Assessing the Impact of Parity in the Large Group Employer-Sponsored Insurance Market: Final Report. 5. CONCLUSIONS


MHPAEA Had a Positive Impact on Utilization of Outpatient Behavioral Health Services at the Mean

Overall, the findings demonstrate a significant impact of MHPAEA on average utilization of MH and SUD outpatient services. We observed a significant impact on access to SUD outpatient services, demonstrated by a greater percentage of individuals utilizing one or more service. We also observed a significant impact on frequency of service use for both MH and SUD outpatient services. What is important about these results is that these effects were not isolated to high utilizers (as demonstrated by our analyses of outcomes at the 95thpercentile). Because we found effects for outcomes at the mean, we can conclude that the impacts of MHPAEA on outpatient service utilization were broad in scope and evident for the average user of outpatient BH services.

The Impact of MHPAEA Was Particularly Strong for SUD Outpatient Services

Although the impact of MHPAEA on utilization was evident for both MH and SUD outpatient services, we observed a far larger impact on SUD services. For example, the impact of MHPAEA on the frequency of SUD outpatient services was roughly ten times larger than the impact on MH outpatient services. For SUD services, the increase in spending also was due primarily to an increase in the average number of outpatient visits, although here there was also a modest effect of parity on any use of services. This impact on any use of services is not unexpected, given that MHPAEA expanded parity to include SUD services, and thus some insurance companies may have added SUD coverage where previously there was none.

Although these impacts suggest that parity has played a role in eliminating the treatment gap for SUD services, it is important to note that in this study sample in 2014, only 1.2 percent used any SUD services, and only 0.9 percent used outpatient SUD services. The most recent data from the National Survey on Drug Use and Health estimated that one in 12 individuals needed specialty treatment for SUD and only 10.8 percent of these individuals got it.[35] Hence, despite measurable impacts of MHPAEA on SUD service use, there is still a long way to go to eliminate the SUD treatment gap.

Impact of MHPAEA on Utilization of SUD Outpatient Services Was Not Due to the Opioid Use Disorder Epidemic

To assess whether these findings for SUD treatment primarily were due to MHAPEA or were driven instead by the OUD epidemic, we stratified our SUD analyses into treatment for OUD diagnoses compared with treatment for non-OUD SUD diagnoses. These analyses demonstrated identical patterns across the two groups, increasing our confidence that the overall changes in utilization post-parity were due to parity. However, we did observe a greater magnitude of impact for OUD outpatient treatment, suggesting that the influx of individuals with OUD diagnoses during the same time frame as parity implementation interacted to some extent.

We saw very little in terms of clear differences between the OUD and non-OUD diagnosis groups, increasing our confidence that the changes that we saw at the point of parity implementation were due to parity and not to an increase resulting from the OUD crisis.

There Was No Evidence That MHPAEA Resulted in Increased Costs per Service for the Enrollee or the Provider

Although we found that MHPAEA had a positive impact on both insurer and enrollee spending on outpatient BH services, the overall impact of parity on reimbursement rates and out-of-pocket spending per visit was negligible. The average price paid per outpatient service did not increase for the enrollee or the insurer. Parity's impact on increased outpatient service utilization was the driver of the impact on spending for BH outpatient services (for both the insurer and the enrollee).

MHPAEA Led to a Dramatic Shift to Out-of-Network Spending for Outpatient Services Due to Parity

Our analyses of in-network and out-of-network spending demonstrated that the increases in the ratio of out-of-network spending to total outpatient spending were not due primarily to the opioid epidemic. Prior to MHPAEA, the ratio of out-of-network spending to total outpatient spending was declining for non-BH, MH, and SUD. Following MHPAEA, this declining trend continued for non-BH and MH, but the trend for SUD switched directions and began to increase quite substantially. These findings are consistent with McGinty et al.,[36] who found a similar shift to out-of-network spending following MHPAEA. We advanced her findings by exploring whether the opioid use epidemic was driving these differences by isolating the transition period for these findings and by examining the trends over a longer period of time.

