ACA Impact on Uninsured: Survey Estimates

VII. Key Issues: Regulation & Reform >> C. Health Reform >> Affordable Care Act (ACA) >> ACA Impact Analysis >> ACA Impact on Access >> Impact on Uninsured >> Survey Estimates (last updated 6.17.16) 

Overview

While administrative data in principle might provide a more timely snapshot of how many previously uninsured obtain coverage, such information is not readily available for Exchange coverage, off-Exchange non-group coverage, employer-based coverage or Medicaid. Consequently, the most complete and accurate information will come from survey data. Of the sources cited below, both the ACS data (for the prior calendar year) and CPS data (from the March survey, but purportedly characterizing insurance coverage as of the preceding year) are released only once a year in the fall. In contrast, the remaining surveys report more frequently, but also are much smaller and less reliable than either the ACS or CPS. The NHIS and Gallup figures are reported quarterly, while RAND and Urban estimates are reported more sporadically.

Baseline Estimates

  • Commonwealth FundAmerica’s Underinsured: A State-by-State Look at Health Insurance Affordability Prior to the New Coverage Expansions (March 2014). Using 2012-2013 Current Population Survey data, this state profiles report (3.20.14) provides pre-2014 baseline estimates of the number of uninsured and underinsured for each state (counts and % of population) with comparisons to national averages. A separate table breaks down these figures by poverty status (<100%, 100-199%, 200-399%, 400% or more).
  • Kaiser Family FoundationHealth Coverage & UninsuredState Health Facts. Urban Institute and Kaiser Commission on Medicaid and the Uninsured have partnered to produce estimates based on the Census Bureau’s March 2012 and 2013 Current Population Survey (CPS: Annual Social and Economic Supplements). The most recent tabulations provided a snapshot based on pooled estimates for each state for 2011-2012. Breakdowns are available for a variety of difference demographic characteristics and family poverty status.

National Estimates

Government Population Surveys

There are several large population surveys that can be used to calculate reductions in the uninsured. The surveys are listed in order of their sample size. The following figures were derived at the SHADAC Data Center.

American Community Survey (ACS)

The ACS is by far the largest survey that asks health insurance questions: designed to replace the decennial Census with a continuous rolling survey, the 2012 ACS interviewed 2.375 million people in households and 154,000 people in group quarters (thus it represents the entire U.S. population). The ACS collects 12 independent monthly samples (over 200,000 persons interviewed monthly). Annual ACS estimates are released every September (3- and 5-year estimates also are available, but nothing less than 1 year due to the nature of the sampling frame). In any given month, the ACS asks about coverage on the day of the survey, but because this question is asked monthly of different people, the ACS estimate of the uninsured is more akin to a monthly average for a given calendar year.

