Looking Behind Opinion Polling on the Affordable Care Act

 

John Holahan, Genevieve M. Kenney, Michael Karpman, and Ariel Fogel

February 26, 2014

 

At a Glance
  • Americans most likely to have an unfavorable view of the ACA are in the middle- and higher-income groups, have private insurance, are in very good/excellent health, are white, and live in rural areas. Those most likely to have no opinion are the groups most likely to benefit from the law—those in fair/poor health, those with lower incomes, the uninsured, nonwhites and Hispanics, high school graduates (or less), and the young.
  • Those living in states that have adopted the Medicaid expansion or that have a state-based Marketplace were less likely to have an unfavorable opinion of the law than those living in other states.
  • Those who said they were affected by at least one of the early benefit provisions of the law had a more favorable opinion of the ACA than those who said they had not been so affected (or were unsure).
  • Large proportions of uninsured adults, particularly whites and middle-income adults, were more likely to have an unfavorable than a favorable opinion of the law. In addition, the uninsured most likely to benefit from the ACA (e.g., lower- and middle-income adults, including nonwhites and Hispanics) expressed only weak support for it and were more likely to have no opinion than to have a favorable view.

The Affordable Care Act (ACA) introduces new policies designed to increase access to affordable health insurance coverage and medical care. While some of its provisions were implemented within a year or two of its passage in 2010, most of the major coverage changes are being implemented in 2014—including the state option to expand Medicaid, the introduction of new health insurance Marketplaces with accompanying market reforms and coverage subsidies, and a mandate requiring individuals to be insured. These changes combined are expected to greatly reduce the number of Americans who are uninsured, with particularly high reductions in states choosing to expand Medicaid.

 

This brief looks in detail at what different groups of nonelderly adults (age 18–64) think about the ACA. Public opinion across several polls has been unfavorable toward the ACA and grew increasingly so in the wake of the troubled roll out of the new Marketplaces.1 The analysis draws on data collected in September 2013 from the Health Reform Monitoring Survey (HRMS). The data reflect opinions prior to the first open enrollment period, which began on October 1, 2013. Because the sample includes almost 8,000 adults, we are able to probe more deeply than prior efforts, assessing (1) how opinions about the ACA vary with the person’s demographic, health, and socioeconomic characteristics and (2) whether experience with the early provisions of the law has shaped their perceptions. Future HRMS surveys will allow us to assess any changes in these initial patterns.

 

What We Did

 

The September 2013 HRMS asked nonelderly adults the following: “The next questions ask about the new health care law, known as the Affordable Care Act or ‘Obamacare.’ In general, is your opinion of the law very favorable, somewhat favorable, neither favorable nor unfavorable, somewhat unfavorable, or very unfavorable?” Our analysis groups respondents into three groups: (1) favorable (those responding “somewhat” or “very favorable”), (2) no opinion (those responding“neither favorable nor unfavorable”), and (3) unfavorable (those responding “somewhat” or “very unfavorable”).

 

We analyze opinion differences by self-reported characteristics (including health status, family income, current insurance status, race/ethnicity, education, age, and urban/rural residence), paying particular attention to family income, insurance status, and race/ethnicity. We use three family income groups, based on ACA-related categories: family income less than or equal to 138 percent of FPL (the lower-income group, eligible for Medicaid in expansion states); family income between 139 percent and 399 percent of FPL (the middle-income group, eligible for Marketplace-based subsidies unless affordable employer-sponsored insurance is available); and family income at or above 400 percent of FPL (the higher-income group, above the Marketplace subsidy income eligibility cutoff). We use three insurance statuses at the time of the survey: private insurance, Medicaid and other public coverage, and uninsured. We use three race/ethnicity groups: white (defined as white, non-Hispanic); nonwhite (defined as ‘other non-Hispanics,’ predominantly African Americans, but also Asians and Native Americans); and Hispanic. We also conducted regression analyses to assess how much the patterns change when we control for other factors.2 Unless otherwise noted, the variation in opinions about the ACA across the socioeconomic and demographic characteristics reported in the brief hold up when we control for observed differences in age, general health status as reported by the respondent, income, insurance status, education, race/ethnicity, and residential location.

