Consistency of Large Employer and Group Health Plan Benefits with Requirements of the Paul Wellstone and Pete Domenici Mental Health Parity and Addiction Equity Act of 2008. Brief Review of the Existing Literature

11/01/2013

Necessity of Compliance Testing. The history of parity legislation shows that implementation of requirements in this area is not always straightforward and ensuring equitable treatment of MH and SUD treatment is often complicated. Experience with implementation of the Mental Health Parity Act of 199626 is a case in point. The 1996 Act mandated elimination of unequal annual and lifetime dollar limits on MH coverage in employer-sponsored and group health insurance plans. Compliance monitoring found that most health plans complied by eliminating dollar limits but increased restrictions on the number of hospital days or outpatient visits for MH services.27 Findings reported by the U.S. Government Accountability Office (GAO) are representative. Of 863 employer plans responding to its 1999 survey:

  • 14% of employers had not complied with the law by 1999.28
  • 51% reduced the number of outpatient visits covered.
  • 36% reduced hospital days covered.
  • 20% increased outpatient visit copayments.
  • 18% increased the cap on enrollee out-of-pocket costs.29

Research studies focusing on implementation of previous parity requirements such as those applied to FEHBP can complement our other sources of information and enhance our understanding of the impact of MHPAEA.

FEHBP Parity. Monitoring of FEHBP parity implementation3031 revealed that all FEHBPs complied with parity, that no plan reported major problems implementing parity, and that no plan left the program to avoid implementing the policy. Plans enhanced their pre-parity MH/SUD benefits as required by the policy change (84% enhanced MH, 75% enhanced SUD benefits)32 and were more likely to carve-out the behavioral health benefit. Other expected changes (e.g., increased gate keeping at the primary care provider level, reduced provider networks, concurrent or retrospective review, use of disease management programs for MH/SUD care, and increased financial risk sharing) occurred infrequently.

Evaluations of FEHBP parity found no significant increase in total behavioral health spending. Nor did evaluations find an increased probability of any MH/SUD service utilization resulting from parity.33 In fact, the quantity of MH/SUD services patients received may have decreased slightly after parity was introduced. A recent study by Goldman and colleagues found that beneficiaries in plans that were subject to FEHBP parity demonstrated larger reductions in overall behavioral health visits, medication management visits, psychotherapy visits, and prescriptions for behavioral health medications (which the authors assume resulted from increased use of utilization management techniques by plans) following the introduction of parity than did a matched comparison group not subject to FEHBP parity.34 However, introduction of FEHBP parity was associated with a significant decrease in out-of-pocket spending for MH/SUD services.353637

A separate study of the impact of parity on substance abuse treatment in FEHBP plans found that although the rate of out-of-pocket spending declined significantly for substance abuse treatment and more patients had new diagnoses of a SUD, there were no differences in rates of initiation and engagement in treatment under parity and total plan spending per user and average utilization of substance abuse services did not change.38

Researchers have examined the effects of FEHBP parity on specific populations, services, and diagnoses. A recent study examined utilization and costs for individuals having one of three diagnoses representing a continuum of condition severity: bipolar disorder, which was classified as both severe and chronic in nature; major depression, whose severity and chronicity vary considerably in the population; and adjustment disorder, which was classified as a less severe, non-chronic condition.39 Results suggested that, compared to a matched control group, enrollees having each of these conditions demonstrated no significant changes in utilization associated with medication management, inpatient days, or prescriptions following the implementation of parity. In the adjustment disorder group, there was a small, but statistically significant, reduction in psychotherapy utilization. Additional analyses revealed no changes in total behavioral health spending for individuals with bipolar disorder or major depression and small decreases in spending associated with individuals diagnosed with adjustment disorder.40 Out-of-pocket spending related to MH/SUD treatment decreased across all three diagnostic categories vs. the matched control group.

