MAY 23, 2016
Uninsurance Rates and the Affordable Care Act
What does recent research show about changes in uninsurance rates since 2010?
|What's the issue?|
The uninsurance rate for nonelderly adults increased in the decade before the passage of the Affordable Care Act (ACA), driven by declining rates of employer-based coverage, especially during the recession at the end of the decade. The ACA was intended to decrease the percentage of the population without health insurance and to provide "quality, affordable health care for all." The purpose of this brief is to consider how uninsurance rates are changing under the ACA.
|What's the background?|
While reducing the number of uninsured people is just one measure of the ACA's effect, it is arguably the most important metric. Several government surveys can be used to study the number uninsured in the US population, including the Current Population Survey, the National Health Interview Survey, the American Community Survey, and the Medical Expenditure Panel Survey. The various surveys have different survey designs, field periods, health insurance coverage questions, reference periods, and survey modes, making the uninsurance estimates among each of the sources (see Exhibit 1).
(CPS). The CPS's Annual Social and Economic Supplement (ASEC) is conducted by the Census Bureau between February and April each year. Health insurance information has been collected for all household members since 1988, and the sample design makes it possible to produce state-level estimates. Initially, the CPS asked respondents whether members of the household had different types of insurance at any time in the previous calendar year--including employer-based coverage, privately purchased nongroup coverage, Medicare, Medicaid, and so forth. If respondents answered no to each question about coverage, they were assumed to be uninsured. Beginning in 2000, respondents who did not answer positively to any of the insurance coverage questions were asked explicitly if they or the members of their household were uninsured.
Prior to 2014, the CPS was designed to count only the number of people who were uninsured during the entire previous calendar year instead of those who were uninsured at the point the survey was conducted. Because CPS estimates of the share of full-year uninsured individuals were similar to other surveys' point-in-time uninsurance estimates, many whether respondents were forgetting about coverage they had the previous year, were actually answering questions about the coverage they currently had, or both. Beginning in 2014, a redesigned set of health insurance questions was introduced to address these recall errors and provide information on point-in-time coverage. The questions now start by asking about current health insurance coverage, when that coverage began, and which months they have had that coverage. In light of the redesigned instrument, the uninsurance estimates for calendar year 2013 are not directly comparable to estimates in previous years.
(NHIS). The NHIS is sponsored by the National Center for Health Statistics, part of the Centers for Disease Control and Prevention. Data collection for the NHIS began in 1957, with health insurance questions added in 1959, and significant revisions were made in 1997. The NHIS is fielded continuously throughout the year. In addition to questions about health insurance coverage, the NHIS asks a variety of questions on health status, health conditions, and use of services. It asks about a person's health insurance coverage at the point of the survey but also asks whether he or she did not have coverage at any point in the twelve months prior to the survey and how many months he or she went without coverage. Those who are uninsured at the time of the survey are asked how long it has been since they last had coverage. NHIS questions have a different format than the CPS. Respondents are first asked whether they are covered by any kind of health insurance or health care plan and then asked about the type(s) of coverage. The NHIS incorporates several verification questions as part of the instrument, including whether individuals have Medicaid if they do not initially report coverage. Health insurance estimates are available quarterly through the NHIS Early Release Program.
(ACS). The ACS, the newest of these government surveys, has been used to collect information on health insurance coverage since 2008. The ACS is conducted by the Census Bureau. Unlike the CPS, it collects data continually throughout the year. Respondents are asked if they are covered by any of several types of health insurance at the time of the interview. Unlike the CPS and the NHIS, the ACS does not include a verification question to follow up with people who do not report any insurance coverage to confirm that they are uninsured, nor does it include state-specific names for Medicaid or the Children's Health Insurance Program or a separate category for Marketplace coverage under the ACA. The ACS public use file is fifteen times larger than the CPS and about thirty times larger than the NHIS, with a sample of more than three million individuals. Because of its size and its sample design, both state-level and local-level estimates are available from the ACS.
(MEPS). Launched in 1996 and conducted by the Agency for Healthcare Research and Quality, MEPS collects information from households, medical providers, and employers on health insurance coverage, use of health care services, the cost of these services, and how they are paid for. The previous year's respondents to the NHIS constitute the sampling frame for the MEPS Household Component. Although MEPS has a smaller sample size than the other federal surveys, its longitudinal design (households are surveyed five times over two and a half years) provides unique information on transitions between different types of health insurance coverage and uninsurance.