 

We then examine opinions of the law for both those living in states adopting the Medicaid expansion compared with those living in states not expanding, and those living in states that are adopting their own Marketplaces rather than using the federally facilitated Marketplace. We also examine opinions of the law separately for those who indicated that they or a family member had been affected by one of the early ACA provisions compared with those who said they had not been affected or did not know. Benefit provisions implemented between 2010 and 2013 include the prohibition of coverage denial for children because of pre-existing conditions, free preventive care services, free birth control prescriptions, children being permitted to stay on parents’ private insurance until age 26, the end of annual and lifetime dollar limits on benefits, the requirement of easy-to-understand summaries of health plans’ benefits and coverage rules, and the requirement that insurance companies reimburse customers if at least 80 percent of the insurance premiums collected was not spent on health care services. We examine the responses of uninsured adults in additional detail, given that the main purpose of the ACA is to reduce the number of Americans without health insurance coverage.

 

What We Found

 

Americans most likely to have an unfavorable view of the ACA are in the middle- and higher-income groups, have private insurance, are in very good/excellent health, are white, and live in rural areas. Those most likely to have no opinion include the groups most likely to benefit from the law—those in fair/poor health, those with lower incomes, the uninsured, nonwhites and Hispanics, high school graduates (or less), and the young.

 

Overall, 40.5 percent of respondents had an unfavorable view of the law, compared with 28.3 percent with a favorable view and 30.4 percent with no opinion. The share of adults with an unfavorable view of the law found on the HRMS as of September 2013 is very similar to the share of adults reporting an unfavorable view of the law on the Health Tracking Survey that was fielded at the same time (40.5 percent versus 43 percent).3

 

Family Income

Respondents in the higher-income group had a more unfavorable view of the ACA compared with their counterparts in the lower-income group (44.4 percent versus 31.9 percent, respectively [table 1 and figure 1]). Middle-income respondents were also more likely to have an unfavorable opinion of the law than lower-income respondents. Lower-income respondents were about equally split between having a favorable or an unfavorable opinion (29.3 percent versus 31.9 percent); but they were also the most likely to have no opinion (37.4 percent versus 23.5 percent of the higher-income group). However, none of these differences in opinion across income group are significant when we control for other factors, such as insurance status and education, suggesting that other factors contribute to the observed differences in support for the ACA.4

 

 

Insurance Status Pre-reform

Those with private insurance were far more likely to have an unfavorable opinion of the law than those on Medicaid or the uninsured (44.8 percent versus 29.6 percent and 31.3 percent, respectively [table 1 and figure 1]). Those on Medicaid and the uninsured were far more likely than the privately insured to have no opinion (38.8 percent and 37.8 percent, respectively, versus 27.0 percent).5 Only about 3 out of 10 respondents had a favorable view of the law, regardless of insurance status.

 

Race/Ethnicity

Whites were far more likely to have an unfavorable than a favorable opinion of the law, at 51.3 percent versus 23.0 percent (table 1). They were also far more likely to have an unfavorable opinion than nonwhites and Hispanics (20.1 percent and 24.0 percent, respectively). Nonwhites were more likely to have a favorable than an unfavorable opinion (39.7 percent versus 20.1 percent), as were Hispanics (34.2 percent versus 24.0 percent);6 both groups were also considerably more likely than whites to have no opinion.

 

Education

High school graduates, those with some college, and college graduates were more likely to have an unfavorable than a favorable opinion (table 1). The percentage with a favorable opinion increases with education. But only those with post graduate degrees were more likely to have a favorable than an unfavorable opinion (44.5 percent versus 36.9 percent). Those with the least education were the most likely to have no opinion (36.9 percent versus 18.2 percent for those with postgraduate degrees).

 

Self-reported Health Status

Those in very good or excellent health were more likely than those in fair or poor health to have an unfavorable opinion of the law (43.8 percent versus 35.1 percent [table 1]). Those in fair or poor health were also more likely than other health groups to have no opinion (36.3 percent compared with 27.0 percent for those in very good or excellent health).

 

Age

Older nonelderly respondents (age 50–64) were more likely than young adults (age 18–30) to have an unfavorable opinion of the law (43.9 percent versus 36.6 percent [table 1]). Young adults were more likely than older adults to have no opinion (35.3 percent versus 25.4 percent).