  • Accuracy.  The overall response rate in 2012 was 97.3%. The margin of error for the 2012 ACS uninsured estimate is 0.1%. However, there are two important limitations of the survey. First, unlike CPS, NHIS and MEPS, there is no uninsured verification question (people who appear to be uncovered due to their failure to report various types of coverage are not directly asked to confirm whether they are uninsured). Second, state-specific names for Medicaid/CHIP programs are not used; this apparently results in people underreporting Medicaid coverage. There also is some overreporting of non-group coverage, though some researchers have found methods to adjust for both Medicaid underreporting and non-group overreporting.
  • Reductions from 2010 to 2012. The point-in-time estimate of the number of uninsured fell from 47,173,439 (15.4%) in 2010 to 45,465,027 (14.7%) in 2012. The observed decline is statistically significant even though the relative reduction in uninsured risk (4.5%) is not very large. The 0.7% decline applied to the 2012 population implies a net reduction of 2.2 million uninsured between 2010 and 2012.
  • Most Recent Estimates.
    • SHADAC provides summary trend tables for the nation and state in 2014 ACS Tables: State & County Uninsurance Estimates (September 2015).  For the entire population, the uninsured rate has declined from 15.1% in 200815.1% in 200915.5% in 2010, 15.1% in 2011 to 14.8% in 2012 to 14.5% in 2013 and 11.7% in 2014.  When the annual reductions in the rate of being uninsured are applied to the 2014 total population base (313.9 million), they imply that the daily number of uninsured grew by 1.3 million between 2009 and 2010 but has been reduced by 1.3 million in 20110.9 million in 2012, another 0.9 million in 2013 and an additional 8.8 million in 2014 for a cumulative total of 11.9 million since 2010 or 10.7 million if either 2008 or 2009 is used as the base year.
    • Smith, Jessica C. and Carla Medalia, U.S. Census Bureau, Current Population Reports, P60-253, Health Insurance Coverage in the United States: 2014, U.S. Government Printing Office, Washington, DC, 2015. This report includes latest findings from ACS and CPS. Several figures/tables are derived exclusively using ACS data:
      • Fig. 1. Uninsured Rate: 2008 to 2014
      • Fig. 4. Uninsured Rate by Single Year of Age: 2013 and 2014
      • Fig. 5. Uninsured Rate by Poverty Status and Medicaid Expansion of State for Adults
        Aged 19 to 64 Years: 2013 and 2014
      • Fig. 7. Uninsured Rate by State: 2014
      • Fig. 8. Decrease in the Uninsured Rate by State: 2013 to 2014 (details in Table A-1
        Population Without Health Insurance Coverage by State: 2013 and 2014)

Current Population Survey (CPS)

CPS monthly surveys the civilian non-institutionalized population. The Annual Social and Economic Supplement of the CPS collects information on insurance coverage.  The Census Bureau conducts the ASEC over a 3-month period, in February, March, and April, with most data collection occurring in the month of March. All told, including households surveyed in February and April, 99,000 households containing roughly 250,000 people were included in the 2013 ASEC.

  • Accuracy. The non-response rate for ASEC is 20.0%. The margin of  error in 2012 was 0.3%. CPS asks about coverage in the prior calendar year; thus the count of uninsured purportedly represents people uninsured the entire year, but people appear to respond as if they are reporting their coverage as of the time of the survey.
    • Health insurance coverage is likely to be underreported on the Current Population Survey (CPS). While underreporting affects most, if not all, surveys, underreporting of health insurance coverage appears to be a larger problem in the Annual Social and Economic Supplement (ASEC) than in other national surveys that ask about insurance…The CPS ASEC data underreport Medicare and Medicaid coverage compared with enrollment and participation data from the Centers for Medicare and Medicaid Services (CMS). Because the CPS is largely a labor force survey, interviewers receive less training on health insurance concepts than labor concepts. Additionally, many people may not be aware that a health insurance program covers them or their children if they have not used covered services recently.” Income, Poverty, and Health Insurance Coverage in the United States: 2010. DeNavas-Walt, Carmen, Proctor, Bernadette D., Smith, Jessica C., U.S. Census Bureau, September 2011.
  • Reductions from 2010 to 2013. Using SHADAC-Enhanced CPS figures, the uninsured rate for both 2009 (from Spring 2010 survey) and 2010 was 15.9%. In 2012 (Spring 2013 survey), it was 15.0%. The corresponding numbers are 48,308,822 (Spring 2010), 48,826,985 (Spring 2011) and 46,812,709 (Spring 2013), so the net change in number of uninsured depends on whether the figures represent prior year coverage or point-in-time coverage.  The observed decline since 2009/2010 was statistically significant even though the relative reduction in uninsured risk (5.7%) is not very large. The 0.9% decline applied to the spring 2013 population implies a net reduction of 2.8 million uninsured between spring 2010 and spring 2013.
  • Reductions from 2013 to 2014. The CPS data release in September 2014 normally would have allowed calendar year 2012 to be compared to calendar year 2013 (or spring 2014 to spring 2013, depending on how one interprets the responses). But no such comparisons were included for the following reason:
  • Most Recent Estimates.
    • Smith, Jessica C. and Carla Medalia, U.S. Census Bureau, Current Population Reports, P60-253, Health Insurance Coverage in the United States: 2014, U.S. Government Printing Office, Washington, DC, 2015. This report includes latest findings from ACS and CPS. Several figures/tables are derived exclusively using CPS data:
      • Fig. 1. Uninsured Rate: 2008 to 2014 (includes CPS figures for uninsured the entire year for 2013 and 2014 only)
      • Fig. 2. Percentage of People by Type of Health Insurance Coverage and Change From Last Year: 2014
      • Table 1. Coverage Rates by Type of Health Insurance: 2013 and 2014
      • Table 2. Percentage of People by Type of Health Insurance Coverage by Age: 2013 and 2014 (details in Table A-2. Number of People by Type of Health Insurance Coverage by Age: 2013 and 2014)
      • Table 3. Percentage of People by Type of Health Insurance Coverage for Working-Age Adults: 2013 and 2014 (details in Table A-3. Number of People by Type of Health Insurance Coverage for Working-Age Adults: 2013 and 2014)
      • Table 4. Percentage of People by Type of Health Insurance Coverage by Household Income and Income-to-Poverty Ratio: 2013 and 2014 (details in Table A-4. Number of People by Type of Health Insurance Coverage by Household Income and Income-to-Poverty Ratio: 2013 and 2014)
      • Table 5. Percentage of People by Type of Health Insurance Coverage by Selected Demographic Characteristics: 2013 and 2014 (details in Table A-5. Number of People by Type of Health Insurance Coverage by Selected Demographic Characteristics: 2013 and 2014)

