Behavioral Health Insurance Passes Economics Test

Establishing Parity in Coverage for Mental Health Services Fails to Raise Spending

Should mental health and substance-abuse services be insured in the same way that other medical services are? For many years, the opponents of parity—providing equal benefits for behavioral health as for other medical care—have argued that lifting restrictions on coverage would force the costs of health care to rise unacceptably. But the largest nationally representative study to examine the issue suggests that employer-based behavioral health coverage can achieve parity without breaking the bank. In the March 30 New England Journal of Medicine, a team led by researchers at the University of Maryland and HMS shows that providing the same coverage for mental and physical health for federal employees did not increase total costs. There’s a catch, however: in this study, parity went hand-in-hand with managed care, a cost-control system that has been controversial in the behavioral health community.

Photo by Liza Green, HMS Media Services

Clockwise from top, Haiden Huskamp, Sharon-Lise Normand, Vanessa Azzone, and Richard Frank found that federal employees who used behavioral health services paid less money out of pocket under a government-mandated parity program than people in insurance plans without parity.

Push for Parity
Benefits for mental health and substance abuse began to erode in the late 1970s and 1980s. Insurers raised deductibles and imposed stricter limits on hospital stays, office visits, and treatments. Parity advocates have argued that these cuts are rooted in unfair biases against behavioral health care. Today, 34 states have laws that mandate some degree of parity, but the only federal parity law is limited in scope.

In 1999, President Clinton ordered the Office of Personnel Management to ensure parity for the Federal Employees Health Benefits (FEHB) program, which covers 8.5 million current or retired federal employees and their families. As part of the change, he asked that the program be evaluated. Howard Goldman, professor of psychiatry at the University of Maryland, led the Parity Evaluation Research Team, along with a team in the Department of Health Care Policy at HMS and others from UCLA, the Rand Corporation, and Westat, Inc. They saw the federal directive as a perfect natural experiment to test the fears that parity would open a Pandora’s box of spending.

The FEHB program allows employees to choose from a wide variety of insurance plans. For this analysis, the team used data from a set of seven plans based in different parts of the country, including point-of-service and preferred-provider organization plans (two HMOs that were initially chosen had already implemented parity before the study began). The researchers also chose a set of private plans that were outside the FEHB system as controls that did not implement parity during the study period, 1999 to 2004. During this time, the federal plans applied the same deductibles, copayments, and other out-of-pocket costs, and limits on services for behavioral health care as for other medical services.

Drop in Patient Spending
Richard Frank, the Margaret T. Morris professor of health care policy at HMS, led the economic analysis of claims data from the plans. The team conducted a “difference-in-differences” analysis, subtracting the changes experienced by the control group from the changes in the federal plans; this helped them rule out overall trends in cost and use of services.

The team found that implementing parity did not lead to a rush for services or skyrocketing costs. Use of behavioral health services rose across the board during the study period, but people in the federal plans were no more likely to use these services than those in plans that did not have parity. Similarly, spending increased generally, but it actually grew more slowly in three of the plans that implemented parity, and other plans did not differ significantly from controls. Most importantly, Frank said, “People who got sick paid less out of pocket.” Among users of behavioral health services, the parity policy was associated with significant reductions in out-of-pocket spending in five of the seven plans, while the spending increased in only one plan. The study analyzed only one measure of quality of care, the duration of follow-up treatment for depression; the measure did not decline with the implementation of parity and improved slightly in three of the plans, similar to national trends.

The finding that parity did not lead to higher use of services could be seen as a failure to improve access to care. But the authors of the study see the objective of parity differently. “We believe the primary objective of health insurance is to provide financial protection against losses, not necessarily to stimulate demand and spending,” said Goldman. The study showed that it is possible to provide the same financial safety net to people who suffer from behavioral health problems as those who have other medical problems without creating an undue burden on the insurance system.

Managing Costs
Frank emphasized that the results of this study can be understood only in the context of managing care. “It’s managed care that has made parity affordable,” he said. The initial parity directive for the FEHB encouraged plans to manage care; for almost all the plans, that meant contracting with a separate “carve-out” vendor that independently manages behavioral health services, among other strategies. In fact, the one plan in this study that experienced an increase in use did not carve out its behavioral health care.

James Sabin, HMS clinical professor of psychiatry at Brigham and Women’s Hospital and director of teaching in the Department of Ambulatory Care and Prevention, said that the findings of the study imply that “the managed care process facilitates the achievement of an objective.” Sabin also heads the ethics program at Harvard Pilgrim Health Care and has argued for parity in the context of managed care. “I have come to believe over time that all health care should be managed in the right way,” Sabin said. He added that although managed care has been very unpopular within the medical community, with this study, “it’s going to be harder for mental health and substance abuse advocates to simply oppose the use of managed care.”

In fact, many advocates for parity have come to realize that managed care is necessary for easing fears surrounding parity, whether it is ideal for care or not. Kenneth Duckworth, HMS assistant professor of psychiatry at Beth Israel Deaconess Medical Center and medical director for the National Alliance on Mental Illness, said, “Most people would say it’s better to have parity with inconvenience than straight-up discrimination.” Parity, he said, is important as a symbolic achievement, in addition to financially protecting people who need behavioral health services. “There’s a social legitimacy that happens when an illness is considered equal.”

—Courtney Humphries

top