Abstract
This cross-sectional study uses Pregnancy Risk Assessment Monitoring System data to investigate the association between marketplace pregnancy special enrollment and prenatal insurance coverage in New York.
Introduction
The Affordable Care Act established health insurance marketplaces in 2014 for private insurance plan enrollment. Federally required open enrollment is 2 to 3 consecutive months annually.1 Outside federal enrollment periods, state-run exchanges may offer special enrollment for qualifying events.1 Childbirth is a federal qualifying event, but before 2016, no state had included pregnancy.2 Without special enrollment, pregnant people must wait until open enrollment for marketplace coverage, which could delay insurance access.2 In 2016, New York became the first state to establish a pregnancy special enrollment period, later followed by 5 additional states and the District of Columbia.2,3 We investigated the association between establishing a pregnancy special enrollment period and prenatal insurance coverage in New York.
Methods
In this cross-sectional study, we conducted a difference-in-difference analysis using Pregnancy Risk Assessment Monitoring System (PRAMS) data, comparing prenatal insurance coverage before (2014-2015) and after (2017-2019) New York’s policy change vs 17 control states with PRAMS data from 2014 to 2019. We omitted 2016 because the pregnancy period for most births in 2016 overlapped with prepolicy and postpolicy periods. This study followed the STROBE reporting guideline and was considered to be non–human participant research by the Brown University institutional review board.
We hypothesized that the benefits of the policy change would be greatest among pregnant people eligible for marketplace subsidies who had incomes above Medicaid eligibility levels. We restricted the sample to respondents above New York’s pregnancy-related Medicaid eligibility (223% of the federal poverty level) and below 400% of the federal poverty level. The outcomes were prenatal marketplace coverage; employer, private, or military coverage; Medicaid; uninsurance; or other coverage.
We calculated weighted respondent characteristics in New York and control states, including self-reported race and ethnicity from birth certificates. Regression models included an interaction term between birth in New York and postpolicy births. Models included state and year fixed effects, with SEs clustered by state. Multivariable models were adjusted for demographic factors and previous live births. Analyses were weighted to account for the complex survey design. Methods for evaluating model assumptions, categorizing insurance types, and identifying which PRAMS income categories met inclusion criteria are described in eMethods 1 to 3 in the Supplement. Analyses were conducted between April 7 and September 8, 2022, using Stata, version 17 software (StataCorp LLC).
Results
The study included 13 753 births (1167 in New York, 12 586 in control states), representing a weighted total of 712 175 births. Relative to controls (n = 4014), New York respondents (n = 354) were older (20.1% aged 35-39 years vs 14.0%) and less likely to be married (Table 1). In New York, unadjusted estimates of prenatal marketplace coverage significantly increased from 3.3% to 8.6% after the pregnancy special enrollment policy, while uninsurance fell from 1.5% to 0.6%. In control states, there were significant decreases in unadjusted estimates of other prenatal private or military coverage and increases in prenatal Medicaid from prepolicy to postpolicy. In adjusted models, New York’s pregnancy special enrollment period was associated with an increase of 6.3 (95% CI, 4.4-8.2) percentage points in the prenatal marketplace and decreases of 1.4 (95% CI, −2.1 to −0.7) percentage points in uninsurance and 3.0 (95% CI, −3.8 to −2.3) percentage points in other coverage relative to control states (Table 2). The policy was not associated with significant changes in other private, military, or Medicaid coverage.
Table 1. Prepolicy Demographic Characteristics of the Study Sample,a Pregnancy Risk Assessment Monitoring System (PRAMS) 2014-2015.
