Medicaid Expansion For Adults Had Measurable ‘Welcome Mat’ Effects On Their Children

Insurance coverage has increased dramatically for low-income Americans since the implementation of the Affordable Care Act (ACA) in 2014. ¹ This includes gains among low-income children, most of whom were already eligible for public coverage through Medicaid or the Children’s Health Insurance Program (CHIP) before the ACA’s passage in 2010. Between 2013 and 2015, the percentage of children younger than age eighteen who were uninsured in families with incomes below 100 percent and between 100 percent and 199 percent of the federal poverty level fell by 3.4 and 3.9 percentage points, respectively, according to the National Health Interview Survey (NHIS). ¹ While the decline in uninsurance for children was smaller in magnitude than the decline precipitated by the ACA for adults, coverage effects among low-income children in this period were notable because such children experienced very few changes in eligibility when ACA provisions were implemented in 2014. This suggests that the ACA potentially had important “spillover” or “welcome mat” effects, whereby new ACA policies led to an increase in public insurance among children who were already eligible for but not enrolled in public coverage.

Welcome-mat effects could exist among publicly eligible children for several reasons. The ACA was associated with significant outreach at both the federal and state levels to advertise new insurance affordability programs, to inform people about insurance coverage mandates (and penalties for those who do not obtain coverage), and to promote the benefits of enrolling in and maintaining insurance coverage over time. Large-scale marketing efforts could lead to increased enrollment among the already eligible population by decreasing the stigma of public coverage or by encouraging families to cover their children continuously, instead of waiting for a medical event. Outreach could also work to educate families who were previously unaware of their children’s eligibility for public insurance.

In addition, several policies were implemented as part of the ACA to reduce hurdles and to simplify application and eligibility determination processes. The ACA required states to adopt the use of Modified Adjusted Gross Income (MAGI) when determining eligibility for Medicaid, CHIP, and subsidies for private coverage through insurance Marketplaces. This removed the wide variation in income-counting methods used before the implementation of the ACA to determine eligibility across states and across programs within states. States also adopted “no wrong door” policies, under which eligibility is determined for each applicant for all three programs, regardless of which program they apply for. When combined with outreach efforts, these factors can work to increase enrollment of already eligible children by first attracting families to seek insurance coverage through Medicaid, CHIP, or the Marketplace and then by funneling children and their parents into the correct program or programs. Similar factors were credited for large increases in enrollment among Medicaid-eligible children when states implemented CHIP between 1997 and 2000. ^2,3

Finally, although the ACA did little to change eligibility for public programs for already eligible children, parents of such children were among the populations targeted for Medicaid expansions to adults and subsidized Marketplace coverage. In sixteen of the twenty-six states that expanded Medicaid coverage to adults, the Medicaid eligibility threshold of 138 percent of poverty implemented as part of the ACA was larger than the Medicaid threshold applicable to parents before the ACA, and in ten states it represented an increase of more than 50 percentage points. ⁴ Among low-income parents who remained ineligible for Medicaid under the ACA, many became eligible for subsidized Marketplace coverage. In the first year of the ACA, parents with dependent children accounted for a quarter of all adults gaining insurance coverage, with increases occurring in both public and private coverage. ⁵

Prior research has shown strong links between children’s health insurance coverage and their parents’ coverage and public insurance eligibility status. Some of these studies found that children were more likely to be insured when a parent was insured and that take-up of Medicaid among already eligible children increased during prior Medicaid expansions to parents. ^6–9 Children were also more likely to maintain Medicaid coverage over time if their parents were either enrolled in or eligible for the same public program. ^10,11

Considered together, these factors identify three potentially different sources for welcome-mat effects among already eligible children: general ACA effects associated with the vast rollout of new health policies (including the insurance mandate); expansion policy effects associated with the potential for additional investment in outreach and coordination that may have occurred in expansion states; and parental effects associated with joint parent-child eligibility for and coverage by Medicaid.

To date, much of the discussion on post-ACA coverage effects has focused on the first two sources above. We hypothesize, however, that when considering welcome-mat effects among children, it is equally important to consider the third. When the ACA was implemented, parents experienced a wide range of Medicaid eligibility pathways. Many parents gained eligibility for Medicaid in states that adopted the Medicaid expansion to adults, while many others living in nonexpansion states remained ineligible for Medicaid both before and after the ACA. Perhaps less well known, however, is that a sizable number of parents living in both expansion and nonexpansion states had been eligible for Medicaid even before the ACA and remained eligible after its implementation. Prior research found that public coverage increased among Medicaid-eligible children as a result of Medicaid expansions to parents in the 1990s. ^7,8 Those findings suggests that coverage effects for children under the ACA could differ depending on whether their parents gained Medicaid, had always been eligible for Medicaid, or had never been eligible.

