This brief focuses on the health insurance coverage experiences under the Affordable Care Act (ACA) of mothers living with dependent children. Using 1997–2014 data from the National Health Interview Survey, we examine trends in uninsurance among mothers nationally and for mothers in states that did and did not expand Medicaid under the ACA by May 2014. We also assess changes from 2013 to 2014 in the characteristics of uninsured mothers and uninsurance rates for various subgroups of mothers, as well as reasons for being uninsured among mothers without coverage in 2014.
This brief examines the coverage experiences under the Affordable Care Act (ACA) of mothers living with dependent children. Focusing on mothers is important because just over 90 percent of children live with their mother, including about one-third who live with their mother and no other parent.1 Improving the health and well-being of mothers may indirectly benefit children—for example, reducing maternal depression could improve parenting behaviors and children’s health and development outcomes (Cummings and Kouros 2009; Lovejoy et al. 2000).
Though no published research focuses on coverage changes for mothers under the ACA, evidence is available on coverage patterns for all parents. In the years leading up to the passage of the ACA in 2010, the share of parents without insurance coverage had been rising (Rosenbaum and Kenney 2014). Recent research shows, however, that the uninsurance rate for parents declined following the early 2014 implementation of the key coverage provisions of the ACA (Karpman et al. 2015; Karpman, Gates, and Kenney 2016). Previous studies have shown that coverage gains for parents are linked to improved coverage and health care access for their children (Davidoff et al. 2003; Dubay and Kenney 2003). In addition, pre-ACA Medicaid expansions of public coverage to parents and other adults have been shown to increase their own use of preventive care and other health services, reduce out-of-pocket medical costs and medical debt, and improve mental health (Finkelstein et al. 2012; McMorrow et al. 2016).
In this brief, we use data from the National Health Interview Survey (NHIS) to examine trends in uninsurance for mothers between 1997 and 2014, the most recent year for which NHIS data on parental status is available. We provide estimates of uninsurance among mothers nationally and for mothers in states that did and did not expand Medicaid under the ACA by May 2014, overall and among those with family incomes above and below 138 percent of the federal poverty level (FPL), which is the new Medicaid eligibility level established for adults under the ACA. We also assess changes between 2013 and 2014 in the characteristics and uninsurance rates for various subgroups of mothers.
Data and Methods
This analysis uses nationally representative data for the civilian noninstitutionalized population from the NHIS, which is conducted annually by the National Center for Health Statistics. The NHIS consists of three main sections: the family core, sample adult core, and sample child core. Questions on the family core are answered for each member of the family by a knowledgeable adult and include basic demographic information, educational attainment, employment status, general health status, and detailed information on health insurance coverage. Questions on the sample child and sample adult core are asked of one random child age 17 or under (if present; excluding emancipated minors) and one random adult over age 17 in each family; the questions solicit extensive information on health status and details on health care access, use, and affordability. The NHIS is fielded continuously throughout the year and can be used to provide nationally representative annual or quarterly estimates. We use public-use data from the 1997–2014 Integrated Health Interview Series, which provides harmonized versions of NHIS variables across years.2
We track uninsurance rates from 1997 to 2014 for mothers ages 19 to 64, overall and by income as a percentage of FPL.3 We use 1997 as a starting point because earlier estimates from the NHIS are not comparable because of changes in the survey that were implemented at that time. We construct health insurance units (HIUs), which better reflect the units used to determine health insurance eligibility than does the broader family definition available on the NHIS. Hereafter, we use the terms “family” and “HIU” interchangeably, but all analysis uses the HIU. Mothers are defined as women ages 19 to 64 who were identified as the mother of a child age 18 or under in their HIU. Based on this definition, in 2014 the NHIS included data on approximately 14,000 mothers from the family core file and 6,000 mothers from the sample adult file. Mothers are considered uninsured if they do not have coverage through private health insurance, Medicare, Medicaid, military health insurance, or other public insurance at the time of the survey. We also compare changes in uninsurance rates by state Medicaid expansion status between 2013 and 2014 for all mothers and for those with incomes above and below 138 percent of FPL. As of April 2016, 30 states and the District of Columbia had expanded Medicaid to parents and childless adults with incomes up to 138 percent of FPL. For this analysis, we focus on whether states expanded Medicaid as of May 2014, capturing coverage changes for mothers based on whether expanded Medicaid eligibility was available to them for the majority of the year.4
In addition to reporting long-term trends in uninsurance by HIU income, we examine changes between 2013 and 2014 in the composition of uninsured mothers and rates of uninsurance among mothers by age (19 to 25, 26 to 34, 35 to 49, or 50 to 64); race/ethnicity (non-Hispanic white, non-Hispanic black, non-Hispanic other race, or Hispanic); self-reported general health status (excellent or very good, good, or fair or poor); activity limitations; psychological distress5 (severe or moderate distress, or no or mild distress); citizenship status; HIU income (138 percent of FPL or less, above 138 percent and equal to or below 400 percent of FPL, or above 400 percent of FPL); work status (full-time worker, part-time worker, or not working); education level (less than high school, high school, some college, or college graduate); HIU citizenship (any noncitizen in HIU or no noncitizen in HIU); census region (Northeast, Midwest, South, or West); marital status (married; widowed, divorced, or separated; or never married); and whether the mother has one or more young children age 5 and under living in the household.
