Taken as a whole, analyses presented in this report show that employers and health plans have made substantial changes to their plan designs in order to comply with MHPAEA and the IFR. Our sources indicate that by 2011, most ERISA-governed group health plans and health insurance offered in connection with group health plans removed most financial requirements that did not meet MHPAEA standards. Nearly all eliminated the use of separate deductibles for MH/SUD treatment and medical/surgical treatment. The number of plans that apply unequal inpatient day limits, outpatient visit limits or other QTLs for MH/SUD dropped substantially.
Although we document substantial changes since the enactment of MHPAEA, a substantial minority of large employers and health plans still offer some benefits that appear to be inconsistent with MHPAEA and the IFR. Data from 2011 suggests that one out of five large employers required higher copays for in-network outpatient MH/SUD services than for equivalent medical/surgical treatments. Coinsurance rates were still higher for in-network outpatient MH/SUD services than for medical/surgical services in 4% of large employers' plans.
Likewise, preliminary analyses of our limited sample of midsized employer SPDs suggests that in 2010-2011, a substantial minority of the health plans offered by businesses with between 51 and 500 employees required greater cost-sharing for in-network outpatient MH/SUD office visits than for equivalent PCP office visits.
Although the percentage of plans providing benefits that appeared not to conform to MHPAEA's other quantitative limits was much lower in our sample of plans for 2011 compared to 2010, a minority of plans in 2011, between 7% and 9%, still covered fewer MH and SUD inpatient days annually and fewer MH and SUD outpatient visits annually than they covered for medical/surgical conditions.
Assessing compliance with NQTLs is difficult from document review and self-report from employers and plans. We assessed NQTLs through a detailed review of plan documents and responses from an extensive questionnaire administered to plans' MH/SUD and medical/surgical vendors. Our analyses uncovered numerous areas of concern which warrant more intensive investigation. For example, in 2010, nearly three in ten plans used more stringent precertification and utilization management controls for MH/SUD than for medical/surgical conditions. Network management processes were inconsistent, with different standards and processes for including MH/SUD providers in plans' network than were used for medical/surgical providers. MH/SUD provider reimbursement rates were sometimes found to be set at a lower percentage of prevailing community rates than comparable medical/surgical rates. Rates were sometimes determined by the plan based on its internal data, but set medical/surgical reimbursement rates from external, multi-payer databases. Although we were able to identify some areas of non-compliant NQTLs, it is likely that our reliance on these limited sources of information drawn primarily from large employers' health plans resulted in a significant under-identification of non-complaint NQTLs. A careful, in-depth and longitudinal compliance monitoring of plans' NQTL policies and practices would be likely to turn up correctable problems that our analysis could not detect. The California Department of Mental Health's processes for monitoring plans' compliance with California's Mental Health Parity Act included onsite surveys, reviews of claims files, utilization review files, and internal management and performance reports. California was able to detect patterns in practice that could not be identified from the kind of reviews undertaken in the current report: plans incorrectly denying coverage for ER visits; plans were failing to monitor whether beneficiaries had reasonable access to after-hours services; and plans failed to include required information in claim denial letters.69
Some concerns about the impact of MHPAEA were not borne out in our analyses. A very small proportion of employers, between 1% and 2%, dropped or plan to drop coverage for MH or SUD, or for specific MH/SUD diagnoses as a result of MHPAEA. No employers reduced medical/surgical benefits to comply with parity. A very small percentage excluded specific conditions, and most of those were for learning disabilities, developmental delays, and court-ordered services. We did not detect any movement to exclude residential or intensive outpatient services.
Whether the changes that we observed in employers' and health plans' benefit designs constitute compliance with MHPAEA will have to be tested over time in actual practice. Parity should result in greater access to care, improved quality of services, and better outcomes for people with mental illnesses and SUDs.
Limitations. Although it is reasonable to assume that many of the changes we have documented were made in reaction to the implementation of the MHPAEA, it is important also to recognize that other legislative and employer-specific initiatives may have influenced plan design changes that occurred during this time period. Therefore, caution should be used when interpreting these changes as solely attributable to MHPAEA and the IFR.
It is also important to note that many of the comparisons made in this report rely on data obtained from two distinct data sources: the Aon Hewitt database and the Milliman database. Although the general characteristics of employers included in these two databases are similar, there is insufficient information on employers included in each sample to conclude that they are statistically comparable. Therefore, some of the observed differences between these two datasets may be attributable to inherent differences between the two samples, rather than to changes attributable to the implementation of MHPAEA.
In addition, there are significant limitations associated with our analyses of the BLS dataset. One notable limitation is the lack of detailed establishment information provided with the data. The most important characteristic needed to describe differences in establishments is the number of workers at the establishment. Of secondary importance are the industry classification and the physical location of the establishment. We were only provided information on basic industry categories. Therefore, we believe the weights as created, and applied in our analyses, are insufficient to remove all potential bias from the sample.
Our BLS analyses are also limited by the small number of health plans included in each subsample and the amount of information that could be obtained from each SPD. In some cases, plan information was limited to data obtained from a one-page table of benefits, making abstraction of some data points problematic, and further reducing our sample sizes. Because the number of plans included in each subsample is relatively small, it is only possible to detect relatively large changes between the pre-parity and post-parity samples with any certainty. Therefore, caution should be exercised when interpreting the results of these analyses.
Finally, the results of our health plan/vendor interviews should be interpreted with appropriate caution. Participating respondents represent only a small convenience sample of MBHOs. Although they include some of the largest firms in the field, they represent only a fraction of all MBHOs in the United States. Because the MBHOs work with many health plans, the responses tended toward commonalities; they will not reflect the experiences of every patient or plan associated with these MBHOs. Finally, we made no attempt to verify the information provided by respondents. Their comments should be viewed as the informed opinions of employees.
69. 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.