Several studies have addressed welcome-mat effects in Medicaid coverage under the ACA. Descriptive work observing data trends has shown strong evidence of these effects, with the percentage of children who are uninsured at historic lows ¹² and greater improvements in coverage seen among low-income children and children living in expansion states. ^13–15 Two analytic studies provide evidence of welcome-mat effects under the ACA. The first studied the broader population of all nonelderly individuals and identified the relative impact that different ACA policies (such as the coverage mandate, Medicaid expansion, and subsidies for Marketplace coverage) had on coverage increases observed between 2012 and 2015. ¹⁶ The second focused on a more narrow population of publicly eligible children living in California counties selected for an early rollout of the ACA Medicaid expansion. ¹⁷ But to our knowledge, no analytic studies have measured ACA welcome-mat effects for the population of Medicaid-eligible children at the national level, identified the effect that ACA Medicaid expansions to adults who are parents has had on their children’s enrollment, or attempted to differentiate across the various sources of ACA welcome-mat effects.

This article fills that void in the literature, observing changes in public coverage before and after the implementation of the ACA for the population of children already eligible for Medicaid and identifying the sources of those changes in the first and second years of the program, while controlling for a wide set of factors relevant to enrollment decisions. As policy makers determine the future of US health policy, disentangling and quantifying these separate effects can help inform children’s health policy and budget decisions at both the state and federal levels.

Study Data And Methods

Data Sample

We used 2013–15 data from the American Community Survey (ACS), a nationally representative, cross-sectional survey of the US population containing rich information on individuals’ demographic and socioeconomic characteristics. Conducted by the Census Bureau, the ACS is the largest household survey in the United States and releases timely estimates, providing a unique advantage as we studied small subgroups of children across all states through 2015, the second year of ACA implementation. The analyses used ACS weights and balanced repeated replication for standard errors. All results reported in the text were significant at the 5 percent level or higher.

Our sample comprised nondisabled, citizen children ages 0–18 who were eligible for Medicaid (including Medicaid-expansion CHIP) in both the pre- and post-ACA periods. These “already eligible” children faced no changes in their own eligibility over the time period studied. Because a primary focus of our analysis was to measure the impact of the ACA’s adult Medicaid expansion on children’s coverage, we limited the sample to children in families with incomes below 138 percent of poverty with a parent present. We also excluded married minors, children with Medicare coverage, and children eligible for separate (as opposed to Medicaid-expansion) CHIP. Separate CHIP–eligible children below 138 percent of poverty in 2013 became eligible for CHIP-funded Medicaid under the ACA and may have exhibited different patterns of enrollment as they transitioned between the two programs. Our final sample contained 345,207 observations, representing 21.0 million children, or 84 percent of all children below 138 percent of poverty in the ACS. Additional details are available in the online Appendix. ¹⁸

Medicaid-eligible children were categorized into three mutually exclusive groups based on the Medicaid eligibility of their parents: “parent always Medicaid eligible,” “parent newly Medicaid eligible,” and “parent never Medicaid eligible.” Children in the “parent always Medicaid eligible” category had parents who were Medicaid eligible in both the pre- and post-ACA periods. “Parent newly Medicaid eligible” contained children whose parents were Medicaid eligible only in the post-ACA period. Finally, children whose parents were ineligible for Medicaid in both periods were defined as “parent never Medicaid eligible.” Medicaid eligibility for parents included eligibility through programs that predated the ACA (traditional Medicaid and Medicaid family expansions) as well as adult eligibility gained through the ACA Medicaid expansion.

The never eligible category includes a small number of children whose parents were eligible for Medicaid in the pre-ACA period and lost eligibility in the post-ACA period. It also includes parents who may have become eligible for subsidized Marketplace coverage under the ACA. Our results were not sensitive to alternative treatments of these cases. Each child’s expansion-state status was measured as of July 1, separately for 2014 and 2015. (More details on sensitivity tests and expansion-state status are in the Appendix.) ¹⁸

Eligibility Simulation

Eligibility for Medicaid was simulated by comparing family-level income in the American Community Survey interview year to the ACA and state-specific income eligibility thresholds applicable in 2013, ¹⁹ 2014, ⁴ and 2015. ²⁰ Children’s eligibility was calculated using age-specific MAGI-converted thresholds for Medicaid based on 2013 rules. To standardize eligibility determinations over the sample period, we used the 2013 state-reported MAGI-converted thresholds for both parents and children.

