OCTOBER 2015 Insight on the Issues Monitoring the Impact of Health Reform on Americans Ages 50–64: Uninsured Rate Dropped by Nearly Half between December 2013 and March 2015 Laura Skopec Jane Sung Timothy A. Waidmann Olivia Dean Urban Institute AARP Public Policy Institute This Insight on the Issues is the latest in a series that looks at the experience of 50- to 64-year- olds during implementation of the Affordable Care Act. New data from the Urban Institute and the AARP ACA ADDRESSED MAJOR COVERAGE HURDLES Public Policy Institute show that the number of 50- FOR 50- TO 64-YEAR-OLDS to 64-year-olds without health insurance coverage Americans ages 50 to 64 often have a high need dropped by nearly half over the first two open for medical care, but those without employer- enrollment periods of the Affordable Care Act (ACA).1 sponsored insurance (ESI) historically have had This analysis compares results from December 2013, few options for obtaining affordable coverage. during the first ACA open enrollment period,2 with Before implementation of the ACA, many 50- to results from March 2015, after conclusion of the 64‑year‑olds who were not offered an employer- second ACA open enrollment period. sponsored health insurance plan remained uninsured. Many remained uninsured due to In March 2015, the uninsured rate was lower in difficulty gaining affordable coverage through a states that chose to expand their Medicaid programs public program or the nongroup market.3 under the ACA than in states that did not. Overall, gains in Medicaid and nongroup (individual) Several provisions of the ACA make health coverage appear to be responsible for the greatest insurance more accessible and affordable for people drops in the uninsured rate for 50- to 64-year-olds ages 50 to 64 by: between December 2013 and March 2015. The •• Expanding Medicaid to cover childless adults uninsured rate fell among all demographic groups with incomes up to 138 percent of the federal studied, although some subgroups continue to be poverty level (FPL) uninsured at higher rates. •• Ensuring older adults with preexisting health conditions are able to purchase affordable health OCTOBER 2015 insurance coverage in the private nongroup and RESULTS small-group health insurance markets Uninsured Rate Dropped by Nearly Half since •• Prohibiting insurers from varying health December 2013 insurance premiums by health status Between December 2013 and March 2015, the uninsured rate for people ages 50 to 64 fell by •• Limiting cost sharing for coverage of essential 47.4 percent, from 11.6 percent to 6.1 percent health benefits (figure 1). This is consistent with the drop in the •• Limiting the extent to which insurers can charge uninsured rate over the same period among adults older persons higher premiums than younger ages 19 to 64 (from 17.4 percent to 10.1 percent), as persons documented in a different study (Long et al. 2015). •• Making subsidies available to people with The drop in the uninsured rate was accompanied by incomes below 400 percent of FPL4 so they can gains in Medicaid and private nongroup coverage afford to purchase private nongroup insurance among adults ages 50 to 64 (figure 1) between through the health insurance Marketplace December 2013 and March 2015. •• Medicaid: The share of 50- to 64-year-olds with Medicaid coverage increased from 6.1 percent to 9.4 percent. FIGURE 1 Changes in Health Insurance Coverage among 50- to 64-Year-Olds between December 2013 and March 2015 December 2013 March 2015 Drop in Uninsured Increased Health Coverage 11.6% 9.0%** 9.4%** 7.2% 6.1%** 6.1% Uninsured Private nongroup Medicaid Uninsured Private nongroup Medicaid Source: HRMS-AARP Survey, December 2013 and March 2015. Note: ESI, Medicare, and nonspecified coverage not shown. Data are not adjusted for changes in the characteristics of the sample population over time. ** March 2015 estimate is significantly different from December 2013 estimate at the 0.01 level using two-tailed tests. 