| Characteristic | New York | Control states | ||
|---|---|---|---|---|
| Unweighted, No. | Weighted, % (95% CI) | Unweighted, No. | Weighted, % (95% CI) | |
| No. | 354 | 32 532 | 4014 | 191 597 |
| Age, y | ||||
| ≤24 | 38 | 12.6 (8.8-17.8) | 626 | 15.0 (13.5-16.6) |
| 25-29 | 115 | 33.4 (27.5-39.7) | 1459 | 37.8 (35.7-40.0) |
| 30-34 | 108 | 29.7 (24.1-36.0) | 1292 | 30.7 (28.7-32.7) |
| 35-39 | 74 | 20.1 (15.3-25.9) | 538 | 14.0 (12.6-15.6) |
| ≥40 | 19 | 4.2 (2.4-7.3) | 99 | 2.5 (1.9-3.3) |
| Race and ethnicityb | ||||
| Black | 64 | 13.5 (10.1-17.9) | 429 | 8.0 (7.0-9.2) |
| Hispanic | 73 | 20.4 (15.7-26.0) | 393 | 8.7 (7.8-9.7) |
| White | 170 | 54.0 (48.2-59.8) | 2655 | 74.5 (72.9-76.0) |
| Otherc | 47 | 12.0 (8.6-16.6) | 524 | 8.5 (7.6-9.5) |
| Education | ||||
| High school or less | 58 | 16.4 (12.0-22.1) | 725 | 17.9 (16.2-19.7) |
| More than high school | 296 | 83.6 (77.9-88.0) | 3248 | 81.5 (79.7-83.1) |
| Marital status | ||||
| Married | 237 | 68.3 (61.8-74.1) | 3103 | 77.0 (75.0-78.8) |
| Not married | 117 | 31.7 (25.9-38.2) | 909 | 23.0 (21.1-25.0) |
| Survey language | ||||
| English | 339 | 95.6 (92.3-97.6) | 3940 | 98.2 (97.5-98.7) |
| Spanish or Chinese | 15 | 4.4 (2.4-7.7) | 74 | 1.8 (1.3-2.5) |
| No. of previous live births | ||||
| 0 | 171 | 43.4 (37.1-49.9) | 1679 | 39.5 (37.4-41.6) |
| 1 | 128 | 38.7 (32.6-45.3) | 1582 | 41.3 (39.2-43.5) |
| 2 | 42 | 12.8 (8.9-18.0) | 586 | 15.5 (14.0-17.2) |
| ≥3 | 12 | 4.5 (2.3-8.5) | 155 | 3.5 (2.8-4.3) |
N = 4368; weighted N = 224 129. Unweighted sample sizes and weighted proportions are presented using PRAMS survey weights. Sample with full income ranges from 223% to 400% of the federal poverty level in New York (New York City and New York State) and 17 control group states (Alaska, Delaware, Iowa, Illinois, Massachusetts, Maryland, Maine, Missouri, New Hampshire, New Jersey, New Mexico, Pennsylvania, Rhode Island, Utah, Washington, Wisconsin, and Wyoming). Spanish and Chinese were combined for survey language categories because PRAMS surveys were only available in Chinese in New York.
Race and ethnicity were self-reported in the birth certificate files.
Other categories were Alaska Native, American Indian, Asian, Pacific Islander, mixed race, and other race and ethnicity.
Table 2. Estimates of the Association Between New York State’s Marketplace Special Enrollment Period for Pregnancy and Prenatal Insurance Coverage, Pregnancy Risk Assessment Monitoring System (PRAMS) 2014-2019a.
| Prenatal coverage | % (95% CI) | Difference in differences, percentage points (95% CI) | ||||||
|---|---|---|---|---|---|---|---|---|
| New York (n = 1167; weighted n = 108 640) | Control states (n = 12 586; weighted n = 603 535) | |||||||
| Prepolicy | Postpolicy | Difference | Prepolicy | Postpolicy | Difference | Unadjusted | Adjusted | |
| Marketplace | 3.3 (1.5 to 7.1) | 8.6 (6.0 to 12.3) | 5.3 (1.3 to 9.3)b | 6.2 (5.2 to 7.3) | 5.3 (4.6 to 6.1) | −0.9 (−2.2 to 0.4) | 6.2 (4.4 to 8.0)c | 6.3 (4.4 to 8.2)c |
| Employer, other private, or military | 70.3 (64.0 to 75.9) | 64.4 (59.2 to 69.2) | −5.9 (−13.7 to 1.9) | 79.5 (77.7 to 81.3) | 76.2 (74.6 to 77.7) | −3.3 (−5.7 to −1.0)d | −2.6 (−5.3 to 0.1) | −1.2 (−3.9 to 1.6) |
| Medicaid | 18.6 (14.0 to 24.3) | 20.2 (16.4 to 24.7) | 1.6 (−5.0 to 8.2) | 10.6 (9.2 to 12.0) | 12.5 (11.3 to 13.7) | 1.9 (0.1 to 3.8)b | −0.2 (−1.9 to 1.4) | −1.4 (−3.0 to 0.2) |
| Uninsured | 1.5 (0.6 to 3.5) | 0.6 (0.1 to 4.4) | −0.8 (−2.6 to 0.9) | 1.1 (0.8 to 1.6) | 1.5 (1.1 to 2.1) | 0.4 (−0.3 to 1.0) | −1.2 (−1.9 to −0.6)d | −1.4 (−2.1 to −0.7)c |
| Other | 5.5 (3.2 to 9.4) | 3.0 (1.7 to 5.3) | −2.6 (−6.0 to 0.9) | 1.6 (1.2 to 2.2) | 2.0 (1.6 to 2.5) | 0.4 (−0.3 to 1.1) | −2.9 (−3.6 to −2.2)c | −3.0 (−3.8 to −2.3)c |
