[Skip to Content]
[Skip to Content Landing]
Figure.
Trends in Unadjusted Rates of Low Birth Weight and Preterm Birth Outcomes in Medicaid Expansion and Nonexpansion States by Race/Ethnicity
Trends in Unadjusted Rates of Low Birth Weight and Preterm Birth Outcomes in Medicaid Expansion and Nonexpansion States by Race/Ethnicity

Data are from the Vital Statistics Birth Data File (2011-2016). Outcomes were calculated among 8 244 924 births to white women, 2 201 658 births to black women, and 3 944 665 births to Hispanic women. The vertical dotted line indicates the date of expansion for most expansion states on January 1, 2014. Three states included in the analyses (Alaska, Louisiana, and Pennsylvania) expanded at a later date (see eTable 2 in the Supplement for dates).

Table 1.  
Descriptive Statistics of State-Level Characteristics by Medicaid Expansion Status in the Pre–Medicaid Expansion Perioda,b
Descriptive Statistics of State-Level Characteristics by Medicaid Expansion Status in the Pre–Medicaid Expansion Perioda,b
Table 2.  
Descriptive Statistics of Individual-Level Characteristics by Racial/Ethnic Group and Medicaid Expansion Status in the Pre–Medicaid Expansion Perioda
Descriptive Statistics of Individual-Level Characteristics by Racial/Ethnic Group and Medicaid Expansion Status in the Pre–Medicaid Expansion Perioda
Table 3.  
Unadjusted and Adjusted Changes in Birth Outcomes Associated With Medicaid Expansiona,b
Unadjusted and Adjusted Changes in Birth Outcomes Associated With Medicaid Expansiona,b
Table 4.  
Unadjusted and Adjusted Changes in Birth Outcomes Associated With Medicaid Expansion Among Medicaid-Covered Births and Medicaid-Covered Births to Women With at Most a High School Degreea,b
Unadjusted and Adjusted Changes in Birth Outcomes Associated With Medicaid Expansion Among Medicaid-Covered Births and Medicaid-Covered Births to Women With at Most a High School Degreea,b
Supplement.

eMethods. Covariate Variables for Difference-in-Differences and Difference-in-Difference-Differences Models

eTable 1. State Categorization of Medicaid Expansion and Adoption of the 2003 Revised Birth Certificate

eTable 2. Timing of State Adoption of 2003 Revised Birth Certification and for Medicaid Expansion

eTable 3. Adjusted Difference-in-Differences for the Association of Medicaid Expansion With Changes in Birth Outcomes, Sensitivity Analysis With Births During 12-Months Postexpansion Removed

eTable 4. Adjusted Difference-in-Difference-in-Differences Estimates for the Association of Medicaid Expansion With Changes in Birth Outcomes, Sensitivity Analysis With Births During 12-Months Post expansion Removed

eTable 5. Difference-in-differences and difference-in-difference-in-differences estimates for the Association of Medicaid Expansion With Changes Birth Outcomes, State Classification Sensitivity Analyses Excluding States That Adopted 2003 Revised Certificate in 2012 or 2013

eTable 6. Difference-in-Differences and Difference-in-Difference-in-Differences Estimates for the Association of Medicaid Expansion With Changes Birth Outcomes, State Classification Sensitivity Analyses Excluding States With Medicaid Expansion After January 1, 2014

eTable 7. Difference-in-Differences and Difference-in-Difference-in-Differences Estimates for the Association of Medicaid Expansion With Changes Birth Outcomes, State Classification Sensitivity Analyses Including Waiver States

eTable 8. Adjusted Difference-in-Differences and Difference-in-Difference-in-Differences Estimates for the Association of Medicaid Expansion With Birth Outcome Mechanisms

eFigure 1. Exclusion Chart Describing the Number and Percent of Births Excluded to Reach the Final Study Sample

eFigure 2. Relative Percent Change in Unadjusted Rates of Low Birth Weight Prior to Medicaid Expansion and Post-Medicaid Expansion Among Expansion and Nonexpansion States, by Subgroup

eFigure 3. Relative Percent Change in Unadjusted Rates of Very Low Birth Weight Prior to Medicaid Expansion and Post-Medicaid Expansion Among Expansion and Non-expansion States, by Subgroup

eFigure 4. Relative Percent Change in Unadjusted Rates of Preterm Birth Outcomes Prior to Medicaid Expansion and Post-Medicaid Expansion Among Expansion and Nonexpansion States, by Subgroup

eFigure 5. Relative Percent Change in Unadjusted Rates of Very Preterm Birth Outcomes Prior to Medicaid Expansion and Post-Medicaid Expansion Among Expansion and Nonexpansion States, by Subgroup

eFigure 6. Rates of Unadjusted Birth Outcomes Among Births to Women of All Races/Ethnicities, by Expansion Status

eFigure 7. Rates of Unadjusted Birth Outcomes Among Births to Women of All Races/Ethnicities, by Expansion Status, Subgroup of Medicaid-Funded Births

eFigure 8. Rates of Unadjusted Birth Outcomes Among Births to Women of All Races/Ethnicities, by Expansion Status, Subgroup of Medicaid-Funded Births to Women With at Most a High School Degree

