Taking Stock at Mid-Year: Health Insurance Coverage under the ACA as of June 2014
Sharon K. Long, Genevieve M. Kenney, Stephen Zuckerman, Douglas Wissoker, Adele Shartzer, Michael Karpman, Nathaniel Anderson, and Katherine Hempstead
July 29, 2014
The Affordable Care Act (ACA) has brought major changes to the US health insurance system: In January 2014, Medicaid was expanded to nearly all adults with family incomes at or below 138 percent of the federal poverty level (FPL) in 24 states and the District of Columbia,1 and enrollment under the new health insurance Marketplaces created by the ACA officially began in all states and the District of Columbia. By April 19th, enrollment in Marketplace plans was reported at over 8 million (Assistant Secretary for Planning and Evaluation, 2014). Also, reports from the Centers for Medicare and Medicaid Services (CMS) suggest that Medicaid enrollment increased between the beginning of October 2013 and the end of April 2014 (Centers for Medicare and Medicaid Services 2014). However, neither the Marketplace enrollment figures nor the CMS Medicaid report provide an accurate picture of how many uninsured people have gained coverage since open enrollment began, because both sets of enrollment figures may include newly insured people as well as those who had other sources of coverage before 2014.2
We use the June 2014 Health Reform Monitoring Survey (HRMS) to examine changes in health insurance coverage since the beginning of the previous year for nonelderly adults. The HRMS was designed to provide early feedback on ACA implementation to complement the more robust assessments that will be possible when the federal surveys release their estimates of changes in health insurance coverage later in 2014 and in 2015 (Long, Kenney, Zuckerman, Goin, et al. 2014). This brief updates estimates from early March 2014, which showed an estimated gain in coverage of 5.4 million (Long, Kenney, Zuckerman, Wissoker et al. 2014).
What We Did
Our analysis compares the uninsurance rate for nonelderly adults (age 18 to 64) through quarter 2 of 2014 (survey fielded in June 2014) to rates estimated from data collected during 2013. We focus on estimated changes in the uninsurance rate, because estimates of the level of uninsurance often vary across surveys due to differences in the surveys that would be unrelated to the ACA (State Health Access Data Assistance Center 2013). Though we include information for all four quarters of 2013, we focus on changes between quarter 3 of the 2013 HRMS, fielded in September 2013 just before the first Marketplace open enrollment period was initiated on October 1, 2013, and quarter 2 of the 2014 HRMS, fielded in June 2014.3
Though each round of the HRMS is weighted to be nationally representative, it is important in examining changes over time that we base our estimates on comparable samples over time. For example, if the share of those with insurance grows simply because more respondents were older or from higher income groups in one round of the survey, it would be incorrect to associate such a change with the ACA Marketplaces and Medicaid expansions. This is a particular challenge in comparing estimates from survey samples over time because the composition of the sample that is surveyed can change from round to round in ways that are not fully captured in the weights and that may distort the estimates of change.
To control for the potential influence of changes in the characteristics of the HRMS sample, we estimate weighted regression models that control for demographic and socioeconomic characteristics, internet access, and geography.4 We consider changes in insurance coverage for (1) all nonelderly adults,5 (2) adults targeted by the Medicaid expansion and the Marketplaces, (3) adults in states that have and have not adopted the ACA’s optional Medicaid expansion by June 1, 2014, and (4) adults in important demographic and socioeconomic subgroups, including age, gender, race and ethnicity, and family income. Controlling for differences in the characteristics of the respondents through time allows us to remove any variation in insurance coverage caused by changes in the types of people responding to the survey, rather than by changes in the health insurance landscape. In presenting the regression-adjusted estimates, we use the predicted rate of uninsurance in each quarter for the same nationally representative population. For this analysis, we base the nationally representative sample on survey respondents from the most recent 12-month period from the HRMS (i.e., quarters 3 and 4 of 2013 and quarters 1 and 2 of 2014).
