The Costs of Mental Health Care Parity: Research, 1999-2002

Though the Domenici-Wellstone Parity Act of 1996 did not have a major direct effect on mental health care access, it created the momentum for profound shifts in states’ approaches to parity. Today, 34 states have full or partial mental health care parity legislation. Many state legislatures were persuaded, in part, by early evidence that such legislation would not greatly increase spending on mental health care or on overall health costs. In 1998, the Bazelon Center reviewed then-existing research on the costs of mental health parity. Today, the evidence of parity’s affordability has grown substantially. However, some researchers are sounding notes of caution about the risks involved in controlling mental health care costs through managed care.

Continued Evidence of Affordability

• A 2001 US General Accounting Office report on employers in states without parity legislation (and therefore affected only by the 1996 federal parity law) found that only 3% of responding employers claimed that compliance increased their claims costs. Almost no employers had eliminated their mental health benefits in response to the law. GAO, “Mental Health Parity Act: Despite New Federal Standards, Mental Health Benefits Remain Limited.” GAO/HEHS-00-95 (May 2000).
  
• A study of two large employer groups in California that implemented mental health parity on January 1, 2001 showed decreases or only mild increases in overall health care costs. Employer A, which had overall service utilization rates higher than other employer populations both before and after parity, experienced a 1.9% decline in total mental health spending and large reductions in the use of outpatient, intermediate-care, and inpatient services. Employer B started out with lower rates of utilization before parity, and saw substantial increases in service use and spending. However, the increase in mental health care spending was still less than 1% of total health care spending for Employer B. Ronald Branstrom and Roland Sturm, “An Early Case Study of the Effects of California’s Mental Health Parity Legislation.” Psychiatric Services (Oct 2002).
  
• In 2000, the National Advisory Mental Health Council (NAMHC) requested that the Hay Group update its estimate of the average annual premium increase associated with mental health parity. In 1996, the Congressional Budget Office had estimated total cost increases of 5.3% subsequent to mental health parity implementation in an indemnity plan; the number was 4% with the effects of managed care plans on the market reflected. A 1998 review by Sing and colleagues resulted in a parity cost estimate of 3.6%. The results of the 2000 Hay Group study incorporated newer models and assumptions, and showed a predicted cost increase of only 1.4%. National Institute of Mental Health, “Insurance Parity for Mental Health: Cost, Access, and Quality Final Report to Congress by the National Advisory Mental Health Council.” NIH Publication No. 00-4787. Citing Hay Group (2000): Mental Health Benefit Value Comparison MHBVC Version 2.0, March 14, 2000: Report Prepared for National Institute of Mental Health.
  
• Another analysis requested by the NAMHC work group looked at the effects of a state mental health parity mandate (in conjunction with carve-out managed care) for a large employer group. Costs per member were reduced by 39% from pre-parity to the third year of parity, and the portion of the population receiving mental health services increased. Costs dropped by 9% between Year 1 and Year 4 for adults, and by 75% for dependent children, mostly because of decreases in pediatric inpatient stays. Zuvekas SH, Regier DR, Rae D, Rupp A, and Narrow W (2000): The differential impacts of mental health parity and managed care: Evidence from a large natural experiment. (working paper).
  
• In 1997, research by Ronald Sturm showed that the cost to insurers of removing average dollar limits on mental health benefits would be small. Taking away an annual limit of $25,000 per year would raise costs by approximately $1 per employee per year. Removing a stricter annual limit of $10,000 per year would yield increased costs of $4 per enrollee per year. In 1999, Sturm and colleagues performed a similar analysis of substance abuse benefits. Relying on 1996-7 claims from 25 managed care plans with unlimited substance abuse (SA) benefits, they found that eliminating an annual limit of $10,000 per year on substance abuse care would raise insurance payments by about 6 cents per member per year. Eliminating a stricter limit of $1,000 would increase payments by about $3.40. Both changes would greatly increase access to services. Ronald Sturm, “How Expensive us Unlimited Mental Health Care Coverage Under Managed Care?” 278 JAMA 18 (1997), 1533-37. Roland Sturm, Weiying Zhang, and Michael Schoenbaum, “How Expensive are Unlimited Substance Abuse Benefits Under Managed Care?” The Journal of Behavioral Health Services & Research 1999; 26(2).
  