Because of the time lag between data collection and release of estimates from federal surveys, several nonfederal surveys have been used to monitor trends in health insurance coverage following implementation of the ACA's key coverage provisions. These include the Gallup-Healthways Well-Being Index, which collects daily information, and the Health Reform Monitoring Survey, which provided quarterly data between early 2013 and March 2015 and semiannual data since then. Two additional private surveys providing less frequent estimates of the change in uninsurance are conducted by the and the .
. Conducted as part of daily Gallup Poll phone interviews of adults ages eighteen and older, the Gallup-Healthways Well-Being Index has collected information on health insurance since 2008. The large sample size (500 interviews per day since 2013; 1,000 interviews per day previously) and the frequency with which data are collected make it possible to provide quarterly estimates of uninsurance for the national population of adults and semiannual estimates of uninsurance by state. According to the , the share of nonelderly adults who were uninsured fell to a low of 12.9 percent in the first quarter of 2016, down 7.9 percentage points since the fourth quarter of 2013.
(HRMS). The HRMS, which receives its major funding from the Robert Wood Johnson Foundation, was designed by the Urban Institute to provide timely information on the ACA before data from federal surveys become available. Samples include approximately 7,500 nonelderly adults (ages 18-64) and approximately 2,400 children younger than age 18 each round. The HRMS does not provide state-level estimates, although some funders have supported state oversamples to obtain state-specific estimates. Questions on health insurance, access to care, health care affordability, and health status are based on those questions used in government surveys, including the CPS, the NHIS, and the ACS. A unique feature of the HRMS is that each round includes a changing set of topical questions focused on timely ACA policy and implementation issues.
|What's in the law?|
There are several provisions in the ACA aimed at reducing the percentage of people without insurance. For plan years effective after September 23, 2010, the ACA allows young adults to stay on their parents' health plan as dependents until age twenty-six. The law applies to both dependent children as well as children who are no longer considered a dependent for tax filing purposes; married children (although their spouse and any children would not be eligible); and both employer-based insurance plans and individually purchased plans, including Marketplace plans. Prior to this change in law, coverage for young adults on their parents' policies usually ended once they turned nineteen unless they were a full-time student. Full-time students were often allowed to remain on their parents' insurance policies until they graduated or turned twenty-two, whichever came first. At the time the law was passed, young adults had the highest rates of uninsurance and the lowest rates of employer-based coverage.
When the ACA was passed in March 2010, the law required states that had not already done so to expand their Medicaid programs in January 2014 to cover nearly all people with incomes below 138 percent of the federal poverty level (FPL) (currently about $16,400 for an individual and $27,800 for a family of three). However, in 2012 the US Supreme Court struck down that part of the ACA and made the Medicaid expansion optional for states. As of May 2016, thirty-one states and the District of Columbia have expanded their Medicaid programs, although a few states did not expand their programs until mid-2014 or later. Prior to the ACA, most states did not allow nondisabled childless adults to enroll in Medicaid, while other states covered parents only at very low income levels. In states that did not expand Medicaid, coverage for adults remains minimal. According to the , the median income limit for parents' eligibility in states that did not expand Medicaid is just 44 percent of the FPL, or about $8,870 for a family of three.
Along with states having the option of expanding Medicaid, the ACA created state Marketplaces for people to shop for and enroll in private health plans. Individuals and families with incomes between 100 percent and 400 percent of the FPL (currently up to $47,520 for an individual and $80,640 for a family of three) who are not eligible for Medicaid and do not have an affordable offer of employer-based coverage are eligible for federal tax credits to offset their health insurance premiums. Additional cost-sharing subsidies are available to those with family incomes below 250 percent of the FPL. Because the Medicaid expansion was intended to be mandatory, there is a gap in eligibility in states that did not expand their programs. People with incomes between their states' Medicaid eligibility thresholds and 100 percent of the FPL are not eligible for subsidized Marketplace coverage. As of early 2015, more than three million uninsured adults with incomes below 138 percent of the FPL were estimated to fall into a so-called coverage or --living in states that had not expanded Medicaid and not qualifying for either Medicaid or tax credits for Marketplace coverage.