2 OCTOBER 2015 •• Private nongroup: The share of 50- to Coverage Gains Seen across Demographic Groups 64‑year‑olds with private nongroup coverage Between December 2013 and March 2015, the increased from 7.2 percent to 9.0 percent. uninsured rate among 50- to 64-year-olds declined for all demographic groups included in this analysis ESI coverage among this age group remained stable (figure 3). The largest percentage point declines at about 66 percent (data not shown). However, we in the uninsured rate in this age group occurred note that estimating changes in enrollment for among the following groups: specific coverage types from survey data, including coverage increases, is challenging and subject to •• Women: The uninsured rate for 50- to 64-year-old error.5 women dropped from 12.1 percent to 5.5 percent. Uninsured Rate Lowest in Medicaid Expansion •• Individuals with family incomes below States 138 percent of the FPL: The uninsured rate In March 2015, the uninsured rate for 50- to for 50- to 64-year-olds with the lowest incomes 64‑year‑olds was lower in states that chose to dropped from 32.7 percent to 17.8 percent. expand eligibility for their Medicaid programs •• Hispanics: The uninsured rate for Hispanics under the ACA than in states that did not (figure 2). in this age group dropped from 24.9 percent to The uninsured rate was 4.6 percent in Medicaid 11.2 percent. expansion states, compared with 8.7 percent in •• Fair or poor health: The uninsured rate for 50- nonexpansion states. to 64-year-olds with fair or poor health dropped from 18.6 percent to 8.2 percent. FIGURE 2 Uninsured Rate among 50- to 64-Year-Olds Despite Overall Gains in Coverage, Some Gaps in March 2015, by State Medicaid Expansion Remain Status Despite overall declines in the uninsured rate for 50- to 64-year-olds, several subgroups continue 8.7%** to remain uninsured at much higher rates than others (figure 4 and appendix table 1). These same subgroups have also seen the largest coverage gains since December 2013. 4.6% •• Individuals with family incomes at or below 138 percent of the FPL: In March 2015, individuals at or below 138 percent of the FPL were nearly 14 times as likely as those at or above 400 percent of the FPL to be uninsured (17.8 percent and 1.3 percent, respectively). States expanding States not expanding •• Hispanics: In March 2015, Hispanics remained Medicaid Medicaid more than twice as likely to be uninsured as Source: HRMS-AARP Survey, March 2015. non-Hispanic whites (11.2 percent and 5.3 percent, respectively). Note: State Medicaid expansion status as of March 31, 2015. Data are not adjusted for differences in demographic, •• Fair or poor health: In March 2015, 50- to socioeconomic, or health status between expansion and 64-year-olds in fair or poor health were nearly nonexpansion states. twice as likely to be uninsured as those in ** Estimate is significantly different from estimate for excellent or very good health (8.2 percent and states expanding Medicaid at the 0.01 level using two- 4.3 percent, respectively). Fair or poor health was tailed tests. also associated with lower income.6 3 OCTOBER 2015 FIGURE 3 Changes in Uninsured Rate for 50- to 64-Year-Olds by Demographic Group and Health Status between December 2013 and March 2015 December 2013 March 2015 11.1% Sex Male 6.6%** Female 12.1% 5.5%** At or below 138% FPL 32.7% 17.8%** Income 138–399% FPL 13.6% 6.0%** At or above 400% FPL 2.2% 1.3%* 50–54 13.4% 7.2%** Age 55–59 10.8% 5.9%** 60–64 10.3% 5.0%** White, non-Hispanic 9.5% 5.3%** Black, non-Hispanic 11.6% Race/Ethnicity 6.3%** Other, non-Hispanic 13.9% 6.0%** Hispanic 24.9% 11.2%** Excellent or very good 7.7% 4.3%** Health Status Good 13.4% 7.2%** Fair or poor 18.6% 8.2%** Source: HRMS-AARP Survey, December 2013 and March 2015. Note: Data are not adjusted for changes in the characteristics of the sample population over time. */** March 2015 estimate is significantly different from December 2013 estimate at the 0.05/0.01 level using two-tailed tests. These results are consistent with comparisons of state policy makers should consider the following uninsured rates for adults ages 19 to 64 (Shartzer et strategies: al. 2015). •• Encourage non-Medicaid expansion states to CONCLUSIONS AND POLICY RECOMMENDATIONS expand their programs to cover adults under 65 Health insurance coverage for 50- to 64-year-olds with income under 138 percent of the FPL.7 improved significantly between December 2013 and •• Identify and effectively address obstacles to March 2015. However, coverage gaps for vulnerable enrollment in insurance coverage among groups remain. In order to improve coverage and Hispanics.8 reduce disparities within this age group, federal and 4 OCTOBER 2015 FIGURE 4 Comparisons of