Another recent study of FEHBP parity attempted to identify specific subpopulations of beneficiaries who benefited most from the introduction of parity. Applying growth mixture modeling techniques to FEHBP data, Neelon and colleagues concluded that the effects of parity differed depending on an individual's pre-parity utilization patterns. Three distinct subgroups emerged: "low-spenders," (who had low levels of utilization of MH/SUD services in the pre-parity period) -- their utilization of MH/SUD services declined in the post-parity period; "moderate-spenders," (who had moderate pre-parity spending) -- their spending increased following the implementation of parity; and "high-users," (who had high MH/SUD spending during the pre-parity) -- their spending continued to be high in the post-parity period.41 Another study found that among enrollees who received MH treatment for a severe mental illness (e.g., schizophrenia, bipolar disorder, depression), the odds of using any MH/SUD services in subsequent years were more than 1.3 times greater than two matched control groups.42 The relative odds of using inpatient MH/SUD services in the parity group were 0.67 times that of the control groups, a decrease consistent with the hypothesis that managed care organizations might have guided patients toward more outpatient services in treating their severely ill enrollees. Prescription usage under parity appears to have increased. Individuals covered under FEHBP parity were 1.4 times more likely to fill any behavioral health prescription compared to their non-FEHBP counterparts. An analysis of the impact of FEHBP parity on rates of treatment for depression found no significant changes in rates of diagnosis of depression following introduction of parity and very little change in measures of the quality of care.43

Several additional evaluations of FEHBP parity have focused on the effects of the program on children and adolescents. Azrin and colleagues concluded that, following the introduction of FEHBP parity, children enrolled in the FEHBP program showed no significant increase in MH/SUD utilization compared to a matched control group.44 These findings are consistent with analyses of the impact of state parity laws that show no significant impact on access for children and adolescents.45 In evaluating only children and adolescents with high MH/SUD expenditures in the pre-parity period, a recent study concluded that compared to a matched control group, children enrolled in the FEHBP showed similar patterns of MH/SUD expenditures following the introduction of parity, but a statistically significant reduction (approximately $258 in 2011 dollars) in average out-of-pocket spending associated with MH/SUD services.46

In general, these studies of FEHBP parity found no significant increases in overall MH/SUD utilization rates, initiation or engagement rates, or total MH/SUD spending following the implementation of parity but significant decreases in out-of-pocket costs did result.

Vermont. Compliance monitoring of the MHPAEA can also be guided by the findings of studies examining the effects of state-level parity, such as Vermont.47 The Vermont Parity Act took effect January 1, 1998.48 The Vermont legislation mandated group health insurance to cover MH/SUD treatment equitably with other covered medical treatments (ERISA-governed self-insured plans are exempt from state parity legislation). An evaluation of the law's effects found an increased probability of an individual receiving any outpatient MH services and a decreased likelihood of an individual receiving any substance abuse services following the introduction of parity. The percentage of beneficiaries receiving outpatient MH services increased by a range of 6%-8%. The percentage of individuals receiving any substance abuse services decreased by a range of 16%-29%.49 Results also indicated that, in general, consumer cost-sharing for MH and substance abuse treatment services declined, from 27% to 16% of total costs, following the implementation of parity. The evaluation of the Vermont law's effects found little evidence that the introduction of parity resulted in employers dropping health coverage or switching to self-insured plans to avoid complying with the regulation. Only 0.3% of Vermont employers reported that they dropped health coverage for their employees primarily due to the parity law, and only 0.1% of employers reported that parity played a role in their decision to self-insure (to avoid complying with state law).50

Use of managed care techniques increased following Vermont's implementation of parity. Although one of the two major health plans already used managed care before the implementation of parity, the other health plan also shifted most of its members to a managed behavioral health care carve-out. In one plan, spending increased modestly by 19 cents per member per month (PMPM). Nonetheless, MH/SUD services accounted for only 2.5% of total spending in that plan after parity compared to 2.3% before parity. The other plan experienced a 9% decrease in spending for MH/SUD services following implementation of the state parity law. This decrease in spending was largely attributed to a decrease in SUD treatment service utilization.