 

Table 1. Opinion of the Affordable Care Act among Adults Age 18—64, by Health, Demographic, and Socioeconomic Characteristics

 

Sample Size

Opinion of the ACA

Favorable opinion (%)

No opinion (%)

Unfavorable opinion (%)

Total

7911

28.3

30.4

40.5

***

Self-reported health status

Very good or excellent

4135

28.4

27.0

+++

43.8

***

+++

Good

2737

28.4

33.4

***

37.2

***

Fair or poor^

1014

27.3

36.3

***

35.1

*

Family income

Less than or equal to 138% of FPL^

1691

29.3

37.4

***

31.9

139–399% of FPL

3293

23.9

+++

31.8

***

+++

43.3

***

+++

400% of FPL or above

2927

31.7

23.5

***

+++

44.4

***

+++

Current insurance status

Private insurance

5887

27.5

27.0

+++

44.8

***

+++

Medicaid or other public coverage

893

31.0

38.8

*

29.6

Uninsured^

1131

29.2

37.8

***

31.3

Race/ethnicity

White^

5746

23.0

24.8

51.3

***

Nonwhite, non-Hispanic

1279

39.7

+++

39.3

+++

20.1

***

+++

Hispanic

886

34.2

+++

40.5

**

+++

24.0

***

+++

Education

High school graduate or less^

2329

24.0

36.9

***

37.8

***

Some college or associate degree

2441

24.4

30.0

***

+++

44.9

***

+++

College graduate

1894

33.4

+++

24.6

***

+++

41.6

***

Master's degree or PhD

1247

44.5

+++

18.2

***

+++

36.9

*

Age

18–30^

1538

27.1

35.3

***

36.6

**

31–49

2839

27.8

30.9

**

++

40.5

***

++

50–64

3534

29.8

25.4

**

+++

43.9

***

+++

Urban/rural

Lives in an MSA^

6838

29.6

31.4

38.1

**

Does not live in an MSA

1073

20.4

+++

24.3

+++

54.4

***

+++

Source: Health Reform Monitoring Survey, quarter 3 2013.
Notes: Data do not include those who did not answer the questions (less than 1%).? FPL is federal poverty level.? MSA is metropolitan statistical area. The percentages for the three categories of opinion of the ACA do not add to 100% because a small number of respondents (0.9% for the entire sample) did not report an opinion.?
*/**/*** Estimate differs significantly from those with a favorable opinion at the 0.1/0.05/0.01 levels, using two-tailed tests.
++/+++ Estimate differs significantly from the reference group, marked by a ^, at the 0.05/0.01 levels, using two-tailed tests. No estimates differed significantly at the 0.1 (+) level.

 

Urban versus Rural

The majority of rural Americans had an unfavorable view of the law (54.4 percent compared with only 38.1 percent for those living in a metropolitan statistical area [table 1]).

 

Those living in states that have adopted the Medicaid expansion or a state-based Marketplace were less likely to have an unfavorable opinion of the law than those living in other states.

 

 

Those living in states that are expanding Medicaid were less likely to have an unfavorable opinion of the law than those in other states (36.6 percent versus 45.2 percent [figure 2]). Respondents living in states either not expanding Medicaid or still deliberating were more likely to have an unfavorable than favorable opinion (45.2 percent versus 26.6 percent). In states that have expanded Medicaid, there were no differences between the shares of adults with favorable and unfavorable opinions.

 

Those living in states that implemented a state-based Marketplace were less likely to have an unfavorable opinion of the ACA than those living in states with a federally facilitated Marketplace (34.1 percent versus 44.1 percent). Those living in states with a federally facilitated Marketplace were significantly more likely to have an unfavorable than favorable opinion (44.1 percent versus 26.6 percent); there were no significant differences between favorable and unfavorable opinions of the law among those living in states with a state-based Marketplace. This pattern raises the question of which came first: Did opinions of state residents prompt policy decisions by state lawmakers or did state policy decisions influence residents’ opinions?

 

Those who said they were affected by at least one early benefit provision of the law had a more favorable opinion of the ACA compared to those who had not been affected or were unsure. But even among that group, the share with an unfavorable opinion exceeded that with a favorable opinion.