1.
Matthews  TJ, MacDorman  MF, Thoma  ME.  Infant mortality statistics from the 2013 period linked birth/infant death data set.  Natl Vital Stat Rep. 2015;64(9):1-30.PubMedGoogle Scholar
2.
Behrman  RE, Butler  AS; Institute of Medicine, Committee on Understanding Premature Birth and Assuring Healthy Outcomes.  Preterm Birth: Causes, Consequences, and Prevention. Washington, DC: National Academies Press; 2007.
3.
Bhutta  AT, Cleves  MA, Casey  PH, Cradock  MM, Anand  KJ.  Cognitive and behavioral outcomes of school-aged children who were born preterm: a meta-analysis.  JAMA. 2002;288(6):728-737. doi:10.1001/jama.288.6.728PubMedGoogle ScholarCrossref
4.
de Kieviet  JF, Piek  JP, Aarnoudse-Moens  CS, Oosterlaan  J.  Motor development in very preterm and very low-birth-weight children from birth to adolescence: a meta-analysis.  JAMA. 2009;302(20):2235-2242. doi:10.1001/jama.2009.1708PubMedGoogle ScholarCrossref
5.
D’Onofrio  BM, Class  QA, Rickert  ME, Larsson  H, Långström  N, Lichtenstein  P.  Preterm birth and mortality and morbidity: a population-based quasi-experimental study.  JAMA Psychiatry. 2013;70(11):1231-1240. doi:10.1001/jamapsychiatry.2013.2107PubMedGoogle ScholarCrossref
6.
Delnord  M, Hindori-Mohangoo  AD, Smith  LK,  et al; Euro-Peristat Scientific Committee.  Variations in very preterm birth rates in 30 high-income countries: are valid international comparisons possible using routine data?  BJOG. 2017;124(5):785-794. doi:10.1111/1471-0528.14273PubMedGoogle ScholarCrossref
7.
MacDorman  MF, Matthews  TJ, Mohangoo  AD, Zeitlin  J.  International comparisons of infant mortality and related factors: United States and Europe, 2010.  Natl Vital Stat Rep. 2014;63(5):1-6.PubMedGoogle Scholar
8.
Murphy  SL, Mathews  TJ, Martin  JA, Minkovitz  CS, Strobino  DM.  Annual Summary of Vital Statistics: 2013-2014.  Pediatrics. 2017;139(6):e20163239. doi:10.1542/peds.2016-3239PubMedGoogle ScholarCrossref
9.
Martin  JA, Hamilton  BE, Osterman  MJK, Driscoll  AK, Drake  P.  Births: final data for 2017.  Natl Vital Stat Rep. 2018;67(8):1-50.PubMedGoogle Scholar
10.
Johnson  K, Applegate  M, Gee  RE.  Improving Medicaid: three decades of change to better serve women of childbearing age.  Clin Obstet Gynecol. 2015;58(2):336-354. doi:10.1097/GRF.0000000000000115PubMedGoogle ScholarCrossref
11.
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. doi:10.1377/hlthaff.2016.1241PubMedGoogle ScholarCrossref
12.
Shah  SI, Brumberg  HL.  Predictions of the affordable care act’s impact on neonatal practice.  J Perinatol. 2016;36(8):586-592. doi:10.1038/jp.2016.93PubMedGoogle ScholarCrossref
13.
Bhatt  CB, Beck-Sagué  CM.  Medicaid expansion and infant mortality in the United States.  Am J Public Health. 2018;108(4):565-567. doi:10.2105/AJPH.2017.304218PubMedGoogle ScholarCrossref
14.
McMorrow  S, Long  SK, Kenney  GM, Anderson  N.  Uninsurance disparities have narrowed for black and Hispanic adults under the Affordable Care Act.  Health Aff (Millwood). 2015;34(10):1774-1778. doi:10.1377/hlthaff.2015.0757PubMedGoogle ScholarCrossref
15.
National Center for Health Statistics. NCHS’ Vital Statistics Natality Birth Data. http://www.nber.org/data/vital-statistics-natality-data.html. Accessed October 1, 2018.
16.
Kaestner  R, Garrett  B, Chen  J, Gangopadhyaya  A, Fleming  C.  Effects of ACA Medicaid expansions on health insurance coverage and labor supply.  J Policy Anal Manage. 2017;36(3):608-642. doi:10.1002/pam.21993PubMedGoogle ScholarCrossref
17.
Simon  K, Soni  A, Cawley  J.  The impact of health insurance on preventive care and health behaviors: evidence from the first two years of the ACA Medicaid expansions.  J Policy Anal Manage. 2017;36(2):390-417. doi:10.1002/pam.21972PubMedGoogle ScholarCrossref
18.
Courtemanche  Ch, Marton  J, Ukert  B, Yelowitz  A, Zapata  D.  Early impacts of the Affordable Care Act on health insurance coverage in Medicaid expansion and non-expansion states.  J Policy Anal Manage. 2017;36(1):178-210. doi:10.1002/pam.21961PubMedGoogle ScholarCrossref
19.
Kaiser Family Foundation. State health facts. Status of state action on the Medicaid expansion decision. https://www.kff.org/health-reform/state-indicator/state-activity-around-expanding-medicaid-under-the-affordable-care-act/. Accessed February 1, 2019.
20.
Armstrong  J.  Women’s Health in the age of patient protection and the Affordable Care Act.  Clin Obstet Gynecol. 2015;58(2):323-335. doi:10.1097/GRF.0000000000000096PubMedGoogle ScholarCrossref
21.
Sommers  BD, Blendon  RJ, Orav  EJ, Epstein  AM.  Changes in utilization and health among low-income adults after Medicaid expansion or expanded private insurance.  JAMA Intern Med. 2016;176(10):1501-1509. doi:10.1001/jamainternmed.2016.4419PubMedGoogle ScholarCrossref
22.
Antonisse  L, Garfield  R, Rudowitz  R, Artiga  S. The effects of Medicaid expansion under the ACA: updated findings from a literature review. Kaiser Family Foundation. https://collections.nlm.nih.gov/master/borndig/101707525/Issue-Brief-The-Effects-of-Medicaid-Expansion-Under-the-ACA-Updated-Findings.pdf. Published February 2017. Accessed October 1, 2018.
23.
Gonzales  S, Sommers  BD.  Intra-ethnic coverage disparities among Latinos and the effects of health reform.  Health Serv Res. 2018;53(3):1373-1386. doi:10.1111/1475-6773.12733PubMedGoogle ScholarCrossref
24.
Yue  D, Rasmussen  PW, Ponce  NA.  Racial/ethnic differential effects of Medicaid expansion on health care access.  Health Serv Res. 2018;53(5):3640-3656. doi:10.1111/1475-6773.12834PubMedGoogle ScholarCrossref
25.
Karaca-Mandic  P, Norton  EC, Dowd  B.  Interaction terms in nonlinear models.  Health Serv Res. 2012;47(1, pt 1):255-274. doi:10.1111/j.1475-6773.2011.01314.xPubMedGoogle ScholarCrossref
26.
Ryan  AM, Burgess  JF  Jr, Dimick  JB.  Why we should not be indifferent to specification choices for difference-in-differences.  Health Serv Res. 2015;50(4):1211-1235. doi:10.1111/1475-6773.12270PubMedGoogle ScholarCrossref
27.
Xu  J, Murphy  SL, Kochanek  KD, Bastian  B, Arias  E.  Deaths: final data for 2016.  Natl Vital Stat Rep. 2018;67(5):1-76.PubMedGoogle Scholar
28.
Riddell  CA, Harper  S, Kaufman  JS.  Trends in differences in US mortality rates between black and white infants.  JAMA Pediatr. 2017;171(9):911-913. doi:10.1001/jamapediatrics.2017.1365PubMedGoogle ScholarCrossref
29.
Boudreaux  MH, Dagher  RK, Lorch  SA.  The association of health reform and infant health: evidence from Massachusetts.  Health Serv Res. 2017;53(4):2406-2424. doi:10.1111/1475-6773.12779PubMedGoogle ScholarCrossref
30.
Henry J. Kaiser Family Foundation. Population distribution by race/ethnicity. https://www.kff.org/other/state-indicator/distribution-by-raceethnicity/. Published 2017. Accessed October 1, 2018.
31.
Howell  EM.  The impact of the Medicaid expansions for pregnant women: a synthesis of the evidence.  Med Care Res Rev. 2001;58(1):3-30. doi:10.1177/107755870105800101PubMedGoogle ScholarCrossref
32.
Lu  MC, Kotelchuck  M, Hogan  V, Jones  L, Wright  K, Halfon  N.  Closing the black-white gap in birth outcomes: a life-course approach.  Ethn Dis. 2010;20(1)(suppl 2):S2-S62, 76.PubMedGoogle Scholar
33.
Braveman  P, Barclay  C.  Health disparities beginning in childhood: a life-course perspective.  Pediatrics. 2009;124(suppl 3):S163-S175. doi:10.1542/peds.2009-1100DPubMedGoogle ScholarCrossref
34.
Roohan  PJ, Josberger  RE, Acar  J, Dabir  P, Feder  HM, Gagliano  PJ.  Validation of birth certificate data in New York State.  J Community Health. 2003;28(5):335-346. doi:10.1023/A:1025492512915PubMedGoogle ScholarCrossref
35.
Northam  S, Knapp  TR.  The reliability and validity of birth certificates.  J Obstet Gynecol Neonatal Nurs. 2006;35(1):3-12. doi:10.1111/j.1552-6909.2006.00016.xPubMedGoogle ScholarCrossref
36.
DeFranco  EA, Lian  M, Muglia  LA, Schootman  M.  Area-level poverty and preterm birth risk: a population-based multilevel analysis.  BMC Public Health. 2008;8(1):316. doi:10.1186/1471-2458-8-316PubMedGoogle ScholarCrossref
37.
Wehby  GL, Lyu  W.  The impact of the ACA Medicaid expansions on health insurance coverage through 2015 and coverage disparities by age, race/ethnicity, and gender.  Health Serv Res. 2018;53(2):1248-1271. doi:10.1111/1475-6773.12711PubMedGoogle ScholarCrossref
38.
Sommers  BD, Maylone  B, Blendon  RJ, Orav  EJ, Epstein  AM.  Three-year impacts of the Affordable Care Act: improved medical care and health among low-income adults.  Health Aff (Millwood). 2017;36(6):1119-1128. doi:10.1377/hlthaff.2017.0293PubMedGoogle ScholarCrossref
Original Investigation
April 23/30, 2019