National Health Interview Survey (NHIS) 

NHIS surveys 35,000 households containing about 87,500 persons from the civilian non-institutionalized population continuously throughout the year, reporting quarterly estimates of the uninsured for the entire population.  

  • Accuracy. The annual response rate of NHIS is close to 90%. The margin of error for the 2012 figure was 0.45% (based on multiplying reported standard error by 1.96). The NHIS reports 3 different uninsured measures: a) uninsured at the time of interview; b) uninsured at least part of the year prior to interview (which includes persons uninsured for more than a year); and c) uninsured for more than a year at the time of interview.
  • Reductions from 2010 to 2014. The point-in-time estimate of the uninsured rate has shown a steady decline from from 16.0% in 2010 to 15.1% in 2011, 14.7% in 2012, 14.4% in 2013, 11.5% in 2014 and 9.0% in the first half of 2015 (Table I); a similar decline has occurred for those uninsured the entire year.  Trends for children and adults 18-64 are reported in the same table.
    • Between 2013 and 2014, the daily number of uninsured declined by 8.8 million (Table II).
    • Between 2014 and first half of 2015, the daily number of uninsured declined an additional 7.5 million (Table II).
    • Thus, the total decline in daily uninsured has been 20.1 million since 2010 or 16.3 million since 2013.  According to Timothy Jost, “Of course, some of the reduction of the uninsured is simply due to more people returning to work and gaining job-based coverage. But some of it is certainly also due to the ACA.”
  • Long-Term Trends. The Council of Economic Advisors has synthesized a variety of sources to estimate an NHIS-consistent uninsured rate back to 1963 when the national uninsured rate was 24.3% (see Table on p. 11). The 11.3% figure in fourth quarter 2014 is the lowest figure during that entire period (the rate was 11.4% in 1974, 1978 and 1980), but as noted above, this is somewhat misleading since the  annualized rate for 2014 is more likely to be 11.6%. Nevertheless, with the exception of the massive decline in uninsured risk between 1963 and 1974 (when the overall uninsured was more than halved), the decline in the uninsured rate from 2010-2014 was faster and steeper than at any other period using this metric (Figure 3).
  • Future Estimates. Figures through June 2015 were reported in November 2015. Estimates for calendar year 2014 were reported in mid-June 2015. Early Release Reports on Detailed Estimates of Health Insurance Coverage are released quarterly.
  • National Center for Policy Analysis (June, 2016). “The best measurement of people who lack health insurance, the National Health Interview Survey published by the Centers for Disease Control and Prevention (CDC), has released early estimates of health insurance for all fifty states and the District of Columbia in 2015. There are two things to note. First: About 70 percent of residents, age 18 to through 64, had ‘health insurance’ in 2015, which is the same rate as persisted until 2006. Obamacare has not achieved a breakthrough in coverage. It has just restored us to where we were less than a decade ago. What has also happened is a significant change from private coverage to government welfare (primarily Medicaid). The shift has been about five percentage points since 2006, and ten percentage points since 1997.… Between 2013 and 2015, the number of people who were uninsured for one year or more declined by 12.7 million, from 30.5 million to 17.8 million. However, the number uninsured for less than a year increased slightly from 16.9 million to 17.7 million.” (NCPA, 6.9.16)