In presenting our findings, we focus on statistically significant changes in insurance coverage over time (defined as differences that are significantly different from zero at the 5 percent level or lower) and highlight changes relative to September 2013, just before the open enrollment period for the Marketplaces began. We provide a 95 percent confidence interval (CI) for key estimates. The basic patterns shown for the regression-adjusted measures are similar to those based solely on simple weighted (unadjusted) estimates. As noted above, we report on the regression-adjusted estimates that compare September 2013 (just before the start of the open enrollment period) to June 2014. In extrapolating from our estimates of changes in uninsurance rates to the number of adults who have gained coverage over the same period, we use projections for the size of the 2014 population from the US Census Bureau.6
What We Found
The number of uninsured nonelderly adults fell by an estimated 8.0 million between September 2013 and June 2014, representing a drop of 22.3 percent in the uninsurance rate. In June 2014, the uninsurance rate for nonelderly adults was estimated to be 13.9 percent (95% CI [12.3, 15.4]) for the nation, a drop of 4.0 percentage points (95% CI [2.6, 5.5]) since September 2013 (figure 1).7 Applying the estimated 4.0 percentage point decrease in the uninsured rate to the estimated number of nonelderly adults in the nation, these results indicate that the number of uninsured adults declined by 8.0 million between September and June (95% CI [5.1 million, 10.8 million]). This is well above the 5.4 million (95% CI [3.2 million, 7.6 million]) drop in uninsurance estimated for early March 2014 (Long et al. 2014b).
States that implemented the ACA's Medicaid expansion saw large and continued declines in uninsurance through June 2014. The uninsurance rate for adults in the states that adopted the ACA's Medicaid expansion dropped 6.1 percentage points (95% CI [4.9, 7.2]) since September 2013, compared with a drop of 1.7 percentage points (95% CI [0.3, 3.0]) for the nonexpansion states. This represents a decline in the uninsurance rate of 37.7 percent in the expansion states, as compared to only 9 percent in the nonexpansion states. The uninsurance rate for adults in the 25 nonexpansion states was 18.3 percent (95% CI [17.0,19.6]) in June 2014, well above the 10.1 percent average in the expansion states (95% CI [9.1, 11.2]) (figure 1). The gap in the uninsurance rate between expansion and nonexpansion states, which had widened between September 2013 and March 2014, grew further between March and June 2014. In March 2014, the uninsurance rate for nonelderly adults in the expansion states was 6.5 percentage points lower than the rate in the nonexpansion states; by June that difference had grown to 8.2 percentage points.8 Consequently, the remaining uninsured adults are becoming increasingly concentrated in the states that have opted to not expand Medicaid, with 60 percent of the uninsured in those states in June 2014 as compared with 50 percent in September 2013 (Kenney et al. 2014).
Large gains in insurance coverage were reported by the low- and middle-income adults targeted by the ACA's key coverage provisions. Between September 2013 and June 2014, insurance coverage increased by 7.3 percentage points (95% CI [3.6, 11.0]) for the adults with family income at or below 138 percent of FPL who were targeted by the ACA's Medicaid expansion and by 5.3 percentage points (95% CI [3.3, 7.3]) for the middle-income adults (139 to 399 percent of FPL) who were targeted by the new subsidies available for health insurance coverage through the Marketplaces (figure 2).
The gains in coverage benefited adults across all age, sex, and race and ethnicity groups, with stronger gains among groups that historically have higher uninsurance rates. As shown in figure 2, there were strong gains in coverage for young adults (age 18 to 30) (up 4.3 percentage points; 95% CI [0.7, 8.0]) and nonwhite, non-Hispanic adults (up 6.9 percentage points; 95% CI [4.3, 9.4]) and Hispanic adults (up 5.3 percentage points; 95% CI [2.3, 8.4]), groups that have historically had higher than average uninsurance rates. Coverage rates increased for both men and women (4.3 percentage points; 95% CI [2.3, 6.2], and 3.5 percentage points; 95% CI [1.8, 5.3], respectively). Historically, men have had a higher rate of uninsurance than women.
