When he was in the Obama administration, Richard Frank helped to write regulations for the legislation that required employer-based health plans to cover mental health and addiction services on parity with medical and preventive health. Eight years later, Frank is pretty pleased with how it all turned out, but he also acknowledges the numerous barriers that still impede the cause of getting behavioral health care to people who really need it.
“It’s had a really big effect,” Frank says of the Mental Health Parity and Addiction Equity Act (MHPAEA) of 2008. Now a professor at Harvard Medical School, Frank was assistant secretary of planning and evaluation at the Department of Health and Human Services under Obama. The first part of MHPAEA went into effect in 2010, just in time to inspire similar provisions in the ACA for the individual and small-group market as well as Medicaid managed care plans.
Some numbers, says Frank: “Coverage for 174 million people has been affected by a combination of MHPAEA and ACA together.” That includes people who had coverage for behavioral health but got better coverage and 20 million who gained coverage under the ACA. “That’s a big deal,” he says.
While Republicans have targeted the ACA individual mandate, they have left mental health parity untouched. Says Andrew Sperling, director of legislative affairs for the National Alliance on Mental Illness (NAMI), “There’s nothing in the pipeline revoking parity at all.”
But achieving true parity for coverage of mental health and substance use disorders has been a slog. America’s Health Insurance Plans (AHIP) has acknowledged that parity is lagging and confusion surrounds key elements of the law. Massachusetts Democratic Rep. Joe Kennedy III cited uneven compliance when he introduced legislation in 2015 to give HHS and other federal agencies audit powers, but the bill went nowhere. Kennedy’s press office says he hopes to reintroduce an updated version soon. His cousin, former Rhode Island congressman Patrick Kennedy, has become an outspoken advocate for mental health and substance abuse causes (see page 23).
Two silos of regulation
Regulation of behavioral health parity occurs in two silos:
- MHPAEA for employer-based plans. This law does not require employer plans to cover mental health and substance abuse services, but if they do and they have more than 50 employees, they must cover behavioral health benefits on par with medical and surgical services. That means the same levels of cost-sharing and similar benefit limits. The Department of Labor handles enforcement under the Employee Retirement and Income Security Act (ERISA).
- ACA for the marketplace plans and Medicaid managed care. The ACA includes treatment for mental health and substance abuse as one of the 10 essential health benefits. Enforcement comes under the Public Health Service Act, which means state insurance commissioners are in charge with assistance from the CMS. Six states—Alabama, Alaska, Florida, Georgia, Pennsylvania, and Wisconsin—have pushed enforcement responsibilities entirely on to CMS because they say they can’t do that job.
Coverage doesn’t mean care
The two laws have extended coverage for behavioral health care to millions of people who didn’t have it before. At the same time, the evidence that more people are proportionally getting the care they need is mixed. The reasons vary. Access to providers and treatment facilities varies across different regions. Insurers say they have trouble recruiting providers to be part of their networks. Critics say the problem is that insurers don’t pay providers enough. The stigma of mental health disorders also discourages people from seeking care, but that’s beyond what legislators and regulators can fix.
“There remain incentives [for health plans] to try to avoid folks with mental and addictive illnesses,” says Richard Frank of Harvard Medical School.
Estimates of the number of people who have better behavioral health benefits or are newly covered are all over the map. Frank says his estimate of 174 million is a cumulative number based on studies he has reviewed along with his own analysis of National Survey on Drug Use and Health data.
Other estimates have been more modest. In a 2016 report, for example, AHIP estimated that 38 million people had access to behavioral health benefits in 2014 based on case studies of 11 AHIP members, but the report is vague on what exactly access means.
Office visit payments vs. Medicare-allowed amounts*
Source: Milliman Research Report, “Addiction and Mental Health vs. Physical Health: Analyzing Disparities in Network Use and Provider Reimbursement Rates,” November 2017
Much of the published data on the impact the ACA has had on behavioral health coverage hasn’t been updated since 2014, the first year of ACA implementation. Timothy Creedon and Benjamin LeCook published a study in Health Affairs in 2016 that found that 81.5% of 2014 respondents with serious psychological distress or substance use disorders had insurance, which they described as “a significant increase over all pre-2014 time periods.” Drawing on the national survey data, they also found that blacks and Hispanics continued to have much lower rates of coverage. Creedon says they are working to update their data but haven’t published them yet.
The MHPAEA–ACA double barrel may have moved the needle on the coverage meter quite a bit, but the gains in actually getting care have been less impressive.