Private Surveys

There are several other much smaller surveys that provide more frequent snapshots of the population and/or focus on those most likely to enroll in the Exchanges or Medicaid. These are listed in order of the length of time these surveys have been conducted.

Gallup-Healthways Well-Being Index

Since 2008, Gallup has measured uninsured status among adults 18 and older.

  • Accuracy.  Interviews are conducted with respondents on landline telephones and cellular phones, with interviews conducted in Spanish for respondents who are primarily Spanish-speaking. Each sample of national adults includes a minimum quota of 50% cellphone respondents and 50% landline respondents, with additional minimum quotas by time zone within region. Landline and cellular telephone numbers are selected using random-digit-dial methods. Landline respondents are chosen at random within each household on the basis of which member had the most recent birthday. For results based on the total sample of national adults, the margin of sampling error is ±1 percentage point at the 95% confidence level.
  • ASPE Reports. The DHHS Office of Assistant Secretary of Planning and Evaluation has relied on the Gallup surveys to estimate the net change in coverage attributable to the Affordable Care Act. They use a baseline period is from Q1 2012 to Q3 2013. All models use nationally-representative survey weights and adjust for age, sex, race, ethnicity, employment, state of residence, marital status, rural location, and a linear time trend in order to control for changes in the economy, population composition, and non-policy factors affecting health insurance coverage.
    • ASPE. Health Insurance Coverage and the Affordable Care Act (9.22.15). A total of  17.6 million uninsured people have gained health insurance coverage as of 9.12.15. This includes 15.3 million adults who gained health insurance coverage since the beginning of open enrollment in October 2013 through September 12, 2015. Over that period, the uninsured rate declined from 20.3 percent to 12.6 percent — a 38 percent (or 7.7 percentage point) reduction in the uninsured rate. An additional 2.3 million young adults age 19-25 had previously gained coverage before October 2013. There are earlier versions of this report, but they rely on a different methodology, so no results are reported here.
      • The latest report provides the estimated gains in coverage since the beginning of open enrollment: as of Q1 2014 (4.3m.), Q3 2014 (10.3m.), Q1 2015 (14.3m.) and Q3 2015 (15.3m.)–to which must be added the 2.3 million young adults who gained coverage prior to the first open enrollment period.
      • However, given that the subsequent Gallup survey found that the Q4 2015 uninsured rate was identical to Q1 2015 (see Latest Report below), the net gain in coverage is only 14.3 m. + 2.3 million young adults=16.3 million.
    • ASPE. Health Insurance Coverage and the Affordable Care Act (5.5.15).
  • Latest Report. U.S. Uninsured Rate 11.9% in Fourth Quarter of 2015 (1.7.16). The most recent reported results are based on telephone interviews conducted as part of the Gallup-Healthways Well-Being Index survey July 1-September 30, 2015, with a random sample of 42,998 adults, aged 18 and older, living in all 50 U.S. states and the District of Columbia.
    • Uninsured rate in fourth quarter 2015 is 11.9%, identical to first quarter 2015 level.
    • “The sharp drop in the uninsured rate seen in the first year after the insurance exchanges opened has leveled off in the second year, with smaller declines seen in 2015 compared with 2014. This validates concerns that similarly large reductions may not be possible in the future because the remaining uninsured are harder to reach or less inclined to become insured more generally.”
    • Uninsured rate has dropped 5.2 percentage points since 4th quarter 2013. The 5.2% decline applied to the spring 2014 population of adults age 18 and older (247.2m) implies a net reduction of 12.9 million uninsured.