N = 13 753; weighted N = 712 175. Weighted proportions are presented using PRAMS survey weights. Sample with full income ranges from 223% to 400% federal poverty level in New York (New York City and New York State) and 17 control group states (Alaska, Delaware, Iowa, Illinois, Massachusetts, Maryland, Maine, Missouri, New Hampshire, New Jersey, New Mexico, Pennsylvania, Rhode Island, Utah, Washington, Wisconsin, and Wyoming). Marketplace coverage includes insurance from the health care exchange or purchased directly from the insurance company. Employer or parental private or military coverage includes private insurance through work or through a parent, TRICARE, or other military insurance. Uninsurance includes individuals with Indian Health Service coverage. Adjusted difference-in-difference model includes age, educational attainment, marital status, language of survey completion, number of previous live births, and state and year fixed effects. Spanish and Chinese were combined for survey language categories because PRAMS surveys were only available in Chinese in New York. 2016 Omitted as a transition year.
Two-sided P < .05.
Two-sided P < .001.
Two-sided P < .01.
Discussion
New York’s pregnancy special enrollment period was associated with increased marketplace coverage and decreased uninsurance among eligible pregnant people. Previous research has shown that gaps in marketplace coverage are common for eligible and enrolled pregnant people and associated with reduced prenatal care receipt.4,5 Study limitations include that coverage was measured as prenatal care payer, which did not capture when coverage started or duration of uninsurance. This analysis can inform policy debates about how to improve pregnancy insurance coverage, care, and health outcomes in the 18 states that can establish their own exchanges’ enrollment periods.3,6
eMethods 1. Difference-in-Difference Model Assumptions
eMethods 2. Sample Selection
eMethods 3. Definition of Outcomes
Data Sharing Statement
References
- 1.Norris L. What are the deadlines for the ACA’s open enrollment period? healthinsurance.org . Published 2022. Accessed September 9, 2022. https://www.healthinsurance.org/faqs/what-are-the-deadlines-for-the-acas-open-enrollment-period/
- 2.Chen A, Spalding M. State creation of special enrollment periods for pregnancy. National Health Law Program . Published January 26, 2017. Accessed September 9, 2022. https://healthlaw.org/wp-content/uploads/2017/01/Pregnancy-SEP-FINAL-1.pdf
- 3.Norris L. Exceptional circumstances for special enrollment. healthinsurace.org . Published 2022. Accessed September 12, 2022. https://www.healthinsurance.org/special-enrollment-guide/exceptional-circumstances-for-special-enrollment/#pregnancy
- 4.Gordon SH, Alger CL, Declercq ER, Garrido MM. The association between continuity of marketplace coverage during pregnancy and receipt of prenatal care. Health Aff (Millwood). 2021;40(10):1618-1626. doi: 10.1377/hlthaff.2021.00581 [DOI] [PubMed] [Google Scholar]
- 5.Eliason EL, Daw JR, Allen HL. Association of Medicaid vs marketplace eligibility on maternal coverage and access with prenatal and postpartum care. JAMA Netw Open. 2021;4(12):e2137383. doi: 10.1001/jamanetworkopen.2021.37383 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Haley JM, Johnston EM, Hill I, Kenney GM, Thomas TW. The Public health insurance landscape for pregnant and postpartum women: state and federal policies in 2020. Urban Institute . Published January 2021. Accessed November 1, 2022. https://www.urban.org/sites/default/files/publication/103561/the-public-health-insurance-landscape-for-pregnant-and-postpartum-women_1.pdf
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eMethods 1. Difference-in-Difference Model Assumptions
eMethods 2. Sample Selection
eMethods 3. Definition of Outcomes
Data Sharing Statement