Finally, we examine the reasons for being uninsured among mothers without coverage. For this question, respondents could provide up to five reasons for being uninsured. Possible responses included: loss of a job or a change of employers by the person in the family with coverage; divorce, separation, or death of a spouse or parent; loss of eligibility because of age or leaving school; lack of coverage because it is not offered by an employer or the employee is not eligible; the cost is too high; an insurance company refused coverage; Medicaid or another medical plan stopped after pregnancy; loss of Medicaid/medical plan because of a new job or an increase in income; loss of Medicaid for other reasons; or another reason for not having coverage. When answering this question, many respondents are describing why they lost coverage years before the survey since many have been uninsured for more than 12 months.
Changes in Uninsurance among Mothers
Uninsurance rates for mothers increased 2.2 percentage points between 1997 and 2013, and by 2013 almost one in five (19.5 percent) mothers was uninsured (figure 1). In contrast, between 2013 and 2014, the uninsurance rate for mothers fell 3.8 percentage points, a decline that was nearly three times as large as any prior year-to-year change since 1997. Moreover, the uninsurance rate for mothers fell from 17.4 percent to 13.6 percent between the first and last quarter of 2014 (data not shown); the 13.6 percent uninsurance rate in the last quarter of 2014 is by far the lowest rate observed since 1997. Though some of the decline between 2013 and 2014 may have been because of the improving economy, the uninsurance rate for mothers had fallen only 0.7 percentage points between its peak in 2010 (20.2 percent) and 2013 (19.5 percent), just before implementation of the ACA’s major coverage provisions.
Figure 2 shows that uninsurance fell among mothers with family incomes above and below 138 percent of FPL between 1997 and 2014. Among low-income mothers, the uninsurance rate fell from 39.4 percent in 1997 to 37.5 percent in 2013 and then dropped an additional 6.2 percentage points between 2013 and 2014. Mothers in moderate-income families (i.e., those with incomes above 138 percent and equal to or below 400 percent of FPL) saw a steady rise in uninsurance between 1997 and 2013 (from 14.8 percent to 19.0 percent), before their uninsurance rate fell sharply in 2014 (to 14.2 percent that year). Uninsurance rates remained under 4 percent for higher-income mothers with incomes above 400 percent of FPL over the study period. In 2014, low-income mothers remained over twice as likely to be uninsured as moderate-income mothers and almost 15 times as likely to be uninsured as higher-income mothers.
Despite having lower uninsurance rates to start, the decline in uninsurance between 2013 and 2014 was greater among mothers living in Medicaid expansion states than those living in nonexpansion states.6 By 2014, the uninsurance rate for mothers in nonexpansion states was almost 1.7 times the rate in states that expanded Medicaid (19.9 percent versus 11.8 percent; figure 3). Figure 3 shows that the varying changes in coverage between expansion and nonexpansion states are largely caused by different trends among low-income mothers. In expansion states, the uninsurance rate for mothers with incomes at or below 138 percent of FPL fell 8.1 percentage points during this period, from 31.3 percent to 23.2 percent; in nonexpansion states, the uninsurance rate for low-income mothers fell 4.9 percentage points. Changes in uninsurance among moderate- and higher-income mothers were similar in each state group: 3.4 percentage points in expansion states and 2.6 percentage points in nonexpansion states.