Urban Institute: Health Reform Monitoring Survey (HRMS)

HRMS has been tracking insurance coverage since the first quarter of 2013.

  • Accuracy. All survey results were obtained using The KnowledgePanel®, a nationally representative, probability-based Internet panel maintained by GfK Custom Research. Approximately 55,000 people in the panel included households with and without Internet access; laptops and free Internet access are provided to those who lack their own Internet access to ensure their participation in the panel.
    • The American Association for Public Opinion Research (AAPOR) cumulative response rate for the HRMS is the product of the panel household recruitment rate, the panel household profile rate, and the HRMS completion rate2—roughly 5 percent each quarter. Thus the HRMS carries with it more risks and potential errors than federal government surveys. A number of potential sources of bias, including nonresponse bias, likely are only partly mitigated through the survey weighting
    • The quarter 3 2013 HRMS has a a sampling margin of error for a 50 percent statistic with 95% confidence of +/- 1.3 for the nonelderly adult sample.
  • June 2014 Results. These were reported 7.10.14:
    • The uninsurance rate for nonelderly adults (age 18–64) was 13.9 percent (95% CI 1) for the nation in June, a drop of 4.0 percentage points (95% CI 2) since September 2013.
    • This represents a drop of 22.3 percent in the uninsurance rate, which translates to a net gain in coverage for about 8.0 million adults (95% CI 3).
  • March 2014 Results. These were reported 4.3.14:
    • Analysis of data from the March 2014 HRMS shows the uninsurance rate for nonelderly adults (age 18–64) was 15.2 percent for the nation, a drop of 2.7 percentage points since September 2013, the month before ACA open enrollment began. This represents a gain in coverage for about 5.4 million adults (95% CI 4).
    • States that implemented the ACA’s Medicaid expansion saw a larger decline: their uninsurance rates for adults dropped 4.0 percentage points since September, compared with a drop of 1.5 percentage points for the nonexpanding states (all changes significant at .05 level or better).
    • Limitations. These early estimates understate the full effects of the ACA on the uninsured for two major reasons. First, the survey does not capture the enrollment surge that occurred at the end of the open enrollment period, because 80 percent of the responses to the March 2014 HRMS were provided by March 6, 2014. Second, these estimates do not reflect the effects of some important ACA provisions (such as the ability to keep dependents on health plans until age 26 and early state Medicaid expansions) that were implemented before 2013.

Commonwealth Fund Survey

July, 2014. The Commonwealth Fund Affordable Care Act Tracking Survey interviewed a nationally representative sample of 19-to-64-year-old adults, including a sample of individuals who were potentially eligible for the law’s new coverage option. The first was conducted by SSRS from July 15 to September 8, 2013, by telephone among a random, nationally representative U.S. sample of 6,132 adults ages 19 to 64.  In the fourth survey, the survey firm SSRS interviewed 4,425 adults by telephone from April 9 to June 2, 2014.