Uninsured Rates for 50- to 64-Year-Old Groups in March 2015 Show Gaps in Coverage 17.8% 11.2% 8.2% 7.2% 6.0% 6.3% 6.0% 5.3% 4.3% 1.3% ≤138% FPL 138%–399% ≥400% FPL Hispanic Black, non- White, non- Other, non- Fair or poor Good Excellent or FPL Hispanic Hispanic Hispanic very good Family Income Race/Ethnicity Health Status Source: HRMS-AARP Survey, March 2015. Note: Data are not adjusted for differences in health status, income, or other characteristics within groups. See appendix table 1 for statistical significance. •• Target outreach to low-income groups to Reform on Americans Ages 50–64” series are from encourage those who may qualify for subsidies to oversamples of 50- to 64-year-old adults (HRMS- explore their options on the Marketplace.9 AARP). The Urban Institute and GfK Custom •• Target outreach to those with educational Research conducted the survey, and AARP provided or language barriers and provide them with funding to increase the sample size for this age information and assistance that meets their group. GfK Custom Research fielded the HRMS- needs. AARP oversample survey in December 2013, March 2014, December 2014, and March 2015 and included The AARP Public Policy Institute will continue to approximately 8,000 adults ages 50 to 64 for each monitor trends in coverage for 50- to 64-year-olds as survey period. federal survey data become available. The HRMS-AARP is weighted to be nationally DATA AND METHODS representative. Results presented here were This analysis uses data collected by the Urban not adjusted for changes in the demographic Institute’s Health Reform Monitoring Survey characteristics of the HRMS-AARP sample (HRMS), a quarterly Internet-based survey of between December 2013 and March 2015. adults under the age of 65 designed to provide Comparisons within subgroups were not adjusted rapid feedback on implementation of the ACA for socioeconomic, geographic, or health status before data from federal surveys are available. The differences across racial and ethnic groups. More survey data used for this paper and other analyses information on the HRMS is available at in AARP’s “Monitoring the Impact of Health http://hrms.urban.org/. 5 OCTOBER 2015 REFERENCES Washington, DC, April 16, 2015. http://hrms.urban.org/ Alonso-Zaldivar, Ricardo. “Healthcare.gov’s EZ briefs/Gains-in-Health-Insurance-Coverage-under-the- Form Not an Easy Route for Legal Immigrants.” The ACA-as-of-March-2015.html. Rundown (blog), October 23, 2014. Pascale, Joanne. “Measurement Error in Health http://www.pbs.org/newshour/rundown/healthcare- Insurance Reporting.” Inquiry 45, no. 4 (2009): 422–37. govs-ez-form-legal-immigrants/. Pascale, Joanne, Rodean, Jonathan, Leeman, Blavin, Fredric, Zuckerman, Stephen, Karpman, Jennifer, Cosenza, Carol, and Schoua-Glusberg, Michael, and Clemans-Cope, Lisa. “Why Are Hispanics Alisu. “Preparing to Measure Health Coverage Slow to Enroll in ACA Coverage? Insights from the in Federal Surveys Post-Reform: Lessons from Health Reform Monitoring Survey.” Health Affairs Massachusetts.” Inquiry 50, no. 2 (2013): 106–23. (blog), March 18, 2014. http://healthaffairs.org/ blog/2014/03/18/why-are-hispanics-slow-to-enroll- Shartzer, Adele, Long, Sharon K., Karpman, in-aca-coverage-insights-from-the-health-reform- Michael, Kenney, Genevieve M., and Zuckerman, monitoring-survey/. Stephen. “QuickTake: Insurance Coverage Gains Cross Economic, Social, and Geographic Call, Kathleen T., Davern, Michael E., Klerman, Boundaries.” Urban Institute QuickTake, Urban Jacob A., and Lynch, Victoria. “Comparing Errors in Institute, Washington, DC, April 27, 2015. Medicaid Reporting across Surveys: Evidence to Date.” http://hrms.urban.org/quicktakes/Insurance- Health Services Research 48, no. 2, pt. 1 (2013): 652–64. Coverage-Gains-Cross-Economic-Social-and- Dorn, Stan, Buettgens, Matthew, and Dev, Jay. “In Geographic-Boundaries.html. States That Don’t Expand Medicaid, Who Gets Weiss, Adam L., Waidmann, Timothy A., and New Coverage Assistance under the ACA and Caswell, Kyle J. “Monitoring the Impact of Health Who Doesn’t?” Urban Institute Timely Analysis of Reform on Americans 50–64: Medicaid Expansion Immediate Health Policy Issues, Urban Institute, and Marketplace Implementation Increased Health Washington, DC, October 2014. Coverage.” AARP Public Policy Institute Insight on http://research.urban.org/uploadedpdf/413248-Who- the