Not surprisingly, low-income adults targeted by the Medicaid expansion had particularly large gains in insurance coverage in the states that expanded Medicaid under the ACA between September 2013 and June 2014 (figure 3). Insurance coverage increased by 13.7 percentage points (95% CI [10.3, 17.1]) for adults with incomes at or below 138 percent of FPL in the expansion states compared with a change of only 1.2 percentage points (95% CI [-2.3, 4.6]) in the nonexpansion states. In contrast, adults in the middle-income group (i.e., with incomes between 139 and 399 percent of FPL) who could potentially qualify for Marketplace subsidies experienced similar gains in coverage in the expansion and nonexpansion states over the period: 6.3 percentage points (95% CI [3.4, 9.1]) versus 4.2 percentage points (95% CI [1.6, 6.7]), respectively.
Gains in coverage were reported for all of the population subgroups examined in the expansion states except for higher-income adults (i.e., adults with family incomes at or above 400 percent of FPL). Particularly strong gains in insurance coverage were reported for young adults, men, and minority adults. In contrast, the only other population subgroup to report significant gains in coverage in the nonexpansion states was older adults (50–64), where coverage increased 3.4 percentage points, as compared with 4.5 percentage points in the expansion states.
What It Means
Significant gains in coverage have occurred since the beginning of the first Marketplace open enrollment period in October 2013, especially in the states that have expanded Medicaid eligibility under the ACA and among the populations targeted by the Medicaid expansion and the new Marketplace subsidies. We estimate that the number of uninsured adults fell by 8.0 million between September 2013 and June 2014, with a 95 percent confidence interval of 5.1 to 10.8 million. This represents a net gain in coverage of roughly 2.6 million since early March, reflecting the surge in enrollment in the Marketplaces at the end of March and ongoing gains in Medicaid enrollment. Altogether, the uninsurance rate declined by 22.3 percent between September 2013 and June 2014.
These estimates of changes between September 2013 and June 2014 do not reflect the effects of some important ACA provisions (such as the ability to keep dependents on a parent’s health plan until age 26 and early state Medicaid expansions) that were implemented before 2013. Therefore, these early estimates will understate coverage changes attributable to the ACA as a whole. Moreover, deriving definitive estimates of the effects of the ACA on health insurance coverage will require accounting for any changes in health insurance coverage that would have occurred independent of the ACA.
The HRMS, like other recent surveys, finds an ongoing decline in the uninsurance rate among nonelderly adults since the ACA rollout began at the end of 2013 (Collins et al. 2014).9 Moreover, information released by CMS indicates larger Medicaid enrollment increases in expansion states than in nonexpansion states, which is consistent with our finding of greater declines in the uninsured in those states (Centers for Medicare and Medicaid Services 2014). Though monitoring surveys like the HRMS and Gallup provide estimates of the changes under the ACA, the estimated magnitude of the change differs given differences in survey design, survey samples, time periods, and other factors. A definitive assessment of the magnitude of the changes in coverage occurring nationally and by state and population subgroups will have to wait for data from the federal surveys, with their stronger designs, larger sample sizes, much higher response rates, and longer time-trends.
Though the gains in coverage so far in 2014 are noteworthy (particularly in the states that expanded Medicaid, where uninsurance dropped 37.7 percent), the uninsurance rate for low-income adults across the nation remains quite high (31.7 percent). In the expansion states, where the uninsurance rate for low-income adults was 23.1 percent at the time of our most recent survey, the opportunity remains for substantial coverage gains under the Medicaid program. In the nonexpansion states, where the uninsurance rate for low-income adults was at 40.0 percent, low-income adults have few paths to coverage in the absence of Medicaid.10 Thus, there is an increasing concentration of the uninsured in those states that have not opted to expand Medicaid (Kenney et al. 2014).