Model

We used triple-difference linear probability models to identify welcome-mat effects, observing public coverage among Medicaid-eligible children before and after the implementation of the ACA while controlling for parental Medicaid eligibility, expansion-state status, state fixed effects, and a wide range of individual- and family-level characteristics (for the child: age, sex, race, and Hispanic origin; for the parent: employment, education, and citizenship). Insurance was measured using mutually exclusive categories in the following hierarchical order: any public coverage, only private coverage, and uninsured. Results were derived using two separate models that both designated 2013 as the base year (pre-ACA) and then used 2014 and 2015, respectively, as the post-ACA year. Results were not sensitive to a specification that combined 2013–15 into a single model, as shown in the Appendix. ¹⁸

Limitations

Our study had the following limitations. First, eligibility for public programs is not available in survey data and thus was simulated. We were unable to simulate changes in Medicaid eligibility that would have resulted from the transition to MAGI income-counting rules under the ACA or to accurately identify parents who would have been eligible for Marketplace subsidies. More discussion of these issues is in the Appendix. ¹⁸ Second, as is common with difference models, we could not be certain that year effects did not pick up additional factors beyond the implementation of the ACA. Finally, prior research has raised concerns that some ACS respondents may misreport public coverage as private. ¹⁶ To address this point, we supplemented our analysis with results for the uninsured, thought to be a more reliable ACS measure, and we relied on consistent patterns in reporting of public/private coverage in the ACS during the sample period 2013–15.

Study Results

Already Eligible Children By Parental Eligibility For Medicaid

At the time of ACA implementation in 2014, 21.0 million nondisabled, citizen children living in families with incomes below 138 percent of poverty were already eligible for Medicaid and faced no changes in eligibility for public coverage between the pre- and post-ACA periods ( Exhibit 1 ). Already eligible children were split relatively evenly across expansion and nonexpansion states, but greater variation was observed across parental eligibility categories. Among already eligible children, 57.5 percent had parents who were also eligible for Medicaid both before and after ACA implementation. These children lived in both expansion and nonexpansion states but were far more prevalent in expansion states (8.6 million compared to 3.5 million).

The remaining 42.5 percent (8.9 million) of already eligible children below 138 percent of poverty had parents who were not eligible for Medicaid before the ACA. Of these children, 2.1 million lived in expansion states and had parents who became newly eligible through the ACA’s Medicaid expansion to adults. The other 6.8 million of these children lived in nonexpansion states and had parents who were never eligible for Medicaid.

Trends In Public Coverage Among Already Eligible Children

Public coverage among our sample of already eligible children increased significantly between 2013 and 2015 (76.5 percent to 79.8 percent; see Exhibit 2 , along with Appendix Exhibit A3 for standard errors). ¹⁸ Children living in expansion states experienced higher rates of public coverage (both pre and post ACA) and displayed a greater boost in coverage after the ACA (see Appendix Exhibits A2 and A3). ¹⁸ These national-level (and state/expansion-level) trends mask significant differences that occurred at the family level based on whether a child’s parents were newly, always, or never eligible for Medicaid. Although public coverage increased between 2013 and 2015 for all children, regardless of their parents’ eligibility, those with always eligible parents had significantly higher coverage rates both before and after ACA implementation. And in a comparison of differences over time, children with the largest changes in public coverage between 2013 and 2015 (parent newly eligible: 5.6 percentage points) experienced increases almost two times as large as those with the smallest changes (parent never eligible: 3.0 percentage points) ( Exhibit 2 , along with Appendix Exhibit A3 for standard errors). ¹⁸

SOURCE American Community Survey, 2013–15. NOTES Nondisabled, citizen children simulated to be eligible for Medicaid using 2013 Modified Adjusted Gross Income (MAGI)-converted thresholds. Public coverage measured at time of the interview. “Low-income” means income below 138 percent of the federal poverty level. State expansion status was defined as of 2014. Estimates using 2015 expansion status are in online Appendix Exhibit A3; see Note  18 in text. ^a For differences across time within each category of parental eligibility; significantly different from 2013 ( $p<0.01$ ). ^b For differences across time within each category of parental eligibility; significantly different from 2014 ( $p<0.01$ ).

