Issue: Prior to the Affordable Care Act (ACA), people with preexisting health conditions could be denied insurance coverage or charged higher rates. If the law is repealed, these protections could be diluted or lost altogether.
Goals: Assess the ACA’s impact on coverage and access for people with preexisting conditions and compare their coverage gains with state high-risk-pool enrollment pre-ACA.
Methods: Analysis of Behavioral Risk Factor Surveillance System data for the period 2011–13 to 2015.
Key Findings and Conclusions: Between 2013 and 2015, 16.5 million nonelderly adults gained coverage following full ACA implementation. Of those, 2.6 million had preexisting conditions that could have otherwise precluded them from coverage because of discriminatory denials and pricing; 9.4 million had conditions that could have otherwise affected insurance cost. We found strong correlations between these coverage gains and access to care. Coverage and access gains for people with preexisting conditions were unrelated to the size or existence of the state high-risk pools that 35 states funded for such individuals pre-ACA. Our findings suggest that proposals to replace current protections for people with preexisting conditions with high-risk pools are unlikely to be sufficient to maintain the ACA’s gains.
Americans with chronic health conditions are at the center of the debate over access to health care coverage. The U.S. Department of Health and Human Services’ Office of the Assistant Secretary for Planning and Evaluation (ASPE) estimates the number of Americans with such “preexisting conditions” ranges from 19 percent to 50 percent of all nonelderly Americans.1,2 This range represents the difference between conditions that fit into a “narrow” definition of preexisting conditions (19%), and a “broad” definition (50%). The narrow definition includes very costly health conditions that would cause insurers to refuse coverage absent the Affordable Care Act’s (ACA) provisions; the broad definition includes slightly less expensive chronic health conditions that could nevertheless make the cost of insurance in the individual market without the ACA largely unaffordable for most patients.
In 2016, the Henry J. Kaiser Family Foundation, in its review of pre-ACA medical underwriting practices, estimated that 27 percent of nonelderly American adults had health conditions that “would likely leave them uninsurable if they applied for individual market coverage.”3 Similarly, a Commonwealth Fund study found that, in 2010, 36 percent of adults ages 19 to 64 who had tried to buy a plan in the individual market over the prior three years were turned down, charged a higher price, or had a condition excluded from their coverage because of a health problem.4
The presence of preexisting conditions is particularly important for the millions of Americans who have gained coverage under the ACA, which Congress and the Trump administration are seeking to repeal.5 The Commonwealth Fund study found significant improvements in the ability of people with health problems to purchase plans on their own in 2016 relative to 2010.6
In this issue brief, we observe whether the coverage gains for people with preexisting conditions also have resulted in better access to care. Better access is defined as a greater likelihood of having a regular health care provider (whether one or more than one clinician) and having less trouble seeing a provider because of the cost.
Prior to passage of the ACA, many states had high-risk pools that sought to provide coverage to individuals locked out of the individual insurance market because of expensive preexisting conditions. Between 2010 and 2013, the ACA funded the Pre-Existing Condition Insurance Program, a set of federally funded high-risk pools to provide interim coverage for those with such conditions. If these pools had been successful in addressing coverage for those with preexisting conditions, we would expect to see a smaller gain in access to care for this population in those states that had previously enrolled substantial shares of the nongroup market in the pools.
Under the ACA, Americans with Preexisting Conditions Gained Coverage and Better Access to Care
For this brief, we considered both the narrow and broad definitions of preexisting conditions. Among the general population surveyed between 2011 and 2015, data from the Behavioral Risk Factor Surveillance System (BRFSS) indicate that 20 percent of Americans have preexisting conditions under the narrow definition and 61 percent of Americans have these conditions under the broader definition. Using the BRFSS data, we estimate that 16.5 million more people were insured in 2015 than in the 2011–2013 period.7 Among this newly insured group, 2.6 million had one or more preexisting conditions under the narrow definition and 9.4 million had one or more under the broader definition (Exhibit 1).