Issues, AARP, Washington, DC, December 2015. Gets-New-Coverage-Assistance-Under-the-ACA-and- http://www.aarp.org/content/dam/aarp/ppi/2014-12/ Who-Doesnt.pdf. medicaid-expansion-and-marketplace-insight95- Long, Sharon K., Karpman, Michael, Kenney, AARP-ppi-health.pdf. Genevieve M., Zuckerman, Stephen, Wissoker, Zuckerman, Stephen, Waidmann, Timothy A., and Douglas, Shartzer, Adele, Anderson, Nathaniel, and Lawton, Emily. “Undocumented Immigrants, Left Hempstead, Katherine. “Taking Stock: Gains in Health Out of Health Reform, Likely to Continue to Grow Insurance Coverage under the ACA as of March as Share of the Uninsured.” Health Affairs 40, no. 10 2015.” Urban Institute Policy Briefs, Urban Institute, (2011): 1,997–2,004. 6 APPENDIX whether they are enrolled in employer-sponsored coverage, Medicaid, or private nongroup coverage (Pascale 2009; APPENDIX TABLE 1 Pascale et al. 2013). Survey-based estimates of enrollment Uninsured Rate in March 2015 for 50- to in public programs such as Medicaid may also be lower than 64-Year-Olds, by Demographic Characteristics enrollment counts in administrative data (Call et al. 2013). and Self-Reported Health Status 6 In March 2015, 42.6 percent of 50- to 64-year-olds in fair or poor health had a family income at or below 138 percent of the FPL, compared with 9.1 percent of those in excellent or Uninsured Rate, very good health (not shown). March 2015 Male^ 6.6% 7 In October 2014, approximately 1.8 million uninsured 45- to Gender 5.5% 64-year-olds were estimated to be ineligible for coverage Female due to lack of Medicaid expansion (Dorn, Buettgens, and Dev At or below 138% FPL^ 17.8% 2014). In nonexpansion states, individuals with incomes below Family Income 138 to 399% FPL 6.0%** 100 percent of the FPL do not qualify for financial assistance At or above 400% FPL 1.3%** to purchase private coverage through health insurance 50–54^ 7.2% Marketplaces, and many do not qualify for Medicaid. Age Group 55–59 5.9% 8 A wide range of issues may contribute to higher uninsured 60–64 5.0%** rates for Hispanic 50- to 64-year-olds, including inadequate outreach and enrollment strategies, and insufficient White, non-Hispanic^ 5.3% mechanisms to address health insurance literacy challenges Race/ Black, non-Hispanic 6.3% and language and cultural differences (Blavin et al. 2014). Ethnicity Other, non-Hispanic 6.0% The ACA exclusion of undocumented immigrants from Hispanic 11.2%** enrollment in the Medicaid expansion and private coverage Excellent or very good^ 4.3% through health insurance Marketplaces is also likely to play a 7.2%* role in the higher uninsured rates, and even legal immigrants Health Status Good have faced difficulties with documentation requirements for Fair or poor 8.2%** coverage in the Marketplaces (Alonso-Zaldivar 2014). See also Zuckerman, Waidmann, and Lawton 2011. Source: HRMS, Q1 2015. 9 Higher uninsured rates among 50- to 64-year-olds in fair or Note: Data are not adjusted for differences in health poor health may be tied to the fact that many in this group status, income, or other characteristics across groups. are low income. As noted earlier in the paper, a significantly higher share of 50- to 64-year-olds in fair or poor health had ^ Denotes reference population. lower incomes. */** Estimate is significantly different from estimate for reference population at the 0.05/0.01 level using two-tailed tests. Insight on the Issues 101, October 2015 © AARP PUBLIC POLICY INSTITUTE 601 E Street, NW 1 Data come from the March 2015 Health Reform Monitoring Washington DC 20049 Survey (HRMS-AARP) oversample. See the Data and Methods Follow us on Twitter @AARPpolicy section of this report for more details on the survey. on facebook.com/AARPpolicy 2 The first Marketplace open enrollment period was October 1, www.aarp.org/ppi 2013, to March 31, 2014. For more reports from the Public Policy Institute, visit http://www.aarp.org/ppi/. 3 Due to preexisting conditions or age, many were denied private health insurance coverage in the nongroup market or charged unaffordable premiums (Weiss, Waidmann, and Caswell 2014). 4 The FPL is currently $11,770 for a single person and $15,930 for a household of two. 5 Challenges in estimating enrollment based on survey data should be recognized when evaluating changes in coverage status. Survey respondents may incorrectly identify 7