  • The sample for the April–June 2014 survey was designed to increase the likelihood of surveying respondents who were most likely eligible for new coverage options under the ACA. As such, respondents in the July–September 2013 survey who said they were uninsured or had individual coverage were asked if they could be recontacted for the April–June 2014 survey.
  • SSRS also recontacted households reached through their omnibus survey of adults who were uninsured or had individual coverage prior to open enrollment. The data are weighted to correct for the stratified sample design, the use of prescreened and recontacted respondents from earlier surveys, the overlapping landline and cellular phone sample frames, and disproportionate nonresponse that might bias results. The resulting weighted sample is representative of the approximately 186.1 million U.S. adults ages 19 to 64.
  • Accuracy.  The first survey had an overall margin of sampling error of +/– 1.8 percent at the 95 percent confidence level.  The survey has an overall margin of sampling error of +/– 2.1 percentage points at the 95 percent confidence level. The landline portion of the main-sample survey achieved a 19 percent response rate and the cellular phone main-sample component achieved a 15 percent response rate
  • Key Findings.
    • The uninsured rate for people ages 19 to 64 declined from 20 percent in the July-to-September 2013 period to 15 percent in the April-to-June 2014 period.
    • An estimated 9.5 million fewer adults were uninsured in the 2014 compared to the 2013 survey.
  • Analysis: The Truth Behind the Obamacare Sign-up Numbers. “The Commonwealth study does not differentiate between people who were previously insured and those who never had health insurance. We don’t know how many of the 20 million had their prior insurance cancelled or changed significantly, forcing them into Medicaid or an exchange or onto their parents’ plans. We also don’t know how many of the 20 million actually paid their new insurance premiums. We only know the number of people who signed up for health insurance. Breaking down the categories of people in the 20 million is also revealing. One million were young adults, under the age of 26, who were placed on their parents’ plans. Six million enrolled in Medicaid or S-CHIP which are taxpayer-funded entitlements. We don’t know how many of these individuals were previously eligible for the programs prior to the passage of the ACA. Five million purchased health insurance in the private market, which basically has nothing to do with Obamacare. That leaves eight million who purchased insurance through the newly-created exchanges. When Obamacare became law in 2010, fifty million Americans did not have health insurance for one reason or another. Taxpayers are now paying trillions of dollars, are enduring a huge health care regulatory burden, and are being misled about Obamacare enrollment numbers.” (Washington Policy Center, 7.11.14)

January, 2015. Findings from the Commonwealth Fund Biennial Health Insurance Survey, 2015. (1.15.15) The sample was designed to generalize to the U.S. adult population and to allow separate analyses of responses of low-income households. This report limits the analysis to respondents ages 19 to 64 (n=4,251). Statistical results are weighted to correct for the stratified sample design, the overlapping landline and cellular phone sample frames, and dis- proportionate nonresponse that might bias results. The data are weighted to the U.S. adult population by age, sex, race/ ethnicity, education, household size, geographic region, population density, and household telephone use, using the U.S. Census Bureau’s 2013 Annual Social and Economic Supplement. The resulting weighted sample is representative of the approximately 182.8 million U.S. adults ages 19 to 64.

  • Accuracy.  The survey has an overall margin of sampling error of +/– 2 percentage points at the 95 percent confidence level. The landline portion of the survey achieved a 15.8 percent response rate and the cellular phone component achieved a 13.6 percent response rate.
  • Key Findings.

RAND Survey

Kaiser Family Foundation Survey 

The Kaiser Family Foundation Survey of Non-Group Health Insurance Enrollees is a series of surveys taking a closer look at the entire non-group market beginning in April, 2014.

  • June Report: Early April to Early May 2014. This survey of 742 individuals finds that 57% of those who bought insurance on the exchanges were previously uninsured. Seven of every ten of this group purchased insurance because of the ACA’s mandates. (Kaiser Family Foundation, June, 2014)

Estimates for Young Adults Age 19-25

Because coverage of young adults through parental plans was one of the first ACA provisions to go into effect (September 2010), initial examinations of changes in uninsured risk naturally focused on this large and identifiable population.

National Health Interview Survey

Because it provides quarterly estimates of coverage, the NHIS has been used to evaluate the impact of the dependent adult mandate on coverage for those 19-25.

