Implications of the Mental Health Parity and Addiction Equity Act

The Mental Health Parity and Addiction Equity Act (MHPAEA) mandates equity in insurance coverage, including both treatment limits (caps on inpatient days and out-patient visits) and financial requirements (cost sharing, deductibles, and out-of-pocket limits), for behavioral health and medical/surgical services. Some insurers and employers, however, have voiced concerns that implementation of the MHPAEA will lead to larger cost increases than those found in previous studies of parity. Generally, these studies have found that parity can be achieved with few if any increases in total health care costs (1). The new federal law goes further than most previous laws by extending parity to managed care techniques that may not be expressed numerically but may nevertheless limit the scope or duration of services—so-called nonquantitative treatment limitations. These management techniques include requirements such as prior authorization, utilization review, or standards for provider participation in a network. Under the new federal law, insurance plans must use the same processes or strategies that they use for medical/surgical benefits to determine how nonquantitative treatment limitations are set. Because some studies suggest that these management techniques are what allowed parity to occur without large cost increases in the past (2), there are concerns that the new federal law would lead to relatively large cost increases. However, there is no published direct evidence to date on the effect of the MHPAEA on health care costs.

In this issue of the Journal, McConnell et al. (3) examine the effects of Oregon’s state parity law on health care costs. This study is novel in that the Oregon law included nonquantitative treatment limitation provisions that are similar to those found in the MHPAEA. It is thus the first study to provide evidence of the effects of parity in the context of nonquantitative treatment limitations, albeit in a single state. The authors find that this parity law did lead to significant changes in the design of benefits among the four plans studied. Limits to the number of outpatient visits and inpatient days appeared in all four plans before parity, but these limits were eliminated after parity. After the nonquantitative treatment limitation provisions in the Oregon law were implemented, the use of management techniques stayed the same or decreased in the insurance plans studied. This contrasts directly with previous studies of parity (1), which have found parity to be associated with increases in the use of management (through contracts with managed behavioral health carve out firms).

In their careful analysis, McConnell et al. (3) find that the Oregon law did not lead to significantly higher health care expenditures. Controlling for secular trends, the authors find that cost increased by $15 per beneficiary, although this increase was not statistically significant. Enrollees in two insurance plans saw a significant decrease in out-of-pocket costs, while those in another plan saw a significant increase, likely because of increased cost sharing overall, reflecting a national trend in health care toward greater cost sharing in general. The authors also find no evidence of meaningful change in the rates of any mental health care service use, which suggests that the nonquantitative treatment limitations in parity requirements will not “break the bank.” As in previous studies, McConnell et al. find that parity has the beneficial effect of improving risk protection (i.e., reducing out-of-pocket spending) at little to no additional increase in costs.

The absence of large expenditure changes observed in the McConnell et al. study (3) increases our confidence that federal parity can be implemented without significant cost increases. A likely explanation for this finding is that mental health benefits are still managed, but in the same way as medical-surgical benefits. In effect, the Oregon and federal laws bring the management techniques that are used in mental health care in line with those used in general medical care.

The MHPAEA, passed in 2008 and implemented in most plans in January 2010, was critically important legislation for many individuals with mental illness or substance use disorders. Previously, the Mental Health Parity Act of 1996 merely banned the annual or lifetime dollar limits on mental health care services that were higher than those for general medical care. While most states had some form of parity law, they were often limited in scope (e.g., covering only public employees or mandating parity for limited diagnoses). In contrast, the federal law extends comprehensive parity to all states, and it requires equal insurance coverage for a broad range of mental health conditions. Unlike most state laws, the federal parity law specifically includes substance abuse treatment services. Moreover, state parity laws do not apply to plans in which employers assume the risk for health care services (“self-insured plans”), but the federal parity law does. This provision is not inconsequential—the Kaiser Family Foundation estimates that approximately 60% of covered workers are in partially or completely self-insured plans (4).

The Affordable Care Act (ACA), passed in March 2010, further extends the reach of federal parity to additional populations beyond those enrolled in privately insured group plans. The ACA requires that plans sold on the state-based insurance exchanges include coverage for mental health and substance use disorder services, and it requires that these services be covered at parity with medical-surgical benefits as defined in MH-PAEA. In addition, the ACA requires that individuals whose income is less than 133% of the federal poverty line be eligible for Medicaid. The law allows many of these newly Medicaid-eligible individuals to be covered under “benchmark plans,” which have historically been less generous than traditional Medicaid. These plans, however, will also be required to cover mental health and substance use disorder services at parity. Because individuals with mental health disorders are more likely to be uninsured, they will disproportionately benefit from these coverage expansions. One estimate suggests that these insurance expansions will result in an additional 1.15 million new mental health service users (5).

It remains unclear, however, which services will be covered under exchange or benchmark plans. The ACA mandates that mental health and substance use disorder services be covered but leaves the details about the essential benefits package to future rule making. Although it did not note specific services to be included in the essential benefits package, an Institute of Medicine panel recommended in September 2011 that the actuarial value of these plans be pegged to those offered in the small employer market, which tend to cover a more narrow range of services than do large employer plans. Although it has not yet been determined, this recommendation may effectively limit the types of mental health and substance use disorder services covered.