy makers that could further improve the stability of health insurance coverage for pregnant and postpartum women, including adoption of the ACA Medicaid expansion, extension of pregnancy Medicaid to one year postpartum, and closing the ACA coverage gap for people with incomes less than 100 percent of the federal poverty level. Additional evidence on the impact of different types of perinatal insurance patterns on access to care and maternal and infant health outcomes could help inform these policy proposals.

ACKNOWLEDGMENTS

This study was supported by a grant awarded to Lindsay Admon (principal investigator), Tyler Winkelman (co-investigator), and Vanessa Dalton (co-investigator) from the Health Resources and Services Administration, Department of Health and Human Services, as part of an award totaling $100,000. The contents of the manuscript are those of the authors and do not necessarily represent the official views of, nor an endorsement by, the Health Resources and Services Administration, the Department of Health and Human Services, or the federal government. Jamie Daw’s work on this project was supported by a Calderone Junior Faculty Prize. The funders had no role in the study design, writing of the report, or the decision to submit the article for publication. Dalton was a paid, expert witness for Bayer and a consultant for Bind. The other authors have no conflicts of interest to report. The authors thank all Pregnancy Risk Assessment Monitoring System study participants and members of the Pregnancy Risk Assessment Monitoring System Working Group.

NOTES

• 1 Eisenberg JM, Power EJ. Transforming insurance coverage into quality health care: voltage drops from potential to delivered quality. JAMA. 2000;284(16):2100–7. Crossref, Medline, Google Scholar
• 2 Daw JR, Hatfield LA, Swartz K, Sommers BD. Women in the United States experience high rates of coverage “churn” in months before and after childbirth. Health Aff (Millwood). 2017;36(4):598–606. Go to the article, Google Scholar
• 3 D’Angelo DV, Le B, O’Neil ME, Williams L, Ahluwalia IB, Harrison LLet al. Patterns of health insurance coverage around the time of pregnancy among women with live-born infants—Pregnancy Risk Assessment Monitoring System, 29 States, 2009. MMWR Surveill Summ. 2015;64(4):1–19. Medline, Google Scholar
• 4 Johnston EM, Strahan AE, Joski P, Dunlop AL, Adams EK. Impacts of the Affordable Care Act’s Medicaid expansion on women of reproductive age: differences by parental status and state policies. Womens Health Issues. 2018;28(2):122–9. Crossref, Medline, Google Scholar
• 5 Clapp MA, James KE, Kaimal AJ, Daw JR. Preconception coverage before and after the Affordable Care Act Medicaid expansions. Obstet Gynecol. 2018;132(6):1394–400. Crossref, Medline, Google Scholar
• 6 Daw JR, Kozhimannil KB, Admon LK. High rates of perinatal insurance churn persist after the ACA. Health Affairs Blog [blog on the Internet]. 2019 Sep 16 [cited 2020 Jul 7]. Available from: https://www.healthaffairs.org/do/10.1377/hblog20190913.387157/full/ Google Scholar
• 7 Shulman HB, D’Angelo DV, Harrison L, Smith RA, Warner L. The Pregnancy Risk Assessment Monitoring System (PRAMS): overview of design and methodology. Am J Public Health. 2018;108(10):1305–13. Crossref, Medline, Google Scholar
• 8 Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation. Prior HHS poverty guidelines and Federal Register references [Internet]. Washington (DC): HHS; [cited 2020 Jul 31]. Available from: https://aspe.hhs.gov/prior-hhs-poverty-guidelines-and-federal-register-references Google Scholar
• 9 Artiga S, Arguello R, Duckett P. Health coverage and care for American Indians and Alaska Natives [Internet]. San Francisco (CA): Henry J. Kaiser Family Foundation; 2013 Oct 7 [cited 2020 Jul 7]. Available from: https://www.kff.org/disparities-policy/issue-brief/health-coverage-and-care-for-american-indians-and-alaska-natives/view/print/ Google Scholar
• 10 Cohen R. Long-term trends in health insurance: estimates from the National Health Interview Survey, 1968–2018 [Internet]. Hyattsville (MD): National Center for Health Statistics; 2019 Jul [cited 2020 Jul 7]. Available from: https://www.cdc.gov/nchs/data/nhis/health_insurance/TrendHealthInsurance1968_2018.pdf Google Scholar
• 11 To access the appendix, click on the Details tab of the article online.
• 12 American College of Obstetricians and Gynecologists. ACOG Committee Opinion No. 762: prepregnancy counseling. Obstet Gynecol. 2019;133(1):e78–89. Crossref, Medline, Google Scholar
• 13 McKinney J, Keyser L, Clinton S, Pagliano C. ACOG Committee Opinion No. 736: optimizing postpartum care. Obstet Gynecol. 2018;132(3):784–5. Crossref, Medline, Google Scholar
• 14 Petersen EE, Davis NL, Goodman D, Cox S, Mayes N, Johnston Eet al. Vital signs: pregnancy-related deaths, United States, 2011–2015, and strategies for prevention, 13 states, 2013–2017. MMWR Morb Mortal Wkly Rep. 2019;68(18):423–9. Crossref, Medline, Google Scholar