Association of State Medicaid Expansion Status With Low Birth Weight and Preterm Birth

Author Affiliations
  • 1Department of Health Policy and Management, University of Arkansas for Medical Sciences, Little Rock
  • 2Institute for Medicaid Innovation, Washington, DC
  • 3Department of Obstetrics and Gynecology, University of Michigan Medical School, Ann Arbor
  • 4Department of Obstetrics and Gynecology, University of Arkansas for Medical Sciences, Little Rock
JAMA. 2019;321(16):1598-1609. doi:10.1001/jama.2019.3678
Key Points

Question  Was state Medicaid expansion associated with differences in rates of low birth weight and preterm birth, both overall and in terms of racial/ethnic disparities?

Findings  In this observational study of 15 631 174 births from 2011 to 2016, state Medicaid expansion was not significantly associated with differences in rates of low birth weight or preterm birth outcomes overall. There were significant reductions in relative disparities for black infants (preterm birth: −0.43 percentage points, very preterm birth: −0.14 percentage points, low birth weight: −0.53 percentage points, and very low birth weight: −0.13 percentage points) but not for Hispanic infants relative to white infants in states that expanded Medicaid compared with those that did not.

Meaning  State Medicaid expansion was not associated with differences in rates of low birth weight or preterm birth outcomes overall, but was associated with improvements in relative disparities for black infants compared with white infants among the states that expanded compared with those that did not.

Abstract

Importance  Low birth weight and preterm birth are associated with adverse consequences including increased risk of infant mortality and chronic health conditions. Black infants are more likely than white infants to be born prematurely, which has been associated with disparities in infant mortality and other chronic conditions.

Objective  To evaluate whether Medicaid expansion was associated with changes in rates of low birth weight and preterm birth outcomes, both overall and by race/ethnicity.

Design, Setting, and Participants  Using US population-based data from the National Center for Health Statistics Birth Data Files (2011-2016), difference-in-differences (DID) and difference-in-difference-in-differences (DDD) models were estimated using multivariable linear probability regressions to compare birth outcomes among infants in Medicaid expansion states relative to non–Medicaid expansion states and changes in relative disparities among racial/ethnic minorities for singleton live births to women aged 19 years and older.

Exposures  State Medicaid expansion status and racial/ethnic category.

Main Outcomes and Measures  Preterm birth (<37 weeks’ gestation), very preterm birth (<32 weeks’ gestation), low birth weight (<2500 g), and very low birth weight (<1500 g).

Results  The final sample of 15 631 174 births (white infants: 8 244 924, black infants: 2 201 658, and Hispanic infants: 3 944 665) came from the District of Columbia and 18 states that expanded Medicaid (n = 8 530 751) and 17 states that did not (n = 7 100 423). In the DID analyses, there were no significant changes in preterm birth in expansion relative to nonexpansion states (preexpansion to postexpansion period, 6.80% to 6.67% [difference: −0.12] vs 7.86% to 7.78% [difference: −0.08]; adjusted DID: 0.00 percentage points [95% CI, −0.14 to 0.15], P = .98), very preterm birth (0.87% to 0.83% [difference: −0.04] vs 1.02% to 1.03% [difference: 0.01]; adjusted DID: −0.02 percentage points [95% CI, −0.05 to 0.02], P = .37), low birth weight (5.41% to 5.36% [difference: −0.05] vs 6.06% to 6.18% [difference: 0.11]; adjusted DID: −0.08 percentage points [95% CI, −0.20 to 0.04], P = .20), or very low birth weight (0.76% to 0.72% [difference: −0.03] vs 0.88% to 0.90% [difference: 0.02]; adjusted DID: −0.03 percentage points [95% CI, −0.06 to 0.01], P = .14). Disparities for black infants relative to white infants in Medicaid expansion states compared with nonexpansion states declined for all 4 outcomes, indicated by a negative DDD coefficient for preterm birth (−0.43 percentage points [95% CI, −0.84 to −0.02], P = .05), very preterm birth (−0.14 percentage points [95% CI, −0.26 to −0.02], P = .03), low birth weight (−0.53 percentage points [95% CI, −0.96 to −0.10], P = .02), and very low birth weight (−0.13 percentage points [95% CI, −0.25 to −0.01], P = .04). There were no changes in relative disparities for Hispanic infants.