 

 

Of those who said that they or a family member had been affected by the early provisions, 34.0 percent had a favorable opinion of the law compared with 24.7 percent for those who said they had not been affected or were unsure (figure 3). The unaffected group was also more likely to have no opinion of the ACA. The two groups were similar, however, in the proportion having an unfavorable opinion (just over 40 percent). Notably, even the individuals who had been affected by the early ACA provisions were more likely to have an unfavorable than a favorable opinion of the ACA as a whole. This implies either (1) that individuals can view specific provisions very positively but still have an unfavorable view of the overall law, or (2) that they do not associate the provision that had affected them or their family with the ACA.

 

Uninsured adults, particularly whites, those living in rural areas, and those with a high school education or less were more likely to have an unfavorable than a favorable opinion of the law. In addition, the uninsured most likely to benefit from the ACA (e.g., lower- and middle-income adults, those in fair or poor health, and young adults ages 18 to 30) expressed only weak support for it and were as or more likely to have no opinion than to have a favorable view.

 

Uninsured adults’ opinions are particularly interesting because a major goal of the ACA is to reduce the number of uninsured. Surprisingly, the uninsured were as likely to have a favorable as to have an unfavorable opinion of the ACA (both about 30 percent) and an even larger share of the uninsured (37.8 percent) said they had no opinion (table 2). Moreover, the picture is not very different for subgroups of the uninsured that are particularly targeted by the coverage provisions of the ACA such as those who could be eligible for Medicaid or Marketplace subsidies, those in fair or poor health, and those who are between the ages of 18 and 30—in all of those groups, the share with a favorable view is just about 30 percent, about the same as the share having an unfavorable view.

 

Uninsured, lower-income adults—who will benefit if their states choose to expand Medicaid—were equally split between favorable and unfavorable opinions, with almost 40 percent having no opinion. Similarly, among middle-income adults who also stand to benefit from the law (because of the availability of income-related subsidies), 36.7 percent had an unfavorable opinion and 36.2 percent had no opinion.

 

There are noticeable differences by race/ethnicity, residential location, and educational attainment. The uninsured who are nonwhite or Hispanic are more likely to have no opinion or to have a favorable opinion of the ACA and less likely to have an unfavorable opinion relative to the uninsured who are white. Similarly, the uninsured who live in a metropolitan statistical area are more likely to have no opinion or to have a favorable opinion of the ACA and less likely to have an unfavorable opinion relative to the uninsured who live outside of a metropolitan statistical area. However, the differences found by residential location do not hold up when we control for other observed factors. Finally, the uninsured with a college degree (or more) were more likely to have a favorable opinion and less likely to have no opinion or to have an unfavorable opinion relative to the uninsured who had only a high school degree or less.

 

Table 2. Opinion of the Affordable Care Act among Uninsured Adults Age 18—64, by Health, Demographic, and Socioeconomic Characteristics

 

Sample Size

Opinion of the ACA

Favorable opinion (%)

No opinion (%)

Unfavorable opinion (%)

Total

1,131

29.2

37.8

***

31.3

 

Self-reported health status

Very good or excellent

448

29.0

35.1

*

++

34.7

 

+

Good

458

30.5

37.2

 

29.8

 

Fair or poor^

219

27.4

43.8

***

27.7

 

Family income

Less than or equal to 138% of FPL^

607

30.2

39.7

***

28.3

 

139–399% of FPL

463

25.9

36.2

***

36.7

**

++

400% of FPL or above†

--

--

--

--

Race/ethnicity

White^

615

20.3

30.4

***

47.8

***

Nonwhite, non-Hispanic

234

38.6

+++

40.5

 

++

18.5

***

+++

Hispanic

282

34.4

+++

45.5

*

+++

18.5

***

+++

Education

High school graduate or less^

562

27.7

40.3

***

30.4

 

Some college or associate degree

374

28.4

34.7

*

+

35.6

 

College degree or higher

195

41.3

++

31.7

 

++

24.1

**

Age

18–30^

316

32.1

36.7

 

29.2

 

31–49

426

27.1

41.3

***

30.8

 

50–64

389

28.4

33.5

 

35.3

*

+

Urban/rural

Lives in an MSA^

954

30.1

39.2

***

28.8

 

Does not live in an MSA

177

24.9

+

31.5

 

