Jul 11, 2017 | Karen Pollitz and Anthony Damico

As the Senate considers the Better Care Reconciliation Act (BCRA), a proposal to repeal and replace the Affordable Care Act (ACA), amendments have been discussed to further change private health insurance market rules that apply under current law.  Under the BCRA, current law health insurance market rules would still apply: Insurers in the non-group health insurance market are prohibited from turning applicants down or charging higher premiums based on health status and from excluding coverage for pre-existing conditions.  In addition, all policies must provide major medical coverage for 10 categories of essential health benefits and must limit the annual out-of-pocket cost sharing (deductibles, co-pays and coinsurance) that people must pay for covered services in network (although states can alter those requirements through waivers).

However, an amendment to the BCRA, suggested by Senator Ted Cruz (R-TX), reportedly would allow insurers in the non-group market to also sell some policies that would not be required to follow all of the ACA market rules.  For example, such policies might not have to follow ACA essential health benefit and cost sharing standards.  In addition, some reports suggest that insurers would not have to sell these policies to people with health conditions or risks and could vary premiums for them based on the health of applicants.

This brief examines the likely impact of such a change on the stability of coverage offered through non-group markets and on the number of individuals who might be affected.

Impact on Consumers Ineligible for Premium Subsidies

If the BCRA were amended to permit insurers to sell ACA-compliant plans alongside plans that did not follow ACA-benefit standards and/or rating and access rules, the likely result would be that the cost of ACA-compliant plans would skyrocket.  The ACA-compliant plans would effectively become a high-risk pool, attracting enrollees when they need costly health benefits – such as maternity care, or drugs to treat cancer or HIV, or therapies to treat mental health and substance abuse disorders – and those with pre-existing conditions who are turned down by non-compliant plans or charged high premiums based on their health.  By contrast, non-compliant plans would attract healthier consumers, at least as long as they didn’t need coverage for such benefits.  Premiums from the healthier enrollees would not be pooled to help keep the price of compliant plans affordable. As a result, premiums for compliant plans would increase significantly, while premiums for non-compliant plans would be substantially lower (though they would also cover fewer benefits).

The Senate BCRA would continue ACA-like premium tax credits to subsidize the cost of coverage for low-and middle-income individuals. Like the ACA, premium tax credits under the BCRA would be tied to the cost of a benchmark marketplace plan, though the benchmark would have higher patient cost-sharing than under the ACA. Eligibility for premium tax credits would be capped at income of 350% of the federal poverty level (FPL), compared to 400% FPL under current law.  Individuals eligible for tax credits would be required to pay a set percentage of their annual income toward a benchmark plan; the premium tax credit amount for each individual would be the difference between the actual cost of the benchmark plan and a person’s required contribution.  Under this formula, as under current law, subsidy-eligible people would generally be shielded from annual premium cost increases, which would instead be absorbed by federal premium tax credits.

While people with pre-existing conditions eligible for premium tax credits would be cushioned from premium increases in compliant plans under the Cruz amendment, those ineligible for credits would not be protected.

According to the National Health Interview Survey, approximately 40% of non-group market participants in 2015, or 6.1 million people, had income above 350% FPL.  Most of these individuals purchased non-group coverage outside of the marketplace. Under the BCRA, these individuals would not be eligible for premium tax credits.

We further estimate that, among the 6.1 million non-group market participants with incomes of at least 350% FPL, 24% (or about 1.5 million) would have pre-existing conditions that would have been considered automatically deniable by insurers prior to the ACA.  These conditions include cancer, diabetes, HIV/AIDS, hepatitis, substance use disorders (including opioid addiction), serious mental illnesses, and pregnancy. (Figure 1)

For these 1.5 million individuals, non-compliant plans would likely either deny coverage outright or charge very high premiums tied to their health. Even if they could obtain coverage in non-compliant plans, it might not cover key benefits, such as maternity care, mental health care, substance use treatment, or prescription drugs, and would not solve the affordability problem.

Among the three-quarters of other market participants with income over 350% FPL, millions would have other types of pre-existing conditions that were not considered automatically declinable prior to ACA, such as high-blood pressure, high cholesterol, asthma, and depression.  Many in this group also could see a substantial increase in out-of-pocket spending for medical care that would offset any premium savings associated with less comprehensive policies.

Methods

To calculate nationwide prevalence rates of declinable health conditions, we reviewed the survey responses of nonelderly adults for all question items shown in Methods Table 1 using the CDC’s 2015 National Health Interview Survey (NHIS).  Approximately 27% of 18-64 year olds, or 52 million nonelderly adults, reported having at least one of these declinable conditions in response to the 2015 survey.  For more details on methods and a list of declinable conditions included in this analysis, see our earlier brief: Pre-existing Conditions and Medical Underwriting in the Individual Insurance Market Prior to the ACA.

The programming code, written using the statistical computing package R, is available upon request for people interested in replicating this approach for their own analysis.