Conclusions and Relevance  Based on data from 2011-2016, state Medicaid expansion was not significantly associated with differences in rates of low birth weight or preterm birth outcomes overall, although there were significant improvements in relative disparities for black infants compared with white infants in states that expanded Medicaid vs those that did not.

Introduction

Quiz Ref IDPrematurity and low birth weight were estimated to contribute to 36% of infant mortality (in 2013)1 and have been associated with increased risk of chronic conditions throughout infancy and into adulthood.2-5 Rates of low birth weight (8.3% as of 2017) and prematurity (9.9%) are higher in the United States than most developed nations,6-9 with non-Hispanic black infants 2.0 times as likely to be born at low birth weight (13.9% vs 7.0%) and 1.5 times as likely to be born prematurely (13.9% vs 9.1%) compared with non-Hispanic white infants.9 Hispanic infants have rates of low birth weight (7.4% as of 2017) and prematurity (9.6%) similar to white infants.9

Since 1990, states have been required to provide Medicaid coverage to low-income pregnant women with family incomes up to 133% of the federal poverty level under the Omnibus Budget Reconciliation Act.10 Because this insurance is dependent on pregnancy, a mother will lose coverage 60 days postpartum if she does not meet the poverty threshold for Medicaid in her state, resulting in frequent transition in and out of Medicaid coverage (“insurance churning”) for many low-income women.11 Under the Affordable Care Act (ACA), states may expand Medicaid to adults with household incomes at or below 138% of the federal poverty level, which may improve continuity of access to health care for low-income women regardless of pregnancy status, potentially improving maternal health and earlier use of prenatal care.11,12 Evaluation of the association between ACA Medicaid expansion and infant outcomes is limited, with one study finding reductions in infant mortality in Medicaid expansion relative to nonexpansion states.13

The objective of this study was to evaluate the association between Medicaid expansion and rates of low birth weight and prematurity overall and among racial/ethnic minorities relative to non-Hispanic white (hereafter “white”) infants. It was hypothesized that adverse outcomes would be reduced in expansion relative to nonexpansion states with greater declines among non-Hispanic black (hereafter “black”) infants given greater increases in insurance coverage post-ACA and greater rates of adverse birth outcomes among black infants.9,14

Methods

The University of Arkansas for Medical Sciences Institutional Review Board deemed this study as non–human subjects research, and informed consent was waived.

Data Sources

The primary data source was the National Center for Health Statistics (NCHS) Vital Statistics Birth Data Files (2011-2016), which contain data abstracted from an estimated 99% of live births from all 50 states and the District of Columbia.15 The Birth Data Files were linked with 2 secondary data sources (2011-2016) at the state and year levels (see eMethods in the Supplement for exceptions): (1) area-level demographics and access to health care services came from weighted averages of county-level data from the Health Resources and Services Administration’s Area Health Resource File database and (2) state-level health outcomes came from the Centers for Disease Control and Prevention’s Behavioral Risk Factor Surveillance System Prevalence Data.

Study Sample
Included States

The standardized birth certificate data collection process and collected data were changed in 2003, with states varying in their adoption of the 2003 revision of the US Standard Certificate of Life Birth.15 Births from states that did not use the 2003 version by January 1, 2013 (at least 1 year prior to Medicaid expansion), were excluded. Among states that adopted the revised version in 2012 or 2013, births were excluded if they occurred in a year prior to the state’s first full year of adoption. States were additionally excluded if they used a Section 1115 waiver for Medicaid expansion. See the Supplement for state classifications (eTable 1 in the Supplement) and dates of expansion and 2003 revised birth certificate adoption (eTable 2 in the Supplement).

Sample of Births

Analyses were limited to births to women aged 19 years and older to ensure the woman was an adult (age ≥18) at the time of conception. Births with missing information on 1 or more variables were excluded (eFigure 1 in the Supplement). A χ2 test was conducted to test for a difference in the percentage of births that were excluded between Medicaid nonexpansion and expansion states.

Given the substantial disparities in adverse birth outcomes among racial/ethnic minorities, this study assessed changes in relative disparities by comparing outcomes in black infants and in Hispanic infants relative to white infants. Mutually exclusive racial/ethnic categories were created using the single-race variable and an indicator of Hispanic ethnicity. Quiz Ref IDThe NCHS recommends collecting race/ethnic information directly from the mother with predefined categories and the option to write in an “other” response. Following the National Vital Statistics Reports categorization, a woman who indicated any race in addition to Hispanic ethnicity was categorized as “Hispanic.”9 Analyses were additionally conducted separately for Medicaid-covered births and Medicaid-covered births to women aged 26 years and older with at most a high school degree. Medicaid funding and educational attainment were used as proxies for lower-income groups that may be more likely to benefit from expansion.

Independent Variables

The primary exposure variable was state Medicaid expansion. Previous studies vary on expansion state classification because states differed by expansion date.16-18 Twenty six states and the District of Columbia expanded Medicaid on January 1, 2014; 2 states expanded later in 2014; and 5 states expanded in 2015 or 2016.19 All states that expanded Medicaid in 2014-2016 were defined as expansion states regardless of previous expansion efforts, with the exception of the 8 states with approved Section 1115 waivers for Medicaid expansion by 2016,19 which were excluded from the study (eTable 1 in the Supplement). The month of expansion marked the beginning of the postexpansion period for later expanding states; a date of January 1, 2014, was used for nonexpansion states and for states that expanded on this date.

Dependent Variables

The dependent variables included preterm birth (<37 weeks’ gestation), very preterm birth (<32 weeks’ gestation), low birth weight (<2500 g), and very low birth weight (<1500 g). Gestation was based primarily on the clinical/obstetric estimate, which the NCHS uses as the standard for estimating gestational age.15

It was hypothesized that increased insurance coverage among low-income women may improve infant outcomes through mechanisms such as improved maternal health and earlier access to prenatal care services.12,20 Several exploratory mechanisms were tested including prenatal care initiation in the first trimester, prepregnancy diabetes, prepregnancy hypertension, cigarette use in the first trimester, any infection, and a measure of paternal acknowledgment.