Employers' knowledge of the parity law remained low, even after its implementation. A survey conducted 2 years after the implementation of parity suggested that approximately 50% of all fully-insured employers in Vermont had never heard of the parity law and that nearly three-fifths of all employers had little to no knowledge of the parity law.51 Small and medium-sized businesses were least likely to be familiar with the law, with approximately 70% of those employers having little to no knowledge of the law. Although the two major health plans in Vermont complied with the law on paper, lack of information, confusion, and mistakes by the state's largest plan generated complaints from beneficiaries and providers that led to changes in administration and consumer education in succeeding years.52

Oregon. Oregon's parity law, implemented January 1, 2007,53 mandated that group health insurance plans provide coverage for MH and substance abuse treatment services at the same level as other medical conditions. Results from Oregon are particularly informative for the current project in that the Oregon law, like the MHPAEA, went beyond the regulation of financial and QTLs and specified that plans cannot utilize unequal, NQTLs for MH and substance abuse treatment services compared to medical/surgical services. A recent analysis of the Oregon law suggested that each of the four plans studied made substantial changes to their MH and substance abuse treatment benefits following the implementation of parity. Each plan removed coverage limits related to inpatient and outpatient MH/SUD treatment services. After implementation of the NQTL provisions in the Oregon law, the use of management techniques stayed the same or decreased in the insurance plans studied. These changes were made without significant increases in total MH/SUD treatment spending. Importantly, the researchers found that these effects were achieved without the increased use of utilization management techniques.54 The authors also found no evidence of meaningful change in the rates of any behavioral health care service use.

In a separate analysis of only substance use spending, McConnell55 found that expenditures for alcohol treatment services increased significantly and spending on other drug abuse treatment services did not. The introduction of parity was associated with a small, but not statistically significant, increase in overall substance use treatment spending. In another study analyzing the impact of parity in Oregon on access to various behavioral health specialists, McConnell found that parity was associated with a slight increase (from 0.5% to 0.8%) in behavioral health treatment initiations with masters-level specialists, and relatively few changes for generalist physicians, psychiatrists, and psychologists. Patients were particularly sensitive to distance for non-physician specialists:56 the greater the distance between an individual and a non-physician specialist, the less likely that individual was to receive treatment. Following the introduction of parity, distance to the nearest psychiatrist, masters-level therapist, or psychologist tended to decrease.

California. California's Mental Health Parity Bill, which became effective on July 1, 2000, mandated that all group and individual health plans offer MH coverage as part of their overall health benefits and outlawed the use of MH treatment limitations and cost-sharing requirements that were more restrictive than those for physical health conditions.

The law required that health plans provide MH services to seriously mentally ill (SMI) adults and all children with serious emotional disturbances. Nine specific SMI diagnoses were included in the mandate: anorexia nervosa, bulimia nervosa, bipolar disorder, major depression, obsessive-compulsive disorder, panic disorder, pervasive developmental disorder/autism, schizophrenia, and schizoaffective disorder. SUDs were not covered by the California Parity Act. To assess health plan compliance with the Mental Health Parity Bill, the California Department of Mental Health undertook an intensive review of health plans that included an onsite survey, reviews of claims files, utilization review files, and internal management and performance reports. The report identified several areas of non-compliance. Six out of seven California plans that were subject to the legislation were incorrectly denying coverage for emergency room (ER) visits; five out of seven plans were failing to monitor whether beneficiaries had reasonable access to after-hours services; and five out of seven plans failed to include required information in claim denial letters.57

Trends in MH/SUD Spending and the Costs of Parity. An analysis by Mark and colleagues examined trends in behavioral health spending between 2001 and 2009 for a sample of over 100 large, self-insured employer plans. Results concluded that the average contribution of behavioral health care spending to total health care spending across each of the years examined was 0.3%, and only 2% of employers experienced a rate increase of more than 1% per year attributable to behavioral health costs.58

Given the small contribution of behavioral health care costs to overall health care costs, MHPAEA is expected to result in only very modest increases total health care expenditures. The Congressional Budget Office (CBO) estimated that MHPAEA itself would result in very modest cost increases, approximately 0.4%, in employer-sponsored group health care premiums and 0.2% in Medicaid payments to managed care plans.59 Recent analyses by Mark and colleagues utilizing MarketScan data are consistent with the CBO's estimate. Their analyses have suggested that an overwhelming majority of privately insured beneficiaries who utilized behavioral health care benefits in the pre-parity era did so at a rate that was far below pre-parity health care limits.60 Using econometric models to estimate the detailed effects of the MHPAEA on high-utilization beneficiaries who are likely to use its expanded coverage, these researchers estimated that the MHPAEA will likely increase total health care costs by 0.4%.