+

42.7

***

+++

Source: Health Reform Monitoring Survey, quarter 3 2013.
Notes: Data do not include those who did not answer the questions (less than 2% of the uninsured sample). FPL is federal poverty level. MSA is metropolitan statistical area. The percentages for the three categories of opinion of the ACA do not add to 100% because a small number of respondents (1.7% for the uninsured sample) did not report an opinion. 
*** Estimate differs significantly from those with a favorable opinion at the 0.01 level, using a two-tailed test. No estimates differ significantly at the 0.1/0.05 (*/**) levels.
+/++/+++ Estimate differs significantly from the reference group, marked by a ^, at the 0.1/0.05/0.01 levels, using two-tailed tests.
† Data are not presented for respondents with incomes at or above 400% of FPL because the cell size was too small to provide reliable estimates.

 

What It Means

 

The fact that so many of the groups targeted by the ACA coverage provisions have no opinion about the ACA and that they are equally likely to have an unfavorable opinion as to have a favorable opinion suggests that public education and outreach efforts are falling well short of reaching and informing the ACA target populations who stand to benefit from the coverage provisions. Large percentages of those in fair or poor health, those with lower incomes, the uninsured, racial/ethnic minorities, and the young have no opinions one way or the other about the law. This is consistent with an earlier Urban Institute brief showing limited knowledge of several of the law’s coverage provisions, particularly among lower-income and uninsured Americans (Long and Goin 2014). Some groups who could benefit significantly from the law, such as middle-income Americans, even have strongly negative views of it. Similarly, many uninsured with incomes less than or equal to 138 percent of FPL, a key target group of the law, have no opinion.


Racial and ethnic differences in opinions of the law are among the largest of those we examined in this analysis. Understanding the opposition to the ACA among the groups that are targeted by the coverage provisions of the ACA, and particularly among groups where opposition is stronger, seems key to understanding the overall unfavorable opinions of the ACA revealed in national polls. These results also point to the potential for large shifts in opinions over time, given that 30 percent of the adults indicated that they had neither a favorable nor an unfavorable opinion about the ACA. It is clear that support for the ACA is relatively weak among all groups that we examined—even among the low- and moderate- income uninsured adults and the uninsured who are in poor or fair health—who stand to benefit directly from the provisions of the ACA that are being implemented now. Understanding and addressing the lack of support for the ACA among these groups will be essential to increasing ACA engagement and achieving the increases in coverage sought under the law.

 

References

 

Long, Sharon K., and Dana Goin. 2014. “Most Adults Are Not Aware of Health Reform’s Coverage Provisions.” Washington, DC: Urban Institutel.

 

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

 

John Holahan is an Institute fellow and Genevieve M. Kenney is codirector and senior fellow in the Urban Institute’s Health Policy Center. Michael Karpman and Ariel Fogel are research associates in the Health Policy Center.  The authors appreciate the advice and suggestions of Stephen Zuckerman, Katherine Hempstead and Linda Blumberg and the input of Douglas Wissoker on statistical issues.

 

Note


1Organizations that have been tracking public opinion of the Affordable Care Act include Kaiser, Gallup, and Rand. See “Health Tracking Poll: Exploring the Public’s Views on the Affordable Care Act (ACA),” Kaiser Family Foundation, accessed February 24, 2014; Frank Newport, “Americans Slightly Less Negative about Healthcare Law,” Gallup Politics, December 13, 2013; “RAND Health Reform Opinion Study.” RAND Corporation, accessed February 24, 2014l.
2The regressions include self-reported health status, family income, current insurance status, race/ethnicity, education, age, and an indicator of residence in a metropolitan area.
3The HRMS finds a lower share with a favorable opinion and a higher share with no opinion relative to the Health Tracking Survey. Given the subjective nature of the question, it is likely that the survey mode may affect the pattern of responses observed across different surveys.
4The other categories reported in table 1 remain at least partially significant after regression adjustment, but the differences are reduced in magnitude.
5The difference between the publicly and privately insured in the share with no opinion of the ACA was not statistically significant at conventional levels (p < .10) when we control for other factors.
6This is consistent with findings from Kaiser Family Foundation, “Health Tracking Poll.”

Ford Foundation Urban Institute Robert Wood Johnson Foundation