In addition to the health insurance coverage expansions in the ACA, the law made significant changes to private insurance markets. Among the most significant changes are the elimination of preexisting condition exclusions and the ability to charge higher premiums based on medical history. Finally, the law requires people to have health insurance or pay a fine for remaining uninsured. For those who do not have coverage in 2016, the fine is the greater of $695 per adult and $347.50 per child (maximum of $2,085 per family) or 2.5 percent of household income (maximum of the national average annual premium for a bronze level plan).
|What's the debate?|
Measuring the number of people without insurance. There have long been debates about how best to measure the number of uninsured people. In addition to some of the issues discussed above, there is a fairly high number of people in a sample who do not respond or respond to only some portion of a survey. In the most recent CPS, for example, the was 13.4 percent for the core survey and an additional 14.2 percent for the ASEC, meaning more than a quarter of the people surveyed did not answer questions on health insurance. Nonresponse might produce a downward bias in CPS uninsurance estimates--meaning the reported number uninsured on the survey is lower than in reality. Another source of error is the underreporting of Medicaid, which has been shown to produce a small upward bias to uninsurance estimates, resulting in people who are actually on Medicaid being counted as uninsured.
Although concerns about the reliability of the different estimates are warranted, both the ACS and the NHIS show very similar estimates of how the uninsurance rate changed between 2013 and 2014, the period when the subsidized Marketplace plans became available and Medicaid was expanded in close to half the states (see Exhibit 2). In the ACS and the NHIS, the decline was nearly 3 percentage points, the largest one-year decline in uninsurance rates since 1997, based on prior NHIS data.
Young adult coverage. One of the least controversial components of the law was the requirement to allow young adults to remain on their parents' policies up until age twenty-six. Since this part of the law went into effect in the later part of 2010, the uninsurance rate for young adults has dropped significantly (see Exhibit 3). Although some of the decrease in uninsurance rates in this population can be attributed to Medicaid expansion or subsidized coverage through the Marketplaces and improvements in the economy, the dramatic drop in uninsurance rates between 2010 and 2013 appears to be driven in large part by the extension of dependent insurance coverage to young adults.
According to a recent , the uninsurance rate for young adults ages 19-25 declined from more than 30 percent in 2009 to 19 percent in 2014. The authors found a significant decline in the percentage of young adults without insurance from the last quarter of 2010 through late 2011 and a corresponding increase in the number with private coverage mirroring the time when the dependent coverage expansion first took effect. Other age groups had relatively stable insurance rates during that time period. Young adults experienced another significant drop in uninsurance rates in 2014, when the Medicaid expansion and subsidized Marketplace coverage became available.
State variations in uninsurance rates. Historically, uninsurance rates varied from state to state, driven in part by differences in income, education, employment, age, race, and state policies. With states opting out of the Medicaid expansion, state variations in uninsurance rates are more prominent than ever. From the third quarter of 2013 to the third quarter of 2015, the national uninsurance rate for adults ages 18-64 declined by almost 41 percent (see Exhibit 4), according to the HRMS. Medicaid expansion states saw uninsurance rates for adults cut by more than half; states that did not expand Medicaid saw a decrease of only one-quarter.
Of the ten states with the highest uninsurance rates in 2013, five chose not to expand Medicaid (Texas, Florida, Georgia, Oklahoma, and Mississippi), four chose to expand in 2014 (Arizona, California, Nevada, and New Mexico), and one (Alaska) expanded Medicaid in 2015. One year later, two of the Medicaid expansion states (Arizona and California) dropped out of the top ten states with the highest uninsurance rates, replaced with two states that did not expand Medicaid (Montana and Louisiana, both of which are expanding Medicaid in 2016). Looking at the data another way, the states with the greatest percentage-point decline in the percentage uninsured from 2013 to 2014 all expanded their Medicaid programs (see Exhibit 5).
The decline in the uninsurance rate from 2013 to 2014 when most coverage provisions went into effect is one measure of the ACA's effect on providing "quality, affordable health care for all." What remains to be seen is whether the uninsurance rate continues to decline in the coming years. Early release of NHIS data is promising. In 2015, 10.5 percent of the nonelderly population was uninsured. Overall, Gallup data indicate that an estimated twenty million adults have gained coverage since 2010, although not all of that decline can be attributed to the ACA and might be a result of the improving economy.