Early MHPAEA Compliance Analysis. In November 2011, GAO issued an early report on MHPAEA compliance in response to a statutory requirement.61 One hundred sixty-eight employers responded to a GAO survey asking detailed questions about changes in their behavioral health benefits between 2008 and 2010/2011 out of 707 employers who received the survey. Although the findings from this survey are not generalizable given the response rate of 24%, the survey did generate information on some questions regarding diagnoses covered not addressed in other studies. The vast majority of responding employers offered MH/SUD coverage in both 2008 and in 2010/2011, and most employers reported covering the same broad range of MH/SUD diagnoses in their current plan year as they also did in 2008. The remaining employers reported including more broad diagnoses.

In keeping with findings in other studies, employers responding to the GAO survey reported reducing their use of MH/SUD office visit and inpatient day limitations. In 2008, a significant percentage of these employers reported utilizing office visit limitations for SUDs. In 2010/2011, far fewer of these employers reported having such limitations. Likewise, in 2008, a significant percentage of employers reported utilizing limitations on inpatient days related to behavioral health conditions. By 2010/2011, the percentage of employers reporting using such limitations had dropped. The GAO did not assess NQTLs used by employers and health plans. While the results of the GAO survey should be interpreted with caution due to its small sample size and low response rate, the results from the survey suggest that employers were generally able to implement changes required by MHPAEA with little disruption to the insurance market.


  1. P.L. 104-204.

  2. Allen KG. Mental Health Parity Act: Employers' mental health benefits remain limited despite new federal standards. Testimony before the Committee on Health, Education, Labor and Pensions, U.S. Senate. Washington, DC: General Accounting Office, May 18, 2000. (Document no. GAO/T-HEHS-00-113.)

  3. Burnam MA, Escarce JJ. Equity in managed care for mental disorders. Health Aff. 1999; 18:22-31.

    Hennessy KD, Goldman HH. Full parity: Steps toward treatment equity for mental and addictive disorders. Health Aff. 2001; 20:58-67.

  4. Allen KG. Mental Health Parity Act: Employers' mental health benefits remain limited despite new federal standards. Testimony before the Committee on Health, Education, Labor and Pensions, U.S. Senate. Washington, DC: General Accounting Office, May 18, 2000. (Document no. GAO/T-HEHS-00-113.)

  5. Goldman HH, Frank RG, Burnam A et al. Behavioral health insurance parity for federal employees. N Engl J Med. 2006; 354:1378-1386.

  6. Lichtenstein C et al. Evaluation of parity in the Federal Employees Health Benefits (FEHB) program: Final Report. Presented to the U.S. Department of Health and Human Services Administration, Office of the Assistant Secretary Planning and Evaluation, 2004. Available at: http://aspe.hhs.gov/daltcp/reports/parity.htm.

  7. Ibid.  

  8. Goldman HH, Frank RG, Burnam MA et al. Behavioral health insurance parity for federal employees. N Engl J Med. 2006; 354:1378-1386.

  9. Goldman HH, Barry CL, Normand ST, Azzone V, Busch AB, Huskamp HA. Economic grand rounds: The price is right? Changes in the quantity of services used and prices paid in response to parity. Psychiatr Serv. 2012; 63:107-109.

  10. Goldman HH, Frank RG, Burnam MA et al. Behavioral health insurance parity for federal employees. N Engl J Med. 2006; 354:1378-1386.

  11. Azrin ST, Huskamp HA, Azzone V, et al. Impact of full mental health and substance abuse parity for children in the Federal Employees Health Benefits Program. Pediatrics 2007; 119:452-459.

  12. Azrin ST, Huskamp HA, Azzone V et al. Impact of full mental health and substance abuse parity for children in the Federal Employees Health Benefits Program. Pediatrics 2007; 119:452-459.

  13. Azzone V, Frank R, Normand S, Burnam A. Effect of insurance parity on substance abuse treatment. Psychiatr Serv. 2011; 62:129-134.

  14. Busch A, Yoon F, Barry C et al. The effects of parity on mental health and substance use disorder spending and utilization: Does diagnosis matter? Am J Psych. In press.

  15. Busch A, Yoon F, Barry C et al. The effects of parity on mental health and substance use disorder spending and utilization: Does diagnosis matter? Am J Psych. In press.