Patterns of change also varied. Children with always eligible parents experienced a relatively constant rate of increase in public coverage between 2013 and 2015 ( Exhibit 2 ). Children with newly eligible parents had a greater boost in coverage in the first year of ACA implementation (2013–14), while children with never eligible parents exhibited lagged effects, with greater increases seen in the second year (2014–15).

Welcome-Mat Effects By Source

Exhibit 3 presents the results from our models of welcome-mat effects that simultaneously accounted for the implementation of the ACA, variation in parental eligibility for Medicaid, and state-level adult Medicaid expansion policy. Welcome-mat effects are presented as percentage-point changes in public coverage rates between the base year 2013 and end years 2014 and 2015. Our primary results are broken out by general ACA effects (with 2013 as the base year) and parental eligibility effects (with never eligible as the base category). In our secondary model, expansion policy effects can be observed by comparing children with always eligible parents living in expansion and nonexpansion states. (Full model results, including sample characteristics of control variables, are available in the Appendix.) ¹⁸

In the first year of ACA implementation (2014), welcome-mat effects were fully concentrated among children whose parents were also eligible for Medicaid. Children with newly eligible and already eligible parents were 3.3 and 1.4 percentage points, respectively, more likely to experience an increase in public coverage between 2013 and 2014 than children with parents who were never eligible for Medicaid.

By the second year of ACA implementation (2015), welcome-mat effects were more widespread. Public coverage rates were 2.7 percentage points higher among all children in 2015 than in 2013, regardless of their parents’ eligibility for Medicaid. Gains seen among children with newly eligible parents in the first year of implementation continued to outpace all others. These children experienced an additional boost in public coverage of 2.7 percentage points, for a total of 5.4 percentage points between 2013 and 2015, resulting in 120,000 additional children covered (see Appendix Exhibit A9). ¹⁸

By 2015 our primary model no longer exhibited additional coverage effects for children whose parents were already Medicaid eligible before the ACA. However, once we controlled for expansion state status, we found that children with already eligible parents living in expansion states experienced an additional boost of 1.2 percentage points, for a total of 3.9 percentage points compared to children with never eligible parents. This resulted in 350,000 additional children being covered between 2013 and 2015 in expansion states (see Appendix A9). ¹⁸ Their gains were also significantly larger than those for children with already eligible parents living in nonexpansion states (data not shown), which suggests the presence of expansion policy effects.

Using public coverage alone, we were unable to distinguish between the relative sizes of the welcome-mat effect for children whose parents were newly eligible versus those with always eligible parents living in expansion states. However, corresponding changes in the percentage of children uninsured and the percentage with only private coverage suggest that these two groups experienced different coverage patterns under the ACA (not shown, see Appendix Exhibit A8). ¹⁸ Children with newly eligible parents were 4.1 percentage points less likely to be uninsured than children with never eligible parents in 2015 than in 2013 but showed no significant difference in private coverage over the same time period. Alternatively, between 2013 and 2015, children with always eligible parents in expansion states were less likely than children with never eligible parents to be uninsured (−3.0 percentage points) and to have private coverage (−1.0 percentage points).

Impact Of Welcome-Mat Effects

Quantifying these effects ( Exhibit 4 ), we estimated that 710,000 low-income children who were already eligible for Medicaid before the ACA implementation gained public coverage between 2013 and 2015, despite having no changes in their own eligibility. If nonexpansion states had adopted the adult Medicaid expansion and experienced similar effects to those in the model, we predicted that an additional 200,000 low income children already eligible for Medicaid could have gained public coverage by 2015.

A gain of 710,000 translates to 4.6 percent nationwide growth in public coverage among children in our sample, but increases varied significantly when observed by parents’ eligibility for Medicaid. Gains for already eligible children with newly eligible parents (8.1 percent) were more than twice the size of those experienced by those with never eligible parents (3.6 percent) ( Exhibit 4 ). The growth in public coverage can also be broken down into the relative impacts from general ACA policies and parental eligibility effects. We estimated that by the second year of ACA implementation, 76 percent of the increase in public coverage among already eligible children was the result of general ACA policies, and 24 percent was attributed to parental eligibility (8 percent and 16 percent for newly eligible and always eligible parents in expansion states; see Appendix Exhibit A9 and discussion on pages 9–10 of the Appendix). ¹⁸

Sensitivity Testing

Our results were not sensitive to a wide variety of specification tests (available in the Appendix). ¹⁸ These included models that accounted for parents’ eligibility for Marketplace subsidies. Others tested the treatment of states that in the pre-ACA period had either already adopted generous (non-ACA) parental Medicaid eligibility policies or chosen to adopt the ACA adult Medicaid expansion early.