About 12.7 million people were enrolled in a Marketplace plan at the end of the third open enrollment period, up from 11.7 million at the end of the second enrollment period and 8.0 million at the end of the first open enrollment period. However, history has shown that not all people who select plans will pay their premiums to keep them, so the number of enrollees is expected to decrease throughout the year because of attrition. While the rate of increase in enrollment has slowed, it is too early to determine whether Marketplace enrollment has leveled off.
Despite the increased penalty in 2016 for not having insurance, the remaining uninsured people might be hard to enroll. In addition, a significant number of the remaining nonelderly uninsured people (about thirty-three million as of March 2015, according to the CPS) are likely ineligible for either Medicaid or subsidized plans through the Marketplace because they are undocumented immigrants. Recent briefs by the Henry J. Kaiser Family Foundation and Urban Institute estimate that between 15 percent and 16 percent of currently uninsured people are ineligible for coverage because of immigration status. The board of Covered California just proposed allowing undocumented immigrants to purchase unsubsidized health insurance through the California Marketplace, a move that will require state and federal approval before taking effect.
Finally, many of the states with the largest number of uninsured people have publicly expressed no intention at this time to expand Medicaid, leaving many poor adults without access to financial assistance for purchasing coverage. While improved outreach is crucial to reaching uninsured people who are eligible for Medicaid or subsidized Marketplace insurance but have not yet taken up coverage, continued progress in reducing the ranks of the uninsured will also depend on federal and state policy changes that expand the number of people who are eligible for assistance and ensure that assistance is sufficient to make coverage affordable.
|About Health Policy Briefs||
Editorial review by
Health Policy Briefs are produced under a partnership of Health Affairs and the Robert Wood Johnson Foundation.
Cite as: "Health Policy Brief: Uninsurance Rates and the Affordable Care Act," Health Affairs, May 23, 2016.
Sign up for free policy briefs at: www.healthaffairs.org/healthpolicybriefs
- From The Editor-In-Chief
- Delivery System Innovation
- Entry Point: Integrated Care
- Eye On Health Reform
- Innovative Environments
- Successful Diabetes Prevention
- Quality Oncology Care Models
- Aging And Health: Avoiding Hospitalizations
- Reducing Early Deliveries In Texas Medicaid
- Telehealth Increases Access, Possibly Spending
- Los Angeles Improves Access To Specialists
- Patient-Centered Medical Homes Research
- Costa Rica's Primary Care Innovation
- Narrative Matters: Patient-Doctor Bonds
- View Table of Contents »
- Our Climate Change And Health “Moment”: How Philanthropy Can Help 30 Mar 2017
- ACA Round-Up: Bill Would Allow Use Of Tax Credits For Off-Marketplace Plans And More 30 Mar 2017
- Help Wanted: Cost-Effectiveness Research For Health And Wellness 30 Mar 2017
- Health Affairs Web First: ACA Risk Adjustment And Reinsurance Were Well Targeted In First Two Years 29 Mar 2017
- Transgender Health Equity: Patients Say Providers Lack Expertise And Understanding Of Transgender Health Needs 29 Mar 2017
Now Playing The Work/Health RelationshipBriefing February 07, 2017
- When Patients Mentor Doctors: The Story Of One Vital Bond March 06, 2017
- The Work/Health Relationship Forum February 22, 2017
- The Work/Health Relationship February 07, 2017
- The Personal Toll Of Practicing Medicine February 06, 2017
- In Opioid Withdrawal, With No Help In Sight January 09, 2017
- Health Affairs CA Forum: Delivery System Innovation March 22, 2017
- Delivery System Innovation March 07, 2017
- Work And Health Forum February 22, 2017
- Work And Health February 07, 2017
- Direct-To-Consumer Telehealth May Increase Access To Care But Does Not Decrease Spending
- Synthesis Of Research On Patient-Centered Medical Homes Brings Systematic Differences Into Relief
- In Mexico, Evidence Of Sustained Consumer Response Two Years After Implementing A Sugar-Sweetened Beverage Tax
- Medical Groups' Adoption Of Electronic Health Records And Information Systems
- Primary Health Care That Works: The Costa Rican Experience
- Providing More Home-Delivered Meals Is One Way To Keep Older Adults With Low Care Needs Out Of Nursing Homes
- When Patients Mentor Doctors: The Story Of One Vital Bond
- Delivery System Innovation
- Direct-To-Consumer Telehealth May Increase Access To Care But Does Not Decrease Spending
- The Triple Aim: Care, Health, And Cost