  • ASPE Estimates. The Office of Assistant Secretary for Planning and Evaluation in DHHS has issued several reports on this issue.
    • Sommers, Benjamin D. Number of Young Adults Gaining Insurance Due to the Affordable Care Act Now Tops 3 Million. ASPE Issue Brief, June 19, 2012. From September 2010 (3rd quarter 2010) to December 2011 (4th quarter 2011), the percentage of adults 19 to 25 with insurance coverage increased from 64.4% to 74.8% (based on NHIS estimates). This 10.4% increase was applied to 29.692 million adults age 19-25 reported in the March 2011 Current Population Survey (Table 8) to estimate 3.1m. gained coverage.
    • ASPE.  State-Level Estimates of Gains in Insurance Coverage Among Young Adults (6.19.12). New results from the National Health Interview Survey (NHIS) indicate that 3.1 million additional young adults have insurance coverage as of December 2011, due to the provision in the Affordable Care Act that allows 19 through 25 year olds to remain on their parents’ insurance plans. This report provides a state-by-state breakdown using American Community Survey estimates of the number of uninsured young adults age 19-25 in each state in 2009 to allocate the gains by state.
    • ASPE. Health Insurance Coverage and the Affordable Care Act (9.22.15). Again, using NHIS data, from the baseline period through the start of open enrollment in October 2013, the uninsured rate for young adults declined from 34.1 percent to 26.7 percent, which translates to 2.3 million young adults gaining coverage.
  • CDC Estimates. 
    • Reductions from 2010 to 2015. The point-in-time estimate of the uninsured rate for adults age 19-25 has shown a fairly steady decline from from 33.9% in 2010 to 27.9% in 2011, 26.4% in 2012, 26.5% in 2013, 20.0% in 2014 and 15.9% in the first half of 2015 (Table I); a similar decline has occurred for those uninsured the entire year.
      • Between 2013 and 2014, the daily number of uninsured age 19-25 declined by 2.0 million (Table II).
      • Between 2014 and first half of 2015, the daily number of uninsured age 19-25 declined an additional 1.1 million (Table II).
      • Thus, the total decline in daily uninsured has been 5.1 million since 2010 or 3.1 million since 2013.
  • Kessler, GlennThe Fact Checker (3.17.14). Using the CDC report cited above, the author estimates “the overall increase in the percentage of insured young adults suggests a gain of about 2.8 million from the third quarter of 2010 to the first half of 2013.” Given quarterly fluctuations, a more conservative approach would be to use the average figure for 2010 as the base; that would reduce the increase to 2.2 million people.
  • Hogberg, DavidAmerican Spectator (4.2.14). This analysis shows that the 3.1m was overstated since it attributed all gains in coverage to the dependent adult mandate, ignoring gains in public coverage having nothing to do with that mandate. If the latter coverage gains are excluded, the net increase in coverage drops to 2.8 million.
    • Author also notes that earlier versions of CDC report had found the mandate had reduced the number of uninsured by 1 million (September 2011 report) to 2.5 million (December 2011).
    • Using Q3 figures for 2013 from the NHIS, the gain in private coverage for 19-25 year olds is only 2.2 million.
    • The CPS shows that between 2009 and 2012 (i.e., spring 2010 and spring 2013 surveys), the number of uninsured 18-24 year olds declined by 976,000, but 276,000 was due to higher Medicaid enrollment, so the maximum potential gain through parental plans is 700,000. These include a gain of 447,000 in employer-based coverage (the published figures do not distinguish between own coverage at work and being covered as a dependent on a parents’ policy, so it really is unknown what share of that totally truly can be attributed to the mandate).
  • Roy, Avik. The Apothecary (4.03.14). Census Data: Since 2008, There’s Been No Net Change In The Proportion of Young Adults With Health Coverage.  This analysis argues “it’s far from clear that the NHIS survey is accurate. A much larger survey—this one from the U.S. Census Bureau—indicates that the net proportion of young adults with private health coverage is unchanged since 2008. The NHIS survey uses data from roughly 75,000 to 80,000 individuals of all ages. The Census survey, called the Current Population Survey or CPS, is more than three times larger; the CPS uses a sample of around 100,000 households, or around 250,000 individuals.”
    • Between 2008 and 2012, the CPS shows that the percentage of young adults age 18-24 with private coverage has remained constant at 60.5% implying no net gain in coverage attributable to the young adult dependent mandate.
    • Using 2010 (the same baseline used in the ASPE study above), the net gain in private coverage is 2.1 percentage points, implying a net gain in coverage of 630,000 when applied to a 2012 population base.
    • Since 2008, CPS figures show the proportion of 25-34 year olds with private health coverage has declined by 3.1 percentage points (64.9% in 2008 to 61.8% in 2012). If ACA is credited with preventing this from happening to 18-24 year olds, it implies 931,000 gained coverage due to the dependent adult mandate.