We saw a positive and substantial increase in the level of out-of-network spending for non-OUD services, but no significant impact on the level of OUD services at the point of MHPAEA implementation. However, we saw a similar increase over time in the ratio of out-of-network spending to total outpatient spending for both OUD and non-OUD treatment categories. Further, our analysis of the transition period illustrates that the shift to out-of-network spending began well in advance of the implementation of the Interim Final Rules in 2011. These findings suggest the importance of further analyses of provider incentives for providing SUD services in-network versus out-of-network, as well as analyses to determine whether limited capacity or narrower insurer networks are driving these results.

Differential Impacts of MHPAEA on Enrollee Spending for Outpatient Visits for Those with SMI and OUD

Our subpopulation analyses found no impact of MHPAEA on enrollee spending for the SMI population group, suggesting that it continues to constrain cost sharing for those with SMI. The 1996 parity law applied to annual and lifetime limits for those with an MH condition. These findings suggest that the impact of that law on individuals with SMI, and the subsequent MHPAEA legislation, effectively constrained total out-of-pocket spending for both MH and SUD services. We also can assume that the continued patterns of decreased out-of-network spending for those with an MH condition also kept out-of-pocket spending in check for this population.

However, for those with OUD, we observed higher enrollee spending for SUD outpatient services. Perhaps more important, we found that enrollee spending for MH outpatient services among those with OUD also were positively affected post-MHPAEA. These findings suggest that the shift to out-of-network delivery of SUD care also extended to MH services among this population and could potentially be a barrier to care.

We Found No Evidence That Effects of MHPAEA Are Leveling Off Over Time

One methodological advantage of this study is that it extends the post-period to the third quarter of 2015, well beyond the point of other published studies in this area. Our analyses demonstrate that in the large employer-sponsored insurance market, the impact of parity, particularly on SUD outpatient services, is continuing to grow. The trend lines show no evidence that they are leveling off, and impact on these services is evident into 2015. The early focus of changes due to MHPAEA was on QTLs and FRs. This continuing impact into later years may be due to increasing monitoring and compliance with respect to NQTLs, as well as disclosure requirements over time.

Next Steps

Greater access to BH services due to MHPAEA has also resulted in increased spending. This increase in spending is driven primarily by this increased utilization of services. Although spending for SUD services increased substantially, it is important to note that use of SUD services is a small fraction of overall BH service use. Because the effects on spending for MH outpatient services were very moderate, the effects of an increase of SUD insurer spending on overall insurer spending should not be a significant policy concern.

What is important to address and requires further analysis is the dramatic shift toward out-of-network providers for SUD services. Future research should investigate the reasons for this shift and the implication that SUD provider networks are inadequate. Patients may find it increasingly difficult to find in-network providers. Important questions include whether these patterns suggest provider shortages or use of more narrow provider networks by insurers following parity. We also need a better understanding of provider incentives for in-network versus out-of-network SUD services. It will be important to consider whether the rates that insurers pay are insufficient to attract providers to accept insurance. Qualitative and quantitative work is needed to understand why SUD providers opt out of insurance networks.

Similarly, future analyses need to consider why parity has increased out-of-pocket spending for SUD services. Our results suggest that this is due primarily to increased service use and that parity did not increase the average amount paid per visit. Yet general health care trends, for example, through increased use of high-deductible plans, have trended toward higher out-of-pocket spending. More research is needed on the additional burden of the out-of-pocket spending attributed to the increased service use that parity has facilitated, particularly for OUDs to assess whether this cost sharing is a barrier to adequate receipt of SUD treatment.

The following are some next steps for future research:

  • More research is needed on insurer SUD networks, including SUD provider decision making, to better understand the shift to out-of-network spending for SUD outpatient services.

  • Further analysis of the impact of MHPAEA on utilization and spending outcomes, stratified across in-network and out-of-network services is needed.

  • Next steps also could include a deeper dive to examine the impact of MHPAEA on isolated outcomes of interest beyond outpatient services, such as utilization and spending on prescription drugs for SUD.

  • This research focused on large employer-sponsored insurance; it is important to investigate the impact of parity laws on other types of private insurance and other payers such as Medicaid.