While a full accounting of the effects of the ACA on changes in insurance coverage is not possible yet, these findings provide evidence that the Medicaid expansion and the provision of subsidies in the new health insurance Marketplaces are likely contributing to significant increases in health insurance coverage among nonelderly adults. We find larger coverage gains in the states that expanded Medicaid for adults, overall and specifically among adults with incomes at or below 138 percent of FPL, but cannot definitively attribute those differences to the Medicaid expansion because expansion states differed from nonexpansion states in other ways that could affect changes in coverage over this period. In particular, states that expanded Medicaid were also more likely to implement a state-based Marketplace as opposed to relying on the federally facilitated Marketplaces, and the state-based Marketplaces had access to substantially greater resources for outreach and enrollment activities (Blumberg et al. 2014; Hill et al. 2014). Disentangling the effects of the range of state policy decisions being made under the ACA will require waiting for data from the federal surveys.
Assistant Secretary for Planning and Evaluation. 2014. Health Insurance Marketplace: Summary Enrollment Report for the Initial Annual Open Enrollment Period. Washington, DC: United States Department of Health and Human Services.
Blumberg, Linda J., John Holahan, Genevieve M. Kenney, Matthew Buettgens, Nathanial Anderson, Hannah Recht, and Stephen Zuckerman. 2014. Measuring Marketplace Enrollment Relative to Projections. Washington, DC: Urban Institute.
Blumenthal, David, and Sara R. Collins. 2014. “Health Care Coverage under the Affordable Care Act—A Progress Report.” New England Journal of Medicine, 371 (3): 275–81..
Carman, Katherine Grace, and Christine Eibner. 2014. Changes in Health Insurance Enrollment since 2013. Santa Monica, CA: RAND Corporation.
Centers for Medicare and Medicaid Services. 2014. Medicaid and CHIP: April 2014 Monthly Applications, Eligibility Determinations, and Enrollment Report. Baltimore: US Department of Health and Human Services.
Collins, Sara R., Petra W. Rasmussen, and Michelle M. Doty. 2014. Gaining Ground: Americans' Health Insurance Coverage and Access to Care After the Affordable Care Act's First Open Enrollment Period. New York: The Commonwealth Fund.
Decker, Sandra L., Genevieve M. Kenney, and Sharon K. Long. 2014. “Characteristics of Low-Income Adults in States Expanding vs. Not Expanding Medicaid.” JAMA Internal Medicine.
Hill, Ian, Margaret Wilkinson, and Brigette Courtot. 2014. The Launch of the Affordable Care Act in Selected States: Outreach, Education, and Enrollment Assistance. Washington, DC: Urban Institute.
Kenney, Genevieve M., Sharon K. Long, Adele Shartzer, Stephen Zuckerman, Douglas Wissoker, Michael Karpman, and Nathaniel Anderson. 2014. “QuickTake: States Not Expanding Medicaid under the ACA Account for 60 percent of the Nation’s Uninsured in June 2014; Up from 50 Percent in September 2013.” Washington, DC: Urban Institute.
Long, Sharon K., Genevieve M. Kenney, Stephen Zuckerman, Dana E. Goin, Douglas Wissoker, Frederic Blavin, Linda J. Blumberg, Lisa Clemans Cope, John Holahan, and Katherine Hempstead. 2014. TheHealth Reform Monitoring Survey: Addressing Data Gaps to Provide Timely Insights into the Affordable Care Act. Health Affairs, 33 (1): 161–67.
Long, Sharon K., Genevieve M. Kenney, Stephen Zuckerman, Douglas Wissoker, Dana E. Goin, Michael Karpman, and Nathaniel Anderson. 2014. “QuickTake: Number of Uninsured Adults Falls by 5.4 Million since 2013.” Washington, DC: Urban Institute.
Sommers, Benjamin D., Genevieve M. Kenney, and Arnold M. Epstein. 2014. “New Evidence on the Affordable Care Act: Coverage Impacts of Early Medicaid Expansions.” Health Affairs, 33 (1): 78–87.
About the Series
This brief is part of a series drawing on the Health Reform Monitoring Survey (HRMS), a quarterly survey of the nonelderly population that is exploring the value of cutting-edge Internet-based survey methods to monitor the Affordable Care Act (ACA) before data from federal government surveys are available. The briefs provide information on health insurance coverage, access to and use of health care, health care affordability, and self-reported health status, as well as timely data on important implementation issues under the ACA. Funding for the core HRMS is provided by the Robert Wood Johnson Foundation, the Ford Foundation, and the Urban Institute.
