Value-Based Payment and Behavioral Health

By Harold A. Pincus, MD, and Alexa Fleet, MPH

Since the passage of the Affordable Care Act in 2010, a key strategy has been to move the US health care system from one that incentivizes volume with traditional fee-for-service care to one that uses value-based payment (VBP) models. Particularly within behavioral health (BH), the transition to a value-based system faces several unique challenges.

While value is broadly defined as quality over costs, in practice, VBP combines financial elements with flexibility and accountability. Several bundled payment models (e.g., accountable care organizations) apply capitated arrangements allowing flexibility to pay for services not typically paid under fee-for-service, such as care management and those addressing social determinants of health. Accountability can be on a regulatory basis, with plans or practice organizations meeting certain quality standards (e.g., reporting of specified quality measures), as well as financial accountability with variation in how it is applied.

The value of VBP has been openly questioned. Research demonstrates that VBP models can potentially drive cost savings and value improvement in a high-cost health care environment; however, achieving those goals in practice has been difficult. Much can be gleaned from the successes and failures of VBP models in the decade since the Affordable Care Act passed. For VBP to work successfully in BH, we must consider the following challenges.

Quality Measurement and Data Capture

A dearth of valid and feasible quality metrics in BH and a lack of investment in BH measurement development prevent VBP from being applied successfully. Quality metrics, crucial for the success of VBP, are especially challenging in BH with limited practical data sources. Most metrics are process/claims-based measures, few with proven association with health outcomes. Measures based on claims do not give a full picture of the care provided. Typically, there is inadequate BH information captured at the level of granularity necessary to adequately measure quality (e.g., standardized measures of severity/outcomes, content of interventions). BH informatics efforts are less well established than mainstream clinical informatics advancements.

Moreover, developing reliable and valid measures is a complex process requiring effort and costly navigation of a quality measurement–industrial complex to ensure adoption and adoption. There is no clear leadership in developing BH metrics. The National Institute of Mental Health, the National Institute on Drug Abuse, the National Institute on Alcohol Abuse and Alcoholism, and the Substance Abuse and Mental Health Services Administration have no specific responsibilities in developing measures. The US Centers for Medicare & Medicaid Services is the principal lead for measure development but has limited BH expertise.

Substantial barriers exist to accessing the necessary data. Data sources are fragmented across primary care, medical specialists, and BH professionals, as well as payer data with carved-in and carved-out arrangements. Privacy hurdles, such as Health Insurance Portability and Accountability Act and 42 Code of Federal Regulations Part 2, mainly limit access to BH patient information. Moreover, there is limited interoperability across BH ontologies (e.g., Logical Observation Identifiers, Names, and Codes, PsyCSTETS, etc.) and the rest of medicine. The lack of access to adequate data is further limited by the disconnect in the development of electronic health records (EHRs). In 2009, the Health Information Technology for Economic and Clinical Health Act passed, providing substantial resources and incentives for physician practices and hospitals to purchase and meaningfully use EHRs. While non-physician BH professionals with mental health and substance use specialties were left out of this act, due to lateness to the market, BH professionals often purchased specialized EHRs that were less integrated for providing information concurrent with mainstream health care organizations.

Structural Challenges

The fragmented US health care system diffuses accountability across multiple players from individual healthcare professionals to clinics, hospitals, and large clinical practice organizations, to health plans, private purchasers, and public purchasers, as well as state and federal policymakers (and with additional silos across mental health substance abuse, and general medical care). Research shows that VBP plays a role in counteracting a fragmented health care delivery system via increasing care coordination in hospitals. However, even if a mechanism to share accountability was devised, coordination strategies to increase value among the many BH players would remain challenging.

A significant BH workforce shortage is widely recognized and exacerbated by a substantial proportion of BH professionals opting out of public and private insurance. The highest proportion of physicians across specialties that do not accept insurance at all are psychiatrists or Medicare (e.g., lower compensation, administrative burden). Moreover, a large component of the health care professional workforce consists of professionals in solo/small cottage industry practices with limited office infrastructure and a limited capacity or desire to respond to VBP.