• Researchers analyzing thirty medium- and large-size private and public self-insured employers with carve-out benefits through United Behavioral Health found reductions in costs ($0.28/member/year) combined with consistent treatment prevalence, indicating that reductions in spending weren't the result of a loss of access to care. Instead, the cost reductions were linked with increased use of UBH's mental health provider network. Outpatient costs remained fairly constant overall, with a decrease in outpatient episode length and an increase in outpatient users. Inpatient costs decreased by over $2000/year per 1000 members, and inpatient days decreased by 2.5/year per 1000 members. This study addresses many limitations of earlier studies supporting the affordability of mental health care: the scope was national, the employers represented a range of sizes, and actuarial models used were updated to reflect the managed care context. William Goldman, Joyce McCulloch, Brian Cuffel, and Danah Kozma. "More Evidence for the Insurability of Managed Behavioral Health Care." 18 Health Affairs 5 (Sept/Oct 1999), 172-181.
  
• During their efforts to secure parity legislation, Vermont advocates commissioned an actuarial report on the predicted impact of a mental health and substance abuse parity bill. The overall finding was that the average premium increase would be 3.4%. Because of likely employer responses to limit these increases, the expected premium increase was estimated at only 1.4%. For a mental-health-only plan, the predicted premium increases (without employer compensations) was 2.4%. Ken Libertoff, “Fighting for Parity in an Age of Incremental Health Care Reform: A Battle Against Discrimination in the Health Care Industry.” The Vermont Association for Mental Health, January 1999. Citing Ronald Bachman, “An Actuarial Analysis of Comprehensive Mental Health and Substance Abuse Parity and Other Options for Improved Coverage in the State of Vermont.” Coopers and Lybrand, January, 1997.
  
• The President of the Washington Business Group on Health argued that the cost offsets of mental health benefits extend far beyond reductions in medical costs or in inpatient mental health costs. She notes that costs of absenteeism and decreased productivity due to unmet mental health needs are far higher than direct spending for mental health care. Mary Jane England, "Capturing Mental Health Cost Offsets." 18 Health Affairs 2 (March/April 1999), 91-93.
  
• Increased access to mental health services may result in savings through reduced use of medical services. Earlier studies suggests that certain patients, including distressed elderly inpatients, people developing major medical illnesses, adults with alcoholism, and primary care outpatients with numerous somatic problems, may use an excessive amount of medical care because of psychological factors. Utilization managers who are properly trained may be able to capture these savings by appropriate offering of mental health services to those patients whose use of medical services is inappropriate or excessive. The authors note that in underserved populations, such a move might at first lead to an appropriate increase in the use of medical services because people with numerous unmet needs may not have been receiving medical care they needed. Mark Olfson, Merrile Sing, and Herbert J. Schlesinger, "Mental Health/Medical Care Cost Offsets: Opportunities for Managed Care." 18 Health Affairs 2 (March/April 1999), 79-90.

Notes of Caution

• According to a 2002 study, the thirty-four states that have mental health parity legislation use numerous different definitions of mental illness. Broadly, some include “broad-based mental illness,” others “serious mental illness” and others “biologically-based mental illness.” These definitions are not linked to psychiatric diagnostic categories but rather to political and economic decisions in individual states. The authors argue that narrower definitions may actually lead to more costs because they leave a greater burden of untreated mental illness in the state. Marcia Peck and Richard Scheffler, “An Analysis of the Definitions of Mental Health Illness Used in State Parity Laws.” 53 Psychiatric Services 8 (September 2002), 1089-1095.
  
• A study by researchers at the RAND Institute of RWJF data found that the states most likely to have passed parity legislation were those with below-average mental health care utilization. This may be due to state advocates' concerns about low utilization, or to a lack of significant opposition because of relatively low mental health costs in those states. This lower utilization persisted after parity laws were passed, suggesting that parity legislation does not create significant increases in mental health care utilization. Roland Sturm and Rosalie Liccardo Pacula, "State Mental Health Parity Laws: Cause of Consequence of Differences in Use?" 18 Health Affairs 5 (Sept/Oct 1999),182-92.

Prepared by Naomi Seiler for the Bazelon Center for Mental Health Law and submitted December 18, 2002