Discussion

Our results showed clear evidence of welcome-mat effects for children as the ACA was implemented, measured as increased enrollment in public coverage among children who were already eligible for Medicaid. We provide the first national-level, analytical estimates for welcome-mat effects among Medicaid-eligible children and are the first to reveal that the parental eligibility effects found in earlier Medicaid expansions to parents were also present in the ACA expansions. The largest effects in our sample were found among children whose parents gained eligibility for Medicaid through ACA-related Medicaid expansions to adults. These children had significant increases in both the first (2014) and second (2015) years of ACA implementation, outpacing coverage effects among children in families without newly eligible parents by more than double. Broader effects from general ACA policies exhibited more of a lagged pattern, with no significant effects found among the entire population of children until the second year of ACA implementation.

By 2015 the magnitude of ACA welcome-mat effects on public coverage ranged from 2.7 to 5.4 percentage points, depending on children’s exposure to expansion policies and parental eligibility. The 5.4-percentage-point increase among children with newly eligible parents aligned perfectly with welcome-mat effects found for children during the 1995–2002 Medicaid expansions to parents, ⁷ which serves as additional evidence of the strong link between children’s and parents’ eligibility and coverage. Our overall range was consistent with recent research that focused on California during early ACA expansions at the county level and that found public coverage among low-income children increased by approximately 3.2 percentage points. ¹⁷ Our estimates were also consistent with findings that show that Medicaid coverage increased by 3.8 percentage points under the ACA in a broader national-level population of nonelderly, previously eligible individuals. ¹⁶

We also found measurable effects of the ACA on overall insurance coverage. Insurance gains were widespread by 2015. All Medicaid-eligible children in our sample were less likely to be uninsured, with the impact growing to −3.0 percentage points by the second year of ACA implementation (see Appendix Exhibit A8). ¹⁸ Children with newly Medicaid-eligible parents had the greatest improvements: Their likelihood of being uninsured dropped by 4.1 percentage points between 2013 and 2015. Our primary results showed little evidence that private coverage was displaced by public coverage among our sample of Medicaid-eligible children (a phenomenon known as crowd-out). The one exception was found for children with always eligible parents in expansion states, whose 3.9-percentage-point increase in public coverage by 2015 coincided with a 1.0-percentage-point decrease in private coverage.

Our finding that enrollment effects were larger among children in families with joint parent/child Medicaid eligibility speaks to the importance of factoring in family-level decision making when crafting health policy. Consistent with the literature, we found that children were more likely to enroll in and retain public coverage when a parent was eligible for or enrolled in the same program. ^9,11 Extending coverage to families versus individuals simplifies coverage and access decisions for families and decreases hurdles associated with enrollment and renewal. Policy makers recognized the importance of family dynamics under the ACA, standardizing a minimum Medicaid threshold of 138 percent of poverty for all family members and transitioning separate state CHIP–eligible children below 138 percent of poverty into CHIP-funded Medicaid. Our results show that these efforts have made a difference. However, with the future of ACA-related programs currently under debate, it is important to recognize that prior research also predicts that the gains we found among Medicaid-eligible children could be lost if their parents lose Medicaid eligibility or insurance coverage (either public or private), even if there are no changes to eligibility for children themselves. ²¹

Identifying the presence of general ACA welcome-mat effects is also relevant for both current and future policy. It highlights the importance of accounting for the welcome-mat population in fiscal planning and in the structure of funding programs. For example, under the ACA, the relative cost to state and federal governments for new enrollees varied by pre-ACA status. States received a 100 percent federal match for newly eligible adults, while lower pre-ACA match rates applied to previously eligible adults and children in the welcome-mat population. And if insurance coverage is correlated with access to and use of services, state and local delivery systems should be prepared for changes in demand for services among both children and parents.