Overall, approximately 1.6 million mothers gained coverage between 2013 and 2014 as the number of uninsured mothers fell from 7.5 million to 5.9 million (figure 4). Nearly 700,000 mothers with incomes at or below 138 percent of FPL and over 850,000 with incomes above that level gained coverage during this period. In 2014, mothers with incomes at or below 138 percent of FPL constituted 61 percent of remaining uninsured mothers.
Composition of Mothers Remaining Uninsured in 2014
The characteristics of uninsured mothers shifted between 2013 and 2014. A greater proportion of the remaining uninsured were noncitizens in 2014 than in 2013 (39.3 percent versus 33.5 percent), potentially reflecting the restrictive immigration requirements associated with the ACA’s new coverage options or a lack of awareness of new coverage options among lawfully present noncitizens (table 1).
Nearly half (47.4 percent) of mothers who were uninsured in 2014 were Hispanic, and more than two-fifths of uninsured mothers had at least one noncitizen in the family. Almost half (47.6 percent) of uninsured mothers were not working. However, only 20.7 percent were in families with no workers, and this did not change significantly from 2013 (data not shown). The share of uninsured mothers living in the South increased; the share of those living in the West declined. In terms of marital status and family structure, unmarried mothers (including those who were never married and those who were widowed, separated, or divorced) constituted more than 40 percent of the uninsured in 2014, and nearly half of mothers who were uninsured had a child age 5 or under.
When asked about the reasons why they do not have health insurance or stopped having coverage, 41.5 percent of uninsured mothers in 2014 said it was because the cost was too high (figure 5). The next most frequently reported reason was that coverage stopped after pregnancy (18.7 percent). In addition, some mothers identified their inability to get employer-based coverage as a reason for being uninsured, either because they or the person in their family with coverage lost or changed jobs (18.0 percent) or because their current employer does not offer coverage (7.3 percent). Over 5 percent said they had lost Medicaid or other coverage because of a new job or increase in income, and 18.6 percent reported not having coverage for other reasons, such as divorce, separation, or death of a spouse or parent, becoming ineligible because of age or leaving school, denial of coverage from an insurance company, or not needing coverage.
Variation in Uninsurance Rates among Mothers
Table 2 shows that between 2013 and 2014, the uninsurance rate fell for mothers in every subgroup examined except for those with incomes above 400 percent of FPL, those with college degrees, and those in the Northeast—groups that had lower uninsurance rates before implementation of the ACA’s major coverage provisions. The largest percentage-point declines in the uninsurance rate were found among young mothers ages 19 to 25 and 26 to 34 (4.7 percentage points for each age group), black and Hispanic mothers (4.7 and 6.9 percentage points, respectively), those living in the West (7.0 percentage points), and those without a college degree.
However, uninsurance rates remained disproportionately high in 2014 for certain groups of mothers, including those who were ages 19 to 25 (22.4 percent), Hispanic (34.3 percent), noncitizens (45.2 percent), living in the South (21.3 percent), lacking a high school education (38.8 percent), or low-income (31.3 percent). Mothers who had never married were more likely to be uninsured than mothers who were widowed, separated, or divorced and those who were married. In addition, mothers of young children were less likely to be insured than mothers who did not have young children in the household.
Uninsurance rates were highest among mothers who reported less than very good health and among those with moderate or severe psychological distress. However, mothers with activity limitations were about half as likely to be uninsured as those with no activity limitations, potentially reflecting broader Medicaid eligibility for people with disabilities than for those who do not have disabilities.
The number of uninsured mothers fell from 7.5 million in 2013 to 5.9 million in 2014, as the uninsurance rate for mothers reached its lowest point since 1997. The largest declines in uninsurance were found among low- and moderate-income mothers who were targeted by the ACA’s Medicaid expansion and the introduction of subsidized Marketplace coverage, respectively. Coverage gains among low-income mothers were larger in states that expanded Medicaid by May 2014 than in states that did not expand Medicaid, which drove the larger overall decline in uninsurance among mothers found in expansion states. Not all changes in coverage can be attributed to the Medicaid expansion and other ACA policy changes; other factors such as the improving economy likely played a role in the coverage gains between 2013 and 2014 as well.