Under the narrow and broad definitions, those with preexisting conditions accounted for 16 percent and 57 percent, respectively, of the newly insured population. These findings suggest that the newly insured population is neither substantially healthier nor sicker than the general population.
Among the 2.6 million newly insured people under the narrow definition of preexisting conditions were an estimated 364,000 people ever diagnosed with heart attack or heart disease; 536,000 people ever diagnosed with chronic obstructive pulmonary disease (COPD), emphysema, or chronic bronchitis; 770,000 people ever diagnosed with cancer; and 1.4 million people ever diagnosed with diabetes. Among the 9.4 million newly insured under the broad definition were 1.5 million people ever diagnosed with asthma, 3.3 million people ever diagnosed with high blood pressure, and 4.1 million ever diagnosed with depression.
The prevalence of preexisting conditions rises with age. Among adults ages 55 to 64 who gained coverage (in both expansion and nonexpansion states) between 2011 and 2015, nearly 40 percent had a preexisting condition under the narrow definition, and nearly 80 percent had a preexisting condition under the broad definition. The share of those gaining coverage who had a preexisting condition was somewhat higher among white non-Hispanics (65%) than among black non-Hispanics (59%) or Hispanics (54%).
We also found that, among populations with preexisting conditions, these increases in insurance coverage were associated with increased access to care. The share of people reporting cost-related access problems in the past 12 months fell in all the preexisting-condition groups (Exhibit 2). Likewise, the share reporting they had a usual source of care (one or more clinicians they consider to be their personal doctor or health care professional) increased (Exhibit 3).
Improvements in access to care over time might have been a result of other changes happening contemporaneously. To determine that these improvements in access were a consequence of the ACA’s coverage expansions, we compared gains in access among those with preexisting conditions to gains in coverage across states. In Exhibits 4 and 5 respectively, we plot affordability and access to a usual source of care against the share of people with preexisting conditions among the newly insured. Improvements in access among those with preexisting conditions were greatest in states where coverage gains were greatest.
In Republican proposals to replace the ACA’s insurance regulations, high-risk pools are suggested as the mechanism to ensure coverage for those with preexisiting conditions who would otherwise be locked out of the individual insurance market. High-risk pools had existed in 35 states before passage of the ACA, and the ACA included $5 billion in funding for the high-risk pools that operated between 2010 and 2013.8
In this study, we hypothesized that if high-risk pools had been effective in covering people with serious preexisting conditions and improving their access to care, we would expect to find that the populations with preexisting conditions in those states would have already been insured and had access to care prior to the ACA’s full rollout in 2014. To test this proposition, we examined the relationship between the increase in insurance coverage and access to care among those with serious preexisting conditions and prior enrollment in the PCIP and high-risk-pool programs.
We found no relationship between either enrollment in the PCIP or the share of the nongroup market enrolled in high-risk pools and gains in coverage or access post-2014. If anything, people with preexisting conditions living in states with a higher enrollment in the PCIP (or in high-risk pools) saw slightly greater gains in access to care under the full rollout of the ACA (Appendix). That is an indication that these individuals may not have been covered when the PCIP was in operation.
A significant portion of Americans—up to 61 percent—could have preexisting health conditions that affect the price of their health insurance or their ability to get any coverage at all. Should the Affordable Care Act be repealed and its protections for coverage of preexisting conditions removed, millions of Americans could find it difficult to obtain affordable health care. Gains in coverage under the ACA have led to corresponding gains in access to care, as measured both by the reduction in cost-related problems getting care and by the increased proportion of people who report having a usual source of care. By contrast, the high-risk pools in place prior to the ACA’s implementation did not have comparable effects on coverage or access to care. Our results suggest that proposals to use states’ high-risk pools in place of ACA protections will be insufficient to maintain the health care access gains made since 2010.