  16. Neelon B, O'Malley AJ, Normand ST. A Bayesian two-part latent class model for longitudinal medical expenditure data: Assessing the impact of mental health and substance abuse parity. Biometrics 2011; 67:280-289.

  17. Yoon FB, Huskamp HA, Busch AB, Normand ST. Using multiple control groups and matching to address unobserved biases in comparative effectiveness research: An observational study on the effectiveness of mental health parity. Stat Biosci. 2011; 3:63-78.

  18. Huskkamp HA. The impact of parity on major depression quality in the Federal Employees Health Benefit Program. Med Care 2006; 44:506-512.

  19. Azrin ST et al. Impact of full mental health and substance abuse parity for children in the Federal Employees Health Benefits Program. Pediatrics 2007; 119:452-459.

  20. Barry CL, Busch SH. Caring for children with mental disorders: Do state parity laws increase access to treatment. J Ment Health Policy Econ. 2008; 11:57-66.

  21. Azrin ST et al. Impact of full mental health and substance abuse parity for children in the Federal Employees Health Benefits Program. Pediatrics 2007; 119(2):452-459.

  22. Rosenbach M, Lake T, Young C et al. Effects of the Vermont Mental Health and Substance Abuse Parity Law. DHHS Pub. No. (SMA) 03-3822. Rockville, MD: Center for Mental Health Services, Substance Abuse and Mental Health Services Administration, 2003.

  23. Vermont State Legislature. (2000) Section 4089, Mental Illness. Available at http://www.leg.state.vt.us/statutes/titles08/chap107.htm.

  24. Ibid.  

  25. Rosenbach M, Lake T, Young C et al. Effects of the Vermont Mental Health and Substance Abuse Parity Law. DHHS Pub. No. (SMA) 03-3822. Rockville, MD: Center for Mental Health Services, Substance Abuse and Mental Health Services Administration, 2003.

  26. Ibid.  

  27. Rosenbach M, Lake T, Young C et al. Effects of the Vermont Mental Health and Substance Abuse Parity Law. DHHS Pub. No. (SMA) 03-3822. Rockville, MD: Center for Mental Health Services, Substance Abuse and Mental Health Services Administration, 2003.

  28. ORS 743.556; OAR 830-053-1404, 1405, 1325, 1330; SB 1.

  29. McConnell KJ, Gast SH, Ridgely MS et al. Behavioral health insurance parity: Does Oregon's experience presage the national experience with the Mental Health Parity and Addiction Equity Act? Am J Psychiatry. 2012; 169:31-38.

  30. McConnell KJ, Ridgely MS, McCarty D. What Oregon's parity law can tell us about the federal Mental Health Parity and Addiction Equity Act and spending on substance abuse treatment services. Drug Alc Depend. 2012; 124:340-346.

  31. McConnell KJ, Gast SH, McFarland BH. The effect of comprehensive behavioral health parity on choice of provider. Med Care. 2012; 50:527-533.

  32. California Department of Managed Health Care. Mental Health Parity in California. Mental Health Parity Focused Survey Project: A Summary of Survey Findings and Observations. Available at: http://www.hmohelp.ca.gov/library/reports/med_survey/parity/sfor.pdf.

  33. Mark TL, Vandivort-Warren R, Miller K. Mental health spending by private insurance: Implications for the Mental Health Parity and Addictions Equity Act. Psych Serv. 2012; 63(4):313-318.

  34. Congressional Budget Office. Congressional Budget Office Cost Estimate: S558. March 20, 2007.http://www.cbo.gov/system/files/cbofiles/ftpdocs/78xx/doc7894/s558.pdf.

  35. Mark TL, Vandivort-Warren R, Miller K. Mental health spending by private insurance: Implications for the Mental Health Parity and Addictions Equity Act. Psych Serv. 2012; 63(4):313-318.

  36. For its 2012 parity study, the GAO surveyed a stratified random sample of 707 small, medium, large, and very large employers about the MH/SUD covered in their health plans that covered the greatest number of lives for the most current plan year -- either 2011 or 2010 -- as well as for 2008. A total of 168 employers submitted usable survey responses, for a response rate of 24%. It may be difficult to generalize from this sample to the universe of employers and health plans subject to MHPAEA and the IFR.

View full report

Preview
Download

"mhpaeAct_0.pdf" (pdf, 1.34Mb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®