American Community Survey

Using the ACS, the point-in-time estimate of the number of uninsured young adults age 19-25 declined from 9,541,264 in 2010 to 8,017,137 in 2012–a net decline of 1.5m that is only half as large as the figure obtained from NHIS (if measured from 2009, the decline is larger by only 30,000 people). Uninsured risk declined from 31.8% in 2010 to 26.0% in 2012 (margin of error=0.3 for both years).  If the reduction in uninsured risk is applied to the 2012 population of young adults, it implies that there were 1.8m fewer uninsured young adults as a consequence of this reduction in risk.

  • The ACS is a much bigger survey than NHIS, covering 3.5 million people each year; hence the NHIS sampling error is 3.5 times as large as ACS.
  • The ACS asks about coverage on the day of the survey; however, unlike the NHIS, ACS does not use a verification question to confirm uninsured status (while this may produce some errors in measuring the number of uninsured at any given time, it should not materially affect estimates of changes over time since an upward or downward bias in measuring the number of uninsured in any given point in time would affect both measured periods, hence would largely cancel out when subtracting one year from another etc.).

Current Population Survey

Using SHADAC-enhanced CPS figures (which correct some of the known inaccuracies in the regular CPS data), the estimate of the number of all-year uninsured young adults age 19-25 declined from 9,421,442 in 2009 to 8,361,983 in 2012–a net decline of 1.1 million. Uninsured risk declined from 32.2% to 27.7% during that period (margin of error 0.9% in both years). However, if the reduction in uninsured risk is applied to the 2012 population of young adults, it implies that there were 1.4m fewer uninsured young adults as a consequence of this reduction in risk.

The number in this age group with private coverage grew from 55.9% in 2009 (March 2010 survey) to 60.2% by 2012 (March 2013). If this 4.3 percentage point gain is applied to the count of young adults 19-25 reported for 2012 (30.702m.), it implies there were 1.4m more young adults with private coverage. However, adults age 26-34 also saw a slight increase in private coverage during this period (from61.8% to 62.0%), so if this difference is subtracted (on grounds that it reflects improvements in the economy rather than ACA), the net gain in private coverage for 19-25 year olds is 4.1 percentage points–an additional 1.3m. with private coverage.

  • The CPS is a much bigger survey than NHIS (hence the NHIS sampling error is 1.75 times as large as CPS), though it has a longer recall period and responses are imputed for about 10% of CPS respondents (versus almost no imputation for NHIS).
  • Note that the CPS asks about coverage during the entire preceding year, hence the number of uninsured technically should represent a count of people without coverage the entire year. However, based on comparing CPS figures with those from other surveys, “many researchers believe that the CPS measure is closer to a point in time measure of uninsurance than a full-year measure.”  If the CPS figures are viewed as point-in-time figures, it means the 2009 estimates actually represent coverage as of February-April 2010, which is when the Annual Social and Economic Supplement survey is conducted. The 2012 figures thus would have represented a snapshot as of spring 2013.
  1. 3, 15.4
  2. 6, 5.5
  3. 1 million, 10.8 million
  4. 2 million, 7.6 million

One Response to ACA Impact on Uninsured: Survey Estimates

  1. Pingback: Did Obamacare reduce un-insurance in 2015, or not? | Insurance News Depot

Leave a Reply

Your email address will not be published. Required fields are marked *