The group of people with preexisting conditions changes over time as its members’ health changes. The group seeking coverage in the individual insurance market also changes: people gain and lose employment and access to employer-sponsored insurance, experience fluctuations in income and eligibility for Medicaid, age in or out of public coverage, and undergo other life changes. Alternatives to the ACA’s prohibitions on underwriting will need to address the needs of these ever-changing populations.
How This Study Was Conducted
The analyses in our study are based on the Behavioral Risk Factor Surveillance System (BRFSS). This survey asks participants simple, point-in-time questions about their health. The questions used for this brief ask participants whether they have “ever [been] told” that they have the chronic health conditions we include in our estimates.
We use BRFSS data from 2011, 2013, and 2015 because the BRFSS has included only the questions on the chronic health conditions we consider in those years. These data include those who were “ever diagnosed” with diabetes, stroke, cancer, COPD, emphysema, chronic bronchitis, angina, kidney disease, heart attack, or heart disease. This lower bound comes to 20 percent of Americans ages 19 to 64. The BRFSS does not include questions for as broad a range of chronic conditions as other data sets, including some conditions that companies have listed as preexisting conditions, so these estimates may be lower than the actual population with preexisting conditions that can lead to discriminatory pricing.
The set of preexisting conditions our study defines as narrow coincides with the conditions cited in a study by the U.S. Department of Health and Human Services’ Office of the Assistant Secretary for Planning and Evaluation (ASPE) as conditions that would categorically lead to a denial by a private insurer. It also accords with conditions included in states’ high-risk pools.
The set of preexisting conditions that our study defines as broad includes conditions that could cause an individual to be charged more for health coverage. In the BRFSS, these conditions were arthritis, asthma, high blood pressure, high cholesterol, obesity, and depression. We also included those who described their general health as “poor.” Under this definition, 61 percent of the population could have a preexisting condition.
Our enrollment rate for the Pre-Existing Condition Insurance Plan was calculated by using enrollment numbers reported by the Center for Consumer Information and Insurance Oversight as of March 31, 2013. We divided those state enrollment numbers by the total state nonelderly adult population in 2013. The 2011 nongroup market enrollment rate in state-run high-risk pool programs used in Appendix Exhibit C is from the Henry J. Kaiser Family Foundation.9
1 Office of the Assistant Secretary for Planning and Evaluation, At Risk: Pre-Existing Conditions Could Affect 1 in 2 Americans: 129 Million People Could Be Denied Affordable Coverage Without Health Reform (U.S. Department of Health and Human Services, Nov. 2011).
2 Office of the Assistant Secretary for Planning and Evaluation, Health Insurance Coverage for Americans with Pre-Existing Conditions: The Impact of the Affordable Care Act (U.S. Department of Health and Human Services, Jan. 2017).
3 G. Claxton, C. Cox, A. Damico et al., Pre-existing Conditions and Medical Underwriting in the Individual Insurance Market Prior to the ACA (Henry J. Kaiser Family Foundation, Dec. 2016).
4 S. R. Collins, M. Z. Gunja, M. M. Doty, and S. Beutel, How the Affordable Care Act Has Improved Americans’ Ability to Buy Health Insurance on Their Own (The Commonwealth Fund, Feb. 2017).
5 R. Abelson, “Donald Trump Says He May Keep Parts of Obama Health Care Act,” New York Times, Nov. 11, 2016.
6 S. R. Collins, M. Z. Gunja, M. M. Doty, and S. Beutel, How the Affordable Care Act Has Improved Americans’ Ability to Buy Health Insurance on Their Own (The Commonwealth Fund, Feb. 2017).
7 Based on our tabulations of data from the Centers for Disease Control and Prevention’s Behavioral Risk Factor Surveillance System (BRFSS).
8 Center for Consumer Information and Insurance Oversight data as of March 31, 2013.
9 K. Pollitz, High-Risk Pools for Uninsurable Individuals (Henry J. Kaiser Family Foundation, Aug. 2016, updated Feb. 2017).