Future work on welcome-mat effects is warranted. We restricted our analysis to the 21.0 million citizen children below 138 percent of poverty who were already eligible for Medicaid, to best target the impact of joint parent/child Medicaid eligibility. The overall welcome mat effect of the ACA on already eligible children is likely to be much larger when noncitizens and publicly eligible children outside of our sample are considered. General ACA welcome-mat effects are likely to exist among noncitizen (1.0 million) and separate CHIP eligible (1.2 million) children below 138 percent of poverty as well as Medicaid- and CHIP-eligible children above that income level (14.4 million citizen children). At higher income levels, Marketplace policies and parental eligibility for subsidized coverage could play a significant role in welcome-mat effects, just as adult Medicaid expansions did in our low-income sample. Finally, given the increase in welcome-mat effects seen over time between the first and second years of ACA implementation, it will be important to continue to observe children’s coverage over time to see if additional gains were made after 2015.

Conclusion

Several policy decisions are on the horizon for children’s coverage, including the funding of CHIP (2017); expiration of maintenance-of-effort requirements for Medicaid and CHIP (2019); and legislative changes to Medicaid, CHIP, and components of the ACA. It is unclear whether enrollment gains seen among Medicaid-eligible children would reverse if ACA policies were rolled back—especially policies affecting parental eligibility and coverage. Nonetheless, our findings can inform policy makers as they craft future health care policy.

ACKNOWLEDGMENTS

An earlier version of this article was presented at the American Society of Health Economists, Sixth Biennial Conference 2016, June 15, 2016, in Philadelphia, PA, and at the Association for Public Policy Analysis and Management, 2016 Fall Research Conference, November 4, 2016, in Washington, DC. The authors appreciate the helpful comments of Joel Cohen, Sandra Decker, Genevieve Kenny, Victoria Lynch, Tom Selden, Benjamin Sommers, and the anonymous reviewers. The views expressed in this article are those of the authors alone, and no official endorsement by the Department of Health and Human Services or the Agency for Healthcare Research and Quality is intended or should be inferred.