• 15 Sommers BD, Gourevitch R, Maylone B, Blendon RJ, Epstein AM. Insurance churning rates for low-income adults under health reform: lower than expected but still harmful for many. Health Aff (Millwood). 2016;35(10):1816–24. Go to the article, Google Scholar
• 16 Federico SG, Steiner JF, Beaty B, Crane L, Kempe A. Disruptions in insurance coverage: patterns and relationship to health care access, unmet need, and utilization before enrollment in the State Children’s Health Insurance Program. Pediatrics. 2007;120(4):e1009–16. Crossref, Medline, Google Scholar
• 17 Sudano JJ, Baker DW. Intermittent lack of health insurance coverage and use of preventive services. Am J Public Health. 2003;93(1):130–7. Crossref, Medline, Google Scholar
• 18 Lavarreda SA, Gatchell M, Ponce N, Brown ER, Chia YJ. Switching health insurance and its effects on access to physician services. Med Care. 2008;46(10):1055–63. Crossref, Medline, Google Scholar
• 19 Clapp MA, James KE, Kaimal AJ. Preconception uninsurance: associations with prenatal care and pregnancy outcomes. Am J Obstet Gynecol. 2019;220(1, Suppl):S492–3. Crossref, Medline, Google Scholar
• 20 Wilcox A, Levi EE, Garrett JM. Predictors of non-attendance to the postpartum follow-up visit. Matern Child Health J. 2016;20(1, Suppl 1):22–7. Crossref, Medline, Google Scholar
• 21 Wherry LR. State Medicaid expansions for parents led to increased coverage and prenatal care utilization among pregnant mothers. Health Serv Res. 2018;53(5):3569–91. Crossref, Medline, Google Scholar
• 22 Currie J, Gruber J. Saving babies: the efficacy and cost of recent changes in the Medicaid eligibility of pregnant women. J Polit Econ. 1996;104(6):1263–96. Crossref, Google Scholar
• 23 McMorrow S, Kenney GM, Long SK, Goin DE. Medicaid expansions from 1997 to 2009 increased coverage and improved access and mental health outcomes for low-income parents. Health Serv Res. 2016;51(4):1347–67. Crossref, Medline, Google Scholar
• 24 McMorrow S, Gates JA, Long SK, Kenney GM. Medicaid expansion increased coverage, improved affordability, and reduced psychological distress for low-income parents. Health Aff (Millwood). 2017;36(5):808–18. Go to the article, Google Scholar
• 25 Clapp MA, James KE, Kaimal AJ, Sommers BD, Daw JR. Association of Medicaid expansion with coverage and access to care for pregnant women. Obstet Gynecol. 2019;134(5):1066–74. Crossref, Medline, Google Scholar
• 26 Brown CC, Moore JE, Felix HC, Stewart MK, Bird TM, Lowery CLet al. Association of state Medicaid expansion status with low birth weight and preterm birth. JAMA. 2019;321(16):1598–609. Crossref, Medline, Google Scholar
• 27 Daw JR, Kolenic GE, Dalton VK, Zivin K, Winkelman T, Kozhimannil KBet al. Racial and ethnic disparities in perinatal insurance coverage. Obstet Gynecol. 2020;135(4):917–24. Crossref, Medline, Google Scholar
• 28 Artiga S, Orgera K, Damico A. Changes in health coverage by race and ethnicity since the ACA, 2010–2018 [Internet]. San Francisco (CA): Henry J. Kaiser Family Foundation; 2020 Mar 5 [cited 2020 Jul 7]. Available from: https://www.kff.org/disparities-policy/issue-brief/changes-in-health-coverage-by-race-and-ethnicity-since-the-aca-2010-2018/ Google Scholar
• 29 Ranji U, Gomez I, Salganicoff A. Expanding postpartum Medicaid coverage [Internet]. San Francisco (CA): Henry J. Kaiser Family Foundation; 2019 May 22 [cited 2020 Jul 7]. Available from: https://www.kff.org/womens-health-policy/issue-brief/expanding-postpartum-medicaid-coverage/ Google Scholar
• 30 Kelly RL. H.R.4996—Helping MOMS Act of 2019 [Internet]. Washington (DC): 116th Congress; 2019 Nov 20 [cited 2020 Jul 7]. Available from: https://www.congress.gov/bill/116th-congress/house-bill/4996 Google Scholar
• 31 Brooks T, Roygardner L, Artiga S, Pham O, Dolan R. Medicaid and CHIP eligibility, enrollment, and cost sharing policies as of January 2020: findings from a 50-state survey [Internet]. San Francisco (CA): Henry J. Kaiser Family Foundation; 2020 Mar 26 [cited 2020 Jul 7]. Available from: https://www.kff.org/medicaid/report/medicaid-and-chip-eligibility-enrollment-and-cost-sharing-policies-as-of-january-2020-findings-from-a-50-state-survey/ Google Scholar
• 32 Pollitz K, Tolbert J, Diaz M. Data note: further reductions in Navigator funding for federal Marketplace states [Internet]. San Francisco (CA): Henry J. Kaiser Family Foundation; 2018 Sep [cited 2020 Jul 7]. Available from: http://files.kff.org/attachment/Data-Note-Further-Reductions-in-Navigator-Funding-for-Federal-Marketplace-States Google Scholar
• 33 Henry J. Kaiser Family Foundation. Medicaid Waiver Tracker: approved and pending Section 1115 waivers by state [Internet]. San Francisco (CA): KFF; 2020 Jun 26 [updated 2020 Jul 23; cited 2020 Jul 31]. Available from: https://www.kff.org/medicaid/issue-brief/medicaid-waiver-tracker-approved-and-pending-section-1115-waivers-by-state/ Google Scholar
• 34 US Citizenship and Immigration Services, Department of Homeland Security. Inadmissibility on public charge grounds. Fed Regist. 2019;84(157):41292–508. Google Scholar