Only a few VBP programs include BH professionals’ incentives. When included, BH generally represents only a small percentage of measures, and frequently limited upside/downside risk. Thus, incentives are less likely to change behavior, especially given the challenges of navigating the health care system for individuals with BH comorbidities. It is often cheaper and easier to take the penalty than to fix the problem. For example, in a study of Medicare’s Merit-Based Incentive Payment System comparing psychiatrists with other outpatient physicians, psychiatrists had significantly lower performance scores and, consequently, were more likely to be penalized.

An additional challenge is the limited availability of valid/meaningable risk adjustment for value-based care in general. It is especially a challenge with BH where clinical and social factors related to the characteristics of the population being served impact quality and costs. The necessary data to adjust for social determinants are often unavailable, and severity measures are neither systematically applied nor generally available in BH records.

There is also a risk for adjustment; while risk adjustment for quality or cost may level the playing field for safety net organizations, it can also indirectly suppress lower standards for vulnerable populations. For example, if risk-adjusting around process measures, entities such as safety-net health care professionals, could be penalized. Not because the entity did not perform well but because larger determinants to care access and success are at play, and the process measure is modeled after 1 population. Some risk adjustment may be appropriate so that these health care professionals are not unfairly penalized. Nonideal options are available when evaluating how to implement risk adjustment.

(continued in next message due to length)

Strategies Going Forward

Multiple strategies must be applied to shape BH for VBP. Designated investment and leadership are needed to develop effective quality metrics and advance the field of BH informatics, along with support for and alignment of BH information systems. Moreover, testing new models for care delivery to make interventions more integrated with incentive structures that motivate BH health care organizations are essential. To reduce fragmentation and encourage coordination, VBP should be deployed through shared accountability across the silos of health care organizations and payers. Workforce expansion and quality measurement can also be simultaneously addressed by leaning into evidence-based integration models and technologies that instantiate routine systematic clinical measurements in BH, such as Great Britain’s Improving Access to Psychological Therapies program. Additionally, equity must be directly addressed by measuring social determinants of health and prioritizing investment in measures that support marginalized populations, particularly ignored in value-based care, including people with serious mental illness.

References

1. Milad MA, Murray KF, Navathe AS, Ryan AM. Value-based payment models in the commercial insurance sector: a systematic review. Health Aff (Millwood). 2022;41(4):540-548. doi:10.1377/hlthaff.2021.00120

2. Werner RM, Emanuel EJ, Hoangmai HP. The future of value-based payment: a road map to 2030. Health Aff Blog. Accessed September 29, 2022. https://www.healthaffairs.org/do/10.1377/forefront.20220725.243398

3. Ranallo PA, Kilbourne AM, Whitley AS, Pincus HA. Behavioral health information technology: from chaos to clarity. Health Aff (Millwood). 2016;35(6):1106-1113. doi:10.1377/hlthaff.2016.0013

4. Heard RL, Carroll N, Hall A. The adoption and spread of hospital care coordination activities under value-based programs. Am J Manag Care. 2019;25(8):397-404.

5. Bishop TF, Press MJ, Keyhani S, Pincus HA. Acceptance of insurance by psychiatrists and the implications for access to mental health care. JAMA Psychiatry. 2014;71(2):176-181. doi:10.1001/jamapsychiatry.2013.2862

6. GAC J, Jayaraj Maddox KE, Bierut LJ, Johnston KJ. Comparison of performance of psychiatrists vs other outpatient physicians in the 2020 US Medicare merit-based incentive payment system. JAMA Health Forum. 2022;3(2):e220016. doi:10.1001/jamahealthforum.2022.0016

7. Goldman ML, Spaeth-Rublee B, Pincus HA. The case for severe mental illness as a disease category. Psychiatr Serv. 2018;69(6):726-728. doi:10.1176/appi.ps.201700138