Despite these gains in coverage, nearly one in six mothers remained uninsured in 2014, and these mothers were disproportionately young, low-income, Hispanic, noncitizens, less educated, not married, and living in the South. One particularly concerning finding is that about one in five mothers who were likely to have the greatest physical and mental health care needs—those who reported being in less than very good health or having moderate or severe psychological distress—were uninsured.
Around two in five uninsured mothers cited cost as a reason for not having coverage. Others mentioned disruptions in coverage following pregnancy, when Medicaid eligibility becomes significantly more restrictive, particularly in states that have not expanded Medicaid under the ACA. More than three in five uninsured mothers have incomes at or below 138 percent of FPL, suggesting that lack of eligibility for Medicaid, or lack of awareness among those who are eligible, is one of the primary barriers to connecting more mothers to coverage.
The decline in the uninsurance rate for mothers after 2013 coincides with a reduction in uninsurance among children in both low-income and moderate-income families, a trend that predated the ACA and persisted following implementation of the ACA’s major coverage provisions (Gates et al. 2016). Given that the ACA’s coverage expansions were primarily targeted toward adults and that many uninsured children were already eligible for Medicaid or CHIP before 2014, it is possible that an increase in coverage among parents has played a role in the continued progress in reducing children’s uninsurance rates. Future briefs will examine changes in coverage for all parents and differences in health care access, service use, and affordability between parents with and without health insurance and will track these outcomes beyond 2014.
- Authors’ tabulations of 2014 National Health Interview Survey data. Parents include biological parents, adoptive parents, and stepparents.↩
- “Integrated Health Interview Series: Version 6.11.” 2016, Minnesota Population Center, University of Minnesota and State Health Access Data Assistance Center, accessed April 7, 2016, http://www.ihis.us.↩
- Our analysis excludes 0.6 percent of mothers of children 18 and under who were under age 19 or over age 64.↩
- States expanding Medicaid as of May 2014 include Arizona, Arkansas, California, Colorado, Connecticut, Delaware, District of Columbia, Hawaii, Illinois, Iowa, Kentucky, Maryland, Massachusetts, Michigan, Minnesota, Nevada, New Jersey, New Mexico, New York, North Dakota, Ohio, Oregon, Rhode Island, Vermont, Washington, and West Virginia. New Hampshire expanded Medicaid in mid-August 2014, and four other states expanded Medicaid by April 2016 (Pennsylvania, Indiana, Alaska, and Montana).↩
- The measure of psychological distress used in this brief is based on mothers’ responses to questions in the sample adult file. Measures are based on scores on the Kessler Psychological Distress Scale (K6).↩
- The difference in the change in the uninsurance rate for expansion states relative to nonexpansion states between 2013 and 2014 was marginally significant (p < 0.10).↩
Cummings, E. Mark, and Chrystyna D. Kouros. 2009. Maternal Depression and Its Relation to Children’s Development and Adjustment.
Davidoff, Amy, Lisa Dubay, Genevieve M. Kenney, and Alshadye Yemane. 2003. “The Effect of Parents’ Insurance Coverage on Access to Care for Low-income Children.” Inquiry 40: 254–68.
Dubay, Lisa, and Genevieve M. Kenney. 2003. “Expanding Public Health Insurance to Parents: Effects on Children’s Coverage under Medicaid.” Health Services Research 38 (5): 1283–1301.
Gates, Jason A., Michael Karpman, Genevieve M. Kenney, and Stacey McMorrow. 2016. Uninsurance among Children: Long-Term Trends and Recent Patterns. Washington, DC: Urban Institute.
Finkelstein, Amy, Sarah Taubman, Bill Wright, Mira Bernstein, Jonathan Gruber, Joseph P. Newhouse, Heidi Allen, Katherine Baicker, and the Oregon Health Study Group. 2012. “The Oregon Health Insurance Experiment: Evidence from the First Year.” Quarterly Journal of Economics 127 (3): 1057–1106.