NOTES

• 1 Cohen RA , Martinez ME , Zammitti EP . Health insurance coverage: early release of estimates from the National Health Interview Survey, 2015 [Internet]. Hyattsville (MD) : Centers for Disease Control and Prevention, National Center for Health Statistics ; 2016 May . Table IV, Percentages (and standard errors) of persons under age 65 who lacked health insurance coverage, had public health plan coverage, and had private health insurance coverage at the time of interview, by poverty status and year: United States, 2010–2015; and Table VI, Percentages (and standard errors) of children aged 0–17 years who lacked health insurance coverage, had public health plan coverage, and had private health insurance coverage at the time of interview, by poverty status and year: United States, 2010–2015; [cited 2017 Aug 19 ]. Available from: https://www.cdc.gov/nchs/data/nhis/earlyrelease/insur201605.pdf Google Scholar
• 2 Dubay L , Guyer J , Mann C , Odeh M . Medicaid at the ten-year anniversary of SCHIP: looking back and moving forward . Health Aff (Millwood) . 2007 ; 26 ( 2 ): 370 – 81 . Go to the article, Google Scholar
• 3 Hudson JL , Selden TS . Children’s eligibility and coverage: recent trends and a look ahead . Health Aff (Millwood) 2007 ; 26 ( 5 ): w618 – 29 . DOI: 10.1377/hlthaff.26.5.w618 . Go to the article, Google Scholar
• 4 Heberlein M , Brooks T , Artiga S , Stephens J. Getting into gear for 2014: shifting new Medicaid eligibility and enrollment policies into drive [Internet]. Washington (DC) : Kaiser Commission on Medicaid and the Uninsured ; 2013 Nov 21 . Appendix Table 2, Medicaid income eligibility limits for parents and other adults as a percent of the federal poverty level as of January 2013 and January 2014; [cited 2017 Aug 16 ]. Available from: https://kaiserfamilyfoundation.files.wordpress.com/2013/11/8516-getting-into-gear-for-2014-shifting-new-medicaid-eligibility-appendix-tables.pdf Google Scholar
• 5 Kenney G , Haley J , Pan C , Lynch V , Buettgens M ( Urban Institute, Washington, DC ). A look at early ACA implementation: state and national Medicaid patterns for adults in 2014 [Internet]. Princeton (NJ) : Robert Wood Johnson Foundation ; 2016 Sep [cited 2017 Aug 10 ]. Available from: http://www.rwjf.org/content/dam/farm/reports/issue_briefs/2016/rwjf431208 Google Scholar
• 6 Davidoff A , Kenney J , Dubay L , Yemane A . Patterns of child-parent insurance coverage: implications for coverage expansions . Washington (DC) : Urban Institute ; 2001 . (New Federalism Series B, No. B-39). Google Scholar
• 7 Aizer A , Grogger J . Parental Medicaid expansions and health insurance coverage . Cambridge (MA) : National Bureau of Economic Research ; 2003 Aug. (Working Paper No. 9907). Crossref, Google Scholar
• 8 Dubay L , Kenney J . Expanding public health insurance to parents: effects on children’s coverage under Medicaid . Health Serv Res . 2003 ; 38 ( 5 ): 1283 – 301 . Crossref, Medline, Google Scholar
• 9 Devoe JE , Marino M , Angier H , O’Malley J , Crawford C , Nelson C , et al. Effect of expanding Medicaid for parents on children’s health insurance coverage: lessons from the Oregon experiment . JAMA Pediatr . 2015 ; 169 ( 1 ): e143 – 5 . Crossref, Google Scholar
• 10 Devoe JE , Crawford C , Angier H , O’Malley J , Gallia C , Marino M , et al. The association between Medicaid coverage for children and parents persists: 2002–2010 . Maternal Child Health . 2015 ; 19 ( 8 ): 1766 – 74 . Crossref, Medline, Google Scholar
• 11 Sommers BD . Insuring children or families: do parental and sibling coverage lead to improved retention of children in Medicaid and CHIP? J Health Econ . 2006 ; 25 ( 6 ): 1154 – 69 . Crossref, Medline, Google Scholar
• 12 Alker J , Chester A . Children’s health coverage rate now at historic high level of 95 percent . Washington (DC) : Georgetown University Health Policy Institute, Center for Children and Families ; 2016 Oct . Google Scholar
• 13 Alker J , Chester A . Children’s health insurance rates in 2014; ACA results in significant improvements . Washington (DC) : Georgetown University Health Policy Institute, Center for Children and Families ; 2015 Oct . Google Scholar
• 14 Kenney GM , Haley JM , Pan CW , Lynch V , Buettgens M . Children’s coverage climb continues: uninsurance and Medicaid/CHIP eligibility and participation under the ACA . Washington (DC) : Urban Institute ; 2016 May . Google Scholar
• 15 Kenney GM , Haley JM , Pan CW , Lynch V , Buettgens M . Medicaid/CHIP participation rates rose among children and parents in 2015 . Washington (DC) : Urban Institute ; 2017 May . Google Scholar
• 16 Frean M , Gruber J , Sommers BD . Premium subsidies, the mandate, and Medicaid expansions: coverage effects of the Affordable Care Act . Cambridge (MA) : National Bureau of Economic Research ; 2016 . (Working Paper No. 22213). Crossref, Google Scholar
• 17 Sommers BD , Chua K , Kenney GM , Long SK , McMorrow S . California’s early coverage expansion under the Affordable Care Act: a county-level analysis . Health Serv Res . 2016 ; 51 ( 3 ): 825 – 45 . Crossref, Medline, Google Scholar
• 18 To access the Appendix, click on the Appendix link in the box to the right of the article online.
• 19 Medicaid.gov . MAGI conversion plans and results [Internet]. Baltimore (MD) : Centers for Medicare and Medicaid Services ; 2013 [cited 2017 Jun 26 ]. Available from: https://www.medicaid.gov/affordable-care-act/eligibility/magi-conversion-plan/index.html Google Scholar
• 20 Brooks T , Touschner J , Artiga S , Stephens J , Gates A . Modern era Medicaid: findings from a 50-state survey of eligibility, enrollment, renewal, and cost-sharing policies in Medicaid and CHIP as of January 2015 [Internet]. Washington (DC) : Kaiser Family Foundation ; 2015 . Table 1, Adult income eligibility limits as a percent of the federal poverty level, January 2015; [cited 2017 Aug 10 ]. Available from: http://www.kff.org/health-reform/report/modern-era-medicaid-findings-from-a-50-state-survey-of-eligibility-enrollment-renewal-and-cost-sharing-policies-in-medicaid-and-chip-as-of-january-2015/ Google Scholar
• 21 Devoe JE , Krois L , Edlund T , Smith J , Carlson NE . Uninsurance among children whose parents are losing Medicaid coverage: results from a statewide survey of Oregon families . Health Serv Res . 2008 ; 43 ( 1 ): 401 – 18 . Crossref, Medline, Google Scholar