For more information on the HRMS and for other briefs in this series, visit www.urban.org/hrms.
About the Authors
Sharon K. Long is a senior fellow, Genevieve M. Kenney and Stephen Zuckerman are senior fellows and co-directors, Adele Shartzer and Michael Karpman are research associates, and Nathaniel Anderson is a research assistant in the Urban Institute’s Health Policy Center. Douglas Wissoker is a senior fellow at the Urban Institute. Katherine Hempstead is a senior program officer at the Robert Wood Johnson Foundation.
The authors gratefully acknowledge the suggestions and assistance of Linda J. Blumberg, John Holahan, Frederic Blavin, Lisa Clemans-Cope, and Bowen Garrett.
1The list of states that have expanded Medicaid is increasing over time as more states decide to implement the ACA expansion. States that expanded Medicaid by June 1, 2014, are AZ, AR, CA, CO, CT, DE, DC, HI, IL, IA, KY, MD, MA, MI (as of April 1), MN, NV, NJ, NM, NY, ND, OH, OR, RI, VT, WA, and WV. Several of those states, including CA, CT, DC, and MN expanded Medicaid to adults under the ACA before 2013. New Hampshire is planning to begin covering adults under the Medicaid expansion in August 2014.
2Caution should be taken in comparing estimates of coverage changes across studies, however. For example, the estimate from a recent analysis by Blumenthal and Collins (2014) is not comparable to the findings presented here, because that estimate counts the number of changes between different types of coverage (e.g., a change from one nongroup insurance plan to another is counted the same as a change from being uninsured to having Medicaid) whereas the estimates here compare the number of uninsured in the September 2013 to the number of uninsured in June 2014, thus estimating the net change in uninsurance over the period.
3Although Marketplace coverage for people enrolling between October and December 2013 did not start until January 2014, some who signed up in the fall may have reported having coverage during the December 2013 HRMS survey. Further, some of those seeking coverage through the Marketplace between October and December 2013 were enrolled in Medicaid.
4Specifically, we control for the variables used in the poststratification weighting of the KnowledgePanel (the internet-based survey panel that underlies the HRMS) and the poststratification weighting of the HRMS, which include sex, age, race and ethnicity, language, education, marital status, whether any children are present in the household, household income, family income as a percent of the federal poverty level, homeownership status, internet access, urban or rural status, and census region.
5In this brief, we are not looking at the effects of the ACA on coverage for children, but recognize that their coverage and well-being may also be affected as their parents enroll in coverage for themselves or by other provisions of the ACA.
6We used 2014 national population predictions available from the US Census Bureau. These files give population projections by race, ethnicity, and sex of all ages from 2012 to 2060 based on estimated birth rates, death rates, and net migration rates over the time period. Using the “Table 1, Middle Series” file (which has a 2014 projected population of 318,892,103), we summed the 2014 population projections for all 18–64 year-olds to arrive at 198,461,688 nonelderly adults in 2014. See US Census Bureau, “2012 National Population Projections: Downloadable Files,” US Department of Commerce, last revised May 15, 2013.
7The estimates of uninsurance reported here differ from some early estimates that have been reported elsewhere. This reflects two factors: (1) we revised the editing process for insurance coverage in quarter 3 2013 to make better use of information from an open-ended follow-up question that was added in quarter 2 2013 to capture type of insurance coverage for those who said they were covered but did not pick a type of coverage from the list that was provided, and (2) the regression-adjusted estimates are always based on the most recent four quarters of data (e.g., for this brief—quarters 3 and 4 2013 and quarters 1 and 2 2014).
8Work with the National Health Interview Survey has shown that uninsurance rates were higher for low-income adults in the nonexpansion states in 2010–2012. The changes in insurance coverage found here have likely exacerbated the coverage gap for low-income adults between expansion and nonexpansion states (Decker et al. 2014).
10In states that did not choose to expand Medicaid, adults with incomes between 100 and 138 percent of FPL who do not have access to an affordable offer of employer coverage are eligible for Marketplace-based subsidies.