Statistical Analysis

Descriptive statistics compared differences among births in Medicaid expansion and nonexpansion states overall and within each racial/ethnic group in the period prior to expansion. A difference-in-differences (DID) approach was used to compare changes in outcomes in expansion vs nonexpansion states, which can control for external secular trends by using nonexpansion states as a counterfactual.18,21,22 DID models were estimated including births of all races/ethnicities as well as for white, black, and Hispanic infants separately. A difference-in-difference-in-differences (DDD) approach was used to assess the change in birth outcomes for minority infants (ie, black and Hispanic infants, separately) in expansion states relative to nonexpansion states compared with the change in birth outcomes for white infants in expansion states relative to nonexpansion states.23,24 It was hypothesized that there would be greater reductions among minority infants, indicated by negative DDD coefficients.

DID and DDD models were estimated using multivariable linear probability models, with standard errors clustered at the state level and controlling for maternal, infant, and state-level factors (Table 1 and Table 2; see eMethods in the Supplement for a complete description) and for state-level dummy variables to control for any unobserved state-level effects that were time invariant. Linear probability models provided estimated percentage-point changes in the treatment (expansion) relative to the control (nonexpansion) states.25 Similar analyses were conducted to evaluate differential changes in the proposed mechanisms (eg, initiation of prenatal care in the first trimester) between Medicaid expansion and nonexpansion states using the same study sample and data as the primary analyses and can be found in eTable 8 in the Supplement.

An assumption of the DID and DDD approaches is that the rate of a given outcome among the treatment and comparison groups is parallel in the time prior to the treatment date.26 This assumption was tested by assessing interactions in the preexpansion period (2011-2013) between state expansion status and a year-based time variable for DID analyses and additionally with a binary racial/ethnic variable for DDD analyses. Line graphs were created to visualize outcomes over time.

This study included 2 sets of robustness checks. First, DID and DDD analyses were reestimated to include a 1-year lag after expansion. Second, different state exclusion criteria were tested by separately including Section 1115 waiver states, excluding states that expanded Medicaid after January 1, 2014, and excluding states that adopted the 2003 revised birth certificate in 2012 or 2013.

Analyses were completed in SAS version 9.4 (SAS Institute Inc). Visualizations were created using Tableau Software 2018.3 (Tableau). A 2-sided P ≤ .05 was considered statistically significant. Because of the potential for type 1 error due to multiple comparisons, findings for analyses of secondary end points should be interpreted as exploratory.

Results

Among the 17 319 894 singleton live births to women aged 19 years and older in eligible states, 9.75% of births (770 618/7 903 838) from the 17 nonexpansion states and 9.75% of births (918 102/9 416 056) from the District of Columbia and 18 expansion states were excluded due to missing information (χ2 test for difference in proportions P = .97), resulting in a final sample of 15 631 174 births (eFigure 1 in the Supplement). Births from 15 states were excluded: 7 states that did not use the 2003 version by 2013, 6 states that used a Section 1115 waiver, and 2 states that met both exclusion criteria. The racial/ethnic subgroups included 8 244 924 births to white women, 2 201 658 births to black women, and 3 944 665 births to Hispanic women.

Table 1 describes differences in state-level characteristics in the period prior to expansion. Relative to nonexpansion states, Medicaid expansion states had higher median incomes ($56 054 vs $49 489; difference: $6566 [95% CI, $3812-$9319]; P < .001), a lower percentage of uninsured females (15.7% vs 19.9%; difference: −4.2 percentage points [95% CI, −6.3 to −2.1]; P < .001), a higher percentage of minority females (35.1% vs 28.2%; difference: 6.9 percentage points [95% CI, 1.2-12.6]; P = .02), a higher number of primary care physicians per 1000 (0.82 vs 0.68; difference: 0.14 [95% CI, 0.10-0.19]; P < .001), and a higher percentage of adults who had their cholesterol checked in the last 5 years (79.6% vs 77.4%; difference: 2.3 percentage points [95% CI, 0.9-3.9]; P = .002).

Table 2 describes maternal and infant characteristics pooled among all races/ethnicities (“overall”) and among white, black, and Hispanic births in the period prior to expansion. Medicaid expansion and nonexpansion states differed with respect to characteristics that are associated with birth outcomes. Expansion states had a higher percentage of births covered by private insurance (50.7% vs 43.9%; difference: 6.8 percentage points [95% CI, 6.7-6.9]), a lower percentage of births to mothers who had attained high school or less education (38.9% vs 41.2%; difference: −2.4 percentage points [95% CI, −2.5 to −2.3]), and a higher percentage of births with prenatal care initiated during the first trimester (77.6% vs 71.9%; difference: 5.7 percentage points [95% CI, 5.7-5.8]).

The leftmost columns of Table 3 provide the unadjusted rates and percentage-point changes in outcomes by state expansion status and pre/postexpansion date, stratified vertically by racial/ethnic group. Among births of all races/ethnicities, adverse outcomes in expansion states declined for preterm birth (6.80% to 6.67%; difference: −0.12 [95% CI, −0.16 to −0.09]), very preterm birth (0.87% to 0.83%; difference: −0.04 [95% CI, −0.05 to −0.02]), low birth weight (5.41% to 5.36%; difference: −0.05 [95% CI, −0.08 to −0.02]), and very low birth weight (0.76% to 0.72%; difference: −0.03 [95% CI, −0.04 to −0.02]). Changes in outcomes in nonexpansion states were not significant for 1 outcome (very preterm birth: 1.02% to 1.03%; difference: 0.01 [95% CI, −0.01 to 0.02]), decreased for 1 outcome (preterm birth: 7.86% to 7.78%; difference: −0.08 [95% CI, −0.12 to −0.04]), and increased for 2 outcomes (low birth weight: 6.06% to 6.18%; difference: 0.11 [95% CI, 0.08-0.15] and very low birth weight: 0.88% to 0.90%; difference: 0.02 [95% CI, 0.00-0.03]). Relative percentage changes for all subgroups can be found in eFigures 2, 3, 4, and 5 in the Supplement.

In the adjusted DID analyses, which describe the association between Medicaid expansion and rates of birth outcomes, negative DID estimates would indicate a greater decline in the rate of preterm birth in expansion relative to nonexpansion states. The adjusted DID estimate for all 4 outcomes was not statistically significant, indicating no association of Medicaid expansion with preterm birth (0.00 percentage points [95% CI, −0.14 to 0.15], P = .98), very preterm birth (−0.02 percentage points [95% CI, −0.05 to 0.02], P = .37), low birth weight (−0.08 percentage points [95% CI, −0.20 to 0.04], P = .20), or very low birth weight (−0.03 percentage points [95% CI, −0.06 to 0.01], P = .14) (Table 3).