Karpman, Michael, Jason A. Gates, and Genevieve M. Kenney. 2016. Time for a Checkup: Changes in Health Insurance Coverage, Health Care Access and Affordability, and Plan Satisfaction among Parents and Children between 2013 and 2015. Washington, DC: Urban Institute.
Karpman, Michael, Genevieve M. Kenney, Nathaniel Anderson, and Sharon K. Long. 2015. QuickTake: The Gap in Uninsurance Rates for Parents and Children Narrowed between September 2013 and March 2015. Washington, DC: Urban Institute.
Lovejoy, M. Christine, Patricia A. Graczyk, Elizabeth O’Hare, and George Neuman. 2000. “Maternal Depression and Parenting Behavior: A Meta-Analytic Review.” Clinical Psychology Review 20 (5): 561–592.
McMorrow, Stacey, Genevieve M. Kenney, Sharon K. Long, and Dana E. Goin. 2016. “Medicaid Expansions from 1997 to 2009 Increased Coverage and Improved Access and Mental Health Outcomes for Low-Income Parents.” Health Services Research. doi: 10.1111/1475-6773.12432.
Rosenbaum, Sara, and Genevieve M. Kenney. 2014. “The Search for a National Child Health Coverage Policy.” Health Affairs 33 (12): 2125–35.
About the Authors
Michael Karpman is a research associate in the Health Policy Center at the Urban Institute. His work focuses on the implications of the Affordable Care Act, including quantitative analysis related to health insurance coverage, access to and affordability of health care, use of health care services, and health status. This work includes efforts to help coordinate and analyze data from the Urban Institute’s Health Reform Monitoring Survey. Before joining Urban in 2013, Karpman was a senior associate at the National League of Cities Institute for Youth, Education, and Families. He received his MPP from Georgetown University.
Jason A. Gates is a research assistant in the Health Policy Center at the Urban Institute. His current work focuses on the effects of expanding coverage on low income populations, children and families. His expertise is with the National Health Interview Survey, and he has experience analyzing the American Community Survey and Behavioral Risk Factor Surveillance System. He received his BA from Dickinson College.
Genevieve M. Kenney is a senior fellow and codirector of the Health Policy Center at the Urban Institute. She has been conducting policy research for over 25 years and is a nationally renowned expert on Medicaid, the Children's Health Insurance Program (CHIP), and broader health insurance coverage and health issues facing low-income children and families. Kenney has led a number of Medicaid and CHIP evaluations, and published over 100 peer-reviewed journal articles and scores of briefs on insurance coverage, access to care, and related outcomes for low-income children, pregnant women, and other adults. In her current research, she is examining implications of the Affordable Care Act, how access to primary care varies across states and insurance groups, and emerging policy questions related to Medicaid and CHIP. She received a master’s degree in statistics and a PhD in economics from the University of Michigan.
Stacey McMorrow is a health economist with extensive experience using quantitative methods to study the factors that affect individual health insurance coverage and access to care as well as the impacts of state and national health reforms on employers and individuals. Her current work uses the Affordable Care Act and past Medicaid expansions to explore the effects of expanding insurance coverage on access to care, service use and health outcomes for various populations. Through this and other work, McMorrow has developed substantial expertise in analyzing data from several federal surveys, including the National Health Interview Survey and the Medical Expenditure Panel Survey. Other research interests include the role of community health centers and safety net providers under health reform, receipt of preventive and reproductive health services among women, barriers to care for low-income children, and the market-level effects of insurance expansions. McMorrow received her PhD in health economics from the University of Pennsylvania in 2009.
This brief was funded by the David and Lucile Packard Foundation and an anonymous donor. We are grateful to them and to all our funders, who make it possible for Urban to advance its mission.
Funders do not determine research findings or the insights and recommendations of Urban experts. Further information on the Urban Institute’s funding principles is available at www.urban.org/support.
The authors are grateful to Patricia Barnes and the staff at the National Center for Health Statistics Research Data Center for their help with this study. The views expressed are those of the authors and should not be attributed to the Research Data Center, the National Center for Health Statistics, the Centers for Disease Control and Prevention, or to the Urban Institute, its trustees, or its funders. They are also grateful to Stephen Zuckerman for providing helpful feedback on an earlier draft.