A study in the January 2017 issue of Psychiatric Services reported that with the implementation of the ACA in 2014, uninsured rates declined 5.4 percentage points for people with mental disorders and 5.1 percentage points for those with substance use disorders, thanks mostly to Medicaid expansion. Gains were largest for adults with incomes up to 200% of the federal poverty level compared with the overall population. Interestingly, though, use of mental health treatment increased by just 2.1 percentage points and the utilization of treatment for substance use disorders remained stagnant. And the number of treatment settings did not change significantly. However, Medicaid payment for substance use disorder treatment increased sizably, going up by 7.6 percentage points.
One indication of the inadequacy of insurance coverage, even with parity, is the number of people who get behavioral health care out of network. People are far more likely to go out of network for behavioral health services than for medical or surgical services, according to a study Milliman just completed for the Mental Health Treatment and Research Institute, a not-for-profit subsidiary of the little-known Bowman Family Foundation in Delaware. Milliman found that between 2013 and 2015, the proportion of out-of-network care for inpatient behavioral health services was almost three to four times higher than medical or surgical services, and out-of-network outpatient behavioral health care was three to almost six times higher.
NAMI late last year released results of an online survey of 3,177 people with behavioral health conditions and their caregivers, and found even worse disparities: Almost 35% said they had trouble finding any mental health therapists who would take their insurance, and 28% said they had to go out of network for such services compared with just 7% for medical specialists and 3% for primary care providers.
This report, says NAMI CEO Mary Giliberti, “confirms a painful reality for millions of adults and children living with mental health conditions that despite passage of a federal health parity law, discrimination is still rampant.”
Why do so many people who need care for mental health and substance abuse disorders go out of network to receive it? Networks may be getting narrower. A 2017 Avalere analysis reported that 31% of ACA exchange plans were either preferred provider organizations or point-of-service plans, down from 52% in 2014. “Network type is not a perfect way to predict the scope of physicians and hospitals covered by a specific plan; however, it is an indicator that exchange plans are moving toward networks with fewer providers,” says Elizabeth Carpenter, senior vice president at Avalere.
Challenges of compliance
The phrase “nonquantitative treatment limitations” is a mouthful, but it comes up in almost any discussion about behavioral health coverage. Quantitative treatment limits are in the parity law’s wheelhouse and cover copayments, deductibles, and caps on provider visits or inpatient days. Nonquantitative treatment limitations, or NQTLs, include prior authorizations, concurrent review, formulary design, network adequacy, and fee levels.
Before federal parity law was passed, quantitative treatment limits on behavioral health coverage were rampant in employer-sponsored plans, but “virtually all plans dropped such limits” afterward, according to a 2017 Psychiatric Services study. But the authors also said that NQTLs have to be fixed to improve access to behavioral health care. Frank, the former Obama administration official, agrees: “If you look at the large-group market, which has the longest history as far as parity goes, you see that plans have done a pretty darn good job at doing things like aligning copayments and coinsurance and limits and deductibles with the regulations,” Frank says.
But NQTLs are a different story, particularly in the individual and small-group markets. “There are a couple reasons for that,” he says. “One, it’s just hard; and the second is there remain incentives to try to avoid folks with mental and addictive illnesses. So plans naturally try to find the limits of the regulations.”
In a letter last year to then-HHS Secretary Tom Price, AHIP executive vice president for clinical affairs Richard Bankowitz, MD, listed the challenges insurers have when it comes to complying with the parity law. He cited needs for flexibility in sharing substance use information among providers and plans to support care coordination, for developing a more robust quality measurement infrastructure, for providing validated accreditation standards, and for flexibility in the implementation of medical management programs.
In a 2017 MCG Health report on behavioral health parity, Monique Yohanan, MD, stated that plans have had a difficult time understanding NQTLs. For one, medical diagnoses are based on objective, evidence-based parameters. A diagnosis of diabetes, for example, is based on a blood test. For mental health and substance abuse disorders, those types of laboratory tests don’t exist. Yohanan notes in the report that defining what exactly “parity” and “equity” mean pose an additional obstacle to compliance. And providing equivalent care to people with serious mental illness can be difficult because they’re far more likely to have socioeconomic issues—discrimination, poverty, and alienation from support systems—than the average medical-surgical patient.
Like Frank says, “it’s just hard.”
Now that the federal parity law has been on the books for about eight years, NAMI and other groups are pushing for tougher enforcement. Every state has laws that address coverage for mental health or substance abuse, or both, but not all mandate parity. NAMI has called on federal and state governments to conduct compliance audits of health plans, and for plans to increase reimbursement rates for mental health providers and encourage wider use of telemedicine for behavioral health. NAMI has also suggested that advanced practice nurses and others with appropriate training be allowed to prescribe mental health medications (see related story on page 20).
Coverage of mental health and substance abuse treatment has come a long way since the passage of MHPAEA. Still, implementation of parity is incomplete. Congress probably won’t do much to remedy the situation, but it doesn’t seem inclined to trash the idea of parity.