Table 3 additionally provides unadjusted and adjusted DDD estimates for black and Hispanic infants, which indicate whether changes in outcomes in expansion vs nonexpansion states differed when comparing black or Hispanic infants with white infants. A significant DDD coefficient indicates a change in the relative disparity for that minority group relative to white infants. Disparities for black infants relative to white infants in Medicaid expansion states compared with nonexpansion states declined for all 4 outcomes, indicated by a negative adjusted DDD coefficient for preterm birth (−0.43 percentage points [95% CI, −0.84 to −0.02], P = .05), very preterm birth (−0.14 percentage points [95% CI, −0.26 to −0.02], P = .03), low birth weight (−0.53 percentage points [95% CI, −0.96 to −0.10], P = .02), and very low birth weight (−0.13 percentage points [95% CI, −0.25 to −0.01], P = .04) (Table 3). There were no changes in relative disparities for Hispanic infants, indicated by no statistically significant DDD estimates for outcomes among Hispanic infants relative to white infants (Table 3).

The Figure displays trends in the 4 birth outcomes stratified by race/ethnicity and by Medicaid expansion status. This Figure was used to visualize trends across the study period as well as whether trends were parallel prior to the expansion date, which was formally tested using linear indicator variables in the DID and DDD models. The DDD comparison of relative disparities in rates of low birth weight between black and white infants as well as 2 DID comparisons among outcomes in black infants (preterm birth and low birth weight) failed the parallel trends test; however, all other significant DID and DDD comparisons in the primary analyses passed this test (Table 3 and Table 4). Line charts for rates pooled among all races/ethnicities and for the subgroups are available in eFigures 6, 7, and 8 in the Supplement.

Table 4 provides the DID and DDD estimates for the 2 Medicaid-covered subgroups. Relative disparities among black infants were reduced for all 4 outcomes among Medicaid-covered births (preterm birth: −0.47 percentage points [95% CI, −0.81 to −0.12], P = .01; very preterm birth: −0.20 percentage points [95% CI, −0.36 to −0.03], P = .03; low birth weight: −0.68 percentage points [95% CI, −1.16 to −0.21], P = .008; and very low birth weight: −0.16 percentage points [95% CI, −0.32 to 0.00], P = .05) and 3 outcomes among Medicaid-covered births to women with at most a high school degree (very preterm birth: −0.38 percentage points [95% CI, −0.62 to −0.13], P = .005; low birth weight: −1.01 percentage points [95% CI, −1.55 to −0.47], P = .001; and very low birth weight: −0.36 percentage points [95% CI, −0.60 to −0.13], P = .004) (Table 4). One DID outcome among white infants in the Medicaid-covered births subgroup (very low birth weight) and 2 DID outcomes among black infants in both Medicaid-covered subgroups (very preterm birth and very low birth weight) failed the parallel trends test. There were no statistically significant DDD estimates among Hispanic infants in either subgroup (Table 4).

The primary findings from this study were quantitatively similar to those from robustness checks including a 1-year lag after expansion (eTables 3 and 4 in the Supplement) and changing state exclusion criteria (eTables 5, 6, and 7 in the Supplement). There was no evidence that differential changes in specific mechanisms, such as prenatal care initiation in the first trimester or prepregnancy diabetes, explained the reduction in disparities between black and white infants in the study (eTable 8 in the Supplement).

Discussion

Quiz Ref IDThis study evaluated whether Medicaid expansion was associated with changes in rates of preterm birth and low birth weight, and found that there were no differences in the change in birth outcomes between expansion and nonexpansion. However, the study found greater reductions in rates of low birth weight and preterm birth outcomes among black infants in expansion states relative to white infants in expansion states, with no significant change in disparities among Hispanic infants.

Quiz Ref IDBlack infants die of complications associated with prematurity and low birth weight at 3.9 times the rate of white infants (257.6 vs 66.3 per 100 000 live births in 2016).27 Reducing the rates of low birth weight and preterm birth among black infants has been suggested as the primary mechanism for decreasing inequities in infant mortality.28 A previous evaluation of infant mortality in Medicaid expansion relative to nonexpansion states found a greater decline among black infants compared with white or Hispanic infants.13 The reduction in relative disparities in low birth weight and prematurity among black infants in expansion states found in the current study provide a potential explanation for the greater decline in infant mortality for black infants.

An evaluation of the association of Massachusetts health care reform with infant mortality and low birth weight found no improvement.29 The findings from the Massachusetts study29 may differ because the percentage of black individuals in many Medicaid expansion states, such as the District of Columbia or Louisiana (46% and 32%, respectively, in 2017) is higher than the percentage in Massachusetts (7%).30 Evaluations of the Medicaid expansions in the 1980s and 1990s found improvements in prenatal care use and mixed evidence regarding improvements in infant outcomes.31Quiz Ref ID The findings here suggest that earlier and continual access to insurance coverage may provide an important opportunity for improving infant outcomes. Because low birth weight and premature births are highly related to complex medical comorbidities throughout childhood and into adulthood, the implications of the improvements associated with Medicaid expansion could potentially be associated with reduced disparities in chronic conditions across the life course.28,32,33

Despite the reductions in disparities for black infants in expansion states, this study did not find improvements for Hispanic infants. While some DDD coefficients for Hispanic infants were positive, these findings were not significant, and many failed to meet the parallel trends assumption. Reasons for the significant increases in unadjusted rates among Hispanic infants in expansion states remain unclear.

Limitations

This study has several limitations. First, the Birth Data Files database consists of administrative data, which may have missing data or lack information on maternal factors potentially related to birth outcomes. Because records with missing data were excluded from the study, missing data could result in selection bias; however, there was no significant difference in the percentage of births excluded from Medicaid expansion and nonexpansion states. Additional variables, such as family income, may provide more accurate subgroups or covariates in regression models. Birth certificate information is collected across multiple settings, with the potential for information misclassification; however, this was mitigated with clustered standard errors and state dummy variables in DID and DDD comparisons. Studies have shown high validity and reliability for birth weight and moderate reliability for gestational age.34,35 To address the concerns of gestational age estimation, this study additionally included very preterm birth, which may be less sensitive to misclassification,36 as well as birth weight, which is not sensitive to the gestational age estimates.

Second, it may take time for the benefits of health insurance coverage to improve maternal health and access to care. Previous studies have found greater improvements in later years after Medicaid expansion37,38; however, including a 1-year lag after expansion (eTables 3 and 4 in the Supplement) did not meaningfully change the overall conclusions of this study.

Third, parallel trends between Medicaid expansion and nonexpansion states are required for valid conclusions from DID or DDD analyses. In the primary analyses reported in this study, 1 statistically significant DDD finding and 2 significant DID findings in the subpopulation of black infants failed to meet the parallel trends test and should be interpreted with caution.

Conclusions

Based on data from 2011-2016, state Medicaid expansion was not significantly associated with differences in rates of low birth weight or preterm birth outcomes overall, although there were significant improvements in relative disparities for black infants compared with white infants in states that expanded Medicaid vs those that did not.

Back to top
Article Information

Corresponding Author: J. Mick Tilford, PhD, Department of Health Policy and Management, University of Arkansas for Medical Sciences, 4301 W Markham, Slot 820-12, Little Rock, AR 72205 (tilfordmickj@uams.edu).

Accepted for Publication: March 13, 2019.

Author Contributions: Drs Brown and Tilford had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: All authors.

Acquisition, analysis, or interpretation of data: Brown, Moore, Bird, Tilford.

Drafting of the manuscript: Brown, Moore, Lowery, Tilford.

Critical revision of the manuscript for important intellectual content: Brown, Moore, Felix, Stewart, Bird, Tilford.

Statistical analysis: Brown, Moore, Bird.

Administrative, technical, or material support: Moore, Felix, Bird, Lowery.

Supervision: Bird, Tilford.

Conflict of Interest Disclosures: Drs Tilford and Stewart reported receiving support from the Arkansas Center for Health Disparities (ARCHD), which is supported by the National Institute on Minority Health and Health Disparities (NIMHD) of the National Institutes of Health (NIH), and from the Translational Research Institute (TRI), which is supported by the National Center for Advancing Translational Sciences of the NIH. Dr Tilford reported receiving copyright income from Trestle Tree Inc and personal fees from Roche. Dr Moore reported serving as the executive director of the Institute for Medicaid Innovation. Dr Lowery reported serving on the board of directors and a having financial interest in 2 private companies, Angel Eye Camera Systems LLC and Air Toco Uterine Monitoring System, and receiving personal fees from General Electric. No other disclosures were reported.

Funding/Support: Drs Tilford and Stewart were supported by grants from the ARCHD, which is supported by the NIMHD of the NIH (award ID 5U54MD002329), and from the TRI (award ID U54TR001629) through the NCATS of the NIH.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

References
1.
Matthews  TJ, MacDorman  MF, Thoma  ME.  Infant mortality statistics from the 2013 period linked birth/infant death data set.  Natl Vital Stat Rep. 2015;64(9):1-30.PubMedGoogle Scholar
2.
Behrman  RE, Butler  AS; Institute of Medicine, Committee on Understanding Premature Birth and Assuring Healthy Outcomes.  Preterm Birth: Causes, Consequences, and Prevention. Washington, DC: National Academies Press; 2007.
3.
Bhutta  AT, Cleves  MA, Casey  PH, Cradock  MM, Anand  KJ.  Cognitive and behavioral outcomes of school-aged children who were born preterm: a meta-analysis.  JAMA. 2002;288(6):728-737. doi:10.1001/jama.288.6.728PubMedGoogle ScholarCrossref
4.
de Kieviet  JF, Piek  JP, Aarnoudse-Moens  CS, Oosterlaan  J.  Motor development in very preterm and very low-birth-weight children from birth to adolescence: a meta-analysis.  JAMA. 2009;302(20):2235-2242. doi:10.1001/jama.2009.1708PubMedGoogle ScholarCrossref
5.
D’Onofrio  BM, Class  QA, Rickert  ME, Larsson  H, Långström  N, Lichtenstein  P.  Preterm birth and mortality and morbidity: a population-based quasi-experimental study.  JAMA Psychiatry. 2013;70(11):1231-1240. doi:10.1001/jamapsychiatry.2013.2107PubMedGoogle ScholarCrossref
6.
Delnord  M, Hindori-Mohangoo  AD, Smith  LK,  et al; Euro-Peristat Scientific Committee.  Variations in very preterm birth rates in 30 high-income countries: are valid international comparisons possible using routine data?  BJOG. 2017;124(5):785-794. doi:10.1111/1471-0528.14273PubMedGoogle ScholarCrossref
7.
MacDorman  MF, Matthews  TJ, Mohangoo  AD, Zeitlin  J.  International comparisons of infant mortality and related factors: United States and Europe, 2010.  Natl Vital Stat Rep. 2014;63(5):1-6.PubMedGoogle Scholar
8.
Murphy  SL, Mathews  TJ, Martin  JA, Minkovitz  CS, Strobino  DM.  Annual Summary of Vital Statistics: 2013-2014.  Pediatrics. 2017;139(6):e20163239. doi:10.1542/peds.2016-3239PubMedGoogle ScholarCrossref
9.
Martin  JA, Hamilton  BE, Osterman  MJK, Driscoll  AK, Drake  P.  Births: final data for 2017.  Natl Vital Stat Rep. 2018;67(8):1-50.PubMedGoogle Scholar
10.
Johnson  K, Applegate  M, Gee  RE.  Improving Medicaid: three decades of change to better serve women of childbearing age.  Clin Obstet Gynecol. 2015;58(2):336-354. doi:10.1097/GRF.0000000000000115PubMedGoogle ScholarCrossref
11.
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. doi:10.1377/hlthaff.2016.1241PubMedGoogle ScholarCrossref
12.
Shah  SI, Brumberg  HL.  Predictions of the affordable care act’s impact on neonatal practice.  J Perinatol. 2016;36(8):586-592. doi:10.1038/jp.2016.93PubMedGoogle ScholarCrossref
13.
Bhatt  CB, Beck-Sagué  CM.  Medicaid expansion and infant mortality in the United States.  Am J Public Health. 2018;108(4):565-567. doi:10.2105/AJPH.2017.304218PubMedGoogle ScholarCrossref
14.
McMorrow  S, Long  SK, Kenney  GM, Anderson  N.  Uninsurance disparities have narrowed for black and Hispanic adults under the Affordable Care Act.  Health Aff (Millwood). 2015;34(10):1774-1778. doi:10.1377/hlthaff.2015.0757PubMedGoogle ScholarCrossref
15.
National Center for Health Statistics. NCHS’ Vital Statistics Natality Birth Data. http://www.nber.org/data/vital-statistics-natality-data.html. Accessed October 1, 2018.
16.
Kaestner  R, Garrett  B, Chen  J, Gangopadhyaya  A, Fleming  C.  Effects of ACA Medicaid expansions on health insurance coverage and labor supply.  J Policy Anal Manage. 2017;36(3):608-642. doi:10.1002/pam.21993PubMedGoogle ScholarCrossref
17.
Simon  K, Soni  A, Cawley  J.  The impact of health insurance on preventive care and health behaviors: evidence from the first two years of the ACA Medicaid expansions.  J Policy Anal Manage. 2017;36(2):390-417. doi:10.1002/pam.21972PubMedGoogle ScholarCrossref
18.
Courtemanche  Ch, Marton  J, Ukert  B, Yelowitz  A, Zapata  D.  Early impacts of the Affordable Care Act on health insurance coverage in Medicaid expansion and non-expansion states.  J Policy Anal Manage. 2017;36(1):178-210. doi:10.1002/pam.21961PubMedGoogle ScholarCrossref
19.
Kaiser Family Foundation. State health facts. Status of state action on the Medicaid expansion decision. https://www.kff.org/health-reform/state-indicator/state-activity-around-expanding-medicaid-under-the-affordable-care-act/. Accessed February 1, 2019.
20.
Armstrong  J.  Women’s Health in the age of patient protection and the Affordable Care Act.  Clin Obstet Gynecol. 2015;58(2):323-335. doi:10.1097/GRF.0000000000000096PubMedGoogle ScholarCrossref
21.
Sommers  BD, Blendon  RJ, Orav  EJ, Epstein  AM.  Changes in utilization and health among low-income adults after Medicaid expansion or expanded private insurance.  JAMA Intern Med. 2016;176(10):1501-1509. doi:10.1001/jamainternmed.2016.4419PubMedGoogle ScholarCrossref
22.
Antonisse  L, Garfield  R, Rudowitz  R, Artiga  S. The effects of Medicaid expansion under the ACA: updated findings from a literature review. Kaiser Family Foundation. https://collections.nlm.nih.gov/master/borndig/101707525/Issue-Brief-The-Effects-of-Medicaid-Expansion-Under-the-ACA-Updated-Findings.pdf. Published February 2017. Accessed October 1, 2018.
23.
Gonzales  S, Sommers  BD.  Intra-ethnic coverage disparities among Latinos and the effects of health reform.  Health Serv Res. 2018;53(3):1373-1386. doi:10.1111/1475-6773.12733PubMedGoogle ScholarCrossref
24.
Yue  D, Rasmussen  PW, Ponce  NA.  Racial/ethnic differential effects of Medicaid expansion on health care access.  Health Serv Res. 2018;53(5):3640-3656. doi:10.1111/1475-6773.12834PubMedGoogle ScholarCrossref
25.
Karaca-Mandic  P, Norton  EC, Dowd  B.  Interaction terms in nonlinear models.  Health Serv Res. 2012;47(1, pt 1):255-274. doi:10.1111/j.1475-6773.2011.01314.xPubMedGoogle ScholarCrossref
26.
Ryan  AM, Burgess  JF  Jr, Dimick  JB.  Why we should not be indifferent to specification choices for difference-in-differences.  Health Serv Res. 2015;50(4):1211-1235. doi:10.1111/1475-6773.12270PubMedGoogle ScholarCrossref
27.
Xu  J, Murphy  SL, Kochanek  KD, Bastian  B, Arias  E.  Deaths: final data for 2016.  Natl Vital Stat Rep. 2018;67(5):1-76.PubMedGoogle Scholar
28.
Riddell  CA, Harper  S, Kaufman  JS.  Trends in differences in US mortality rates between black and white infants.  JAMA Pediatr. 2017;171(9):911-913. doi:10.1001/jamapediatrics.2017.1365PubMedGoogle ScholarCrossref
29.
Boudreaux  MH, Dagher  RK, Lorch  SA.  The association of health reform and infant health: evidence from Massachusetts.  Health Serv Res. 2017;53(4):2406-2424. doi:10.1111/1475-6773.12779PubMedGoogle ScholarCrossref
30.
Henry J. Kaiser Family Foundation. Population distribution by race/ethnicity. https://www.kff.org/other/state-indicator/distribution-by-raceethnicity/. Published 2017. Accessed October 1, 2018.
31.
Howell  EM.  The impact of the Medicaid expansions for pregnant women: a synthesis of the evidence.  Med Care Res Rev. 2001;58(1):3-30. doi:10.1177/107755870105800101PubMedGoogle ScholarCrossref
32.
Lu  MC, Kotelchuck  M, Hogan  V, Jones  L, Wright  K, Halfon  N.  Closing the black-white gap in birth outcomes: a life-course approach.  Ethn Dis. 2010;20(1)(suppl 2):S2-S62, 76.PubMedGoogle Scholar
33.
Braveman  P, Barclay  C.  Health disparities beginning in childhood: a life-course perspective.  Pediatrics. 2009;124(suppl 3):S163-S175. doi:10.1542/peds.2009-1100DPubMedGoogle ScholarCrossref
34.
Roohan  PJ, Josberger  RE, Acar  J, Dabir  P, Feder  HM, Gagliano  PJ.  Validation of birth certificate data in New York State.  J Community Health. 2003;28(5):335-346. doi:10.1023/A:1025492512915PubMedGoogle ScholarCrossref
35.
Northam  S, Knapp  TR.  The reliability and validity of birth certificates.  J Obstet Gynecol Neonatal Nurs. 2006;35(1):3-12. doi:10.1111/j.1552-6909.2006.00016.xPubMedGoogle ScholarCrossref
36.
DeFranco  EA, Lian  M, Muglia  LA, Schootman  M.  Area-level poverty and preterm birth risk: a population-based multilevel analysis.  BMC Public Health. 2008;8(1):316. doi:10.1186/1471-2458-8-316PubMedGoogle ScholarCrossref
37.
Wehby  GL, Lyu  W.  The impact of the ACA Medicaid expansions on health insurance coverage through 2015 and coverage disparities by age, race/ethnicity, and gender.  Health Serv Res. 2018;53(2):1248-1271. doi:10.1111/1475-6773.12711PubMedGoogle ScholarCrossref
38.
Sommers  BD, Maylone  B, Blendon  RJ, Orav  EJ, Epstein  AM.  Three-year impacts of the Affordable Care Act: improved medical care and health among low-income adults.  Health Aff (Millwood). 2017;36(6):1119-1128. doi:10.1377/hlthaff.2017.0293PubMedGoogle ScholarCrossref
×