Payment Rates Create Access Challenges

The Government Accountability Office recently examined whether people with behavioral health coverage could access services easily. The answer was that people may have difficulty finding in-network behavioral health care providers that accept new patients, with low care provider reimbursement rates contributing to this problem (GAO, 2022). For example, while a 1-hour initial appointment with a psychiatrist can cost over $500, Medicare would pay $216.44 and Medicaid would pay $177.44 for a similar visit (Gilberti, 2023). What this disparity leads to is individuals who need mental health or substance use disorder (SUD) services end up using out-of-network health care providers. In effect, individuals with lower incomes who cannot afford out-of-network costs may be unable to access needed care.

One study, for example, found that 17.2 percent of behavioral office visits in 2017 were to an out-of-network behavioral health care provider, compared with 3.2 percent for primary care and 4.3 percent of medical and surgical specialists (Melek et al., 2019). The same study found that the average in-network reimbursement rates for behavioral health office visits were lower than for medical and surgical office visits as a percentage of Medicare-allowed amounts. Another study, using non-self-employed clinical, counseling, and school psychologists as a proxy, found that behavioral health care providers concentrate in metropolitan areas where reimbursement is higher compared with locations associated with lower pay (Mauri et al., 2019).

Traditional Medicare Payments and Incentives

Individuals with traditional Medicare face no administrative barriers, such as prior authorization, to access care. In traditional Medicare, reimbursement for all covered services delivered by behavioral health providers is based on the Medicare physician fee schedule under Medicare Part B, which means traditional Medicare pays clinicians more for doing more. CMS constructs the fee schedules based on the Relative Value Scale for each provided procedure, such as a psychotherapy visit, as well as the clinician type. The relative value depends on the time and effort required to deliver a specific procedure, with CMS and other experts evaluating the time and effort across procedures. In the 2024 Physician Fee Schedule, the proposal to adjust the work RVUs for psychotherapy codes payable under the PFS has been finalized for implementation over a four-year transition period. In response to public comments, the adjustment has also been extended to psychotherapy codes billed alongside an E/M visit and to the HBAI codes. These changes are intended to address longstanding distortions in the valuation of time-based behavioral health services. Consequently, these adjustments represent a significant 20% increase in the fee schedule valuation for these essential behavioral health services (CMS, 2023b). Psychiatrists and psychologists, for example, receive 100 percent of the fee established under the Medicare fee schedule for procedures they supply. The resulting rate of accepting new patients in traditional Medicare is 60 percent for psychiatrists and about 50 percent for psychologists, by far the lowest for any medical specialties (Freed et al., 2023). While new patient acceptance rates are not estimated, in comparison, clinical social workers, marriage and family therapists, and behavioral health counselors are paid at roughly 60 percent of the psychologists’ rate under the Medicare fee schedule, and nurse practitioners, clinical nurse specialists, and physician assistants are paid at roughly 70 percent of the psychiatrists’ rate under the Medicare fee schedule.

Studies have shown that the process of establishing the Relative Value Scale disadvantages psychiatrists and thus all other behavioral health providers, because work values associated with psychiatric services are undervalued. For instance, the median reimbursement for psychiatrists versus nonpsychiatric medical doctors performing the same services in network was 13 to 20 percent lower for psychiatrists (Mark et al., 2018). In response, CMS has recently proposed to increase this component by 19 percent over 4 years (Moran, 2023). In addition, CMS has agreed with an additional concern that the practice cost component of the Relative Value Scale has been underestimated, requesting information to improve the methodology of estimating practice costs (CMS, 2023c).

Medicare Shared Savings Program Incentives

MSSP enrolls a relatively small share of traditional Medicare beneficiaries, and is based on accountable care organization (ACO) models. ACOs are grounded in traditional Medicare but create incentives for the ACO to manage care efficiently and reduce spending. Successful management of savings that adhere to quality standards allows the ACO to share in realized savings, though behavioral health quality metrics are limited and have relatively little weight in the assessments of quality. In addition, ACOs have a financial incentive to use clinicians affiliated with their organization, which may create some access frictions. Thus, there is little financial incentive to ensure adequate access to behavioral health services in the MSSP. To date, the degree to which behavioral health providers are part of ACOs is variable (Busch et al., 2022). Moreover, the evidence suggests there is essentially no impact on behavioral health outcomes resulting from implementing MSSP (Hockenberry et al., 2023). For example, people with depression associated with an ACO were 24 percent less likely to get treatment than those in the simple traditional Medicare program (Hockenberry et al., 2023).

Medicare Advantage Incentives and Administrative Processes

The MA program covers 52 percent of Medicare beneficiaries (Ochieng et al., 2023). MA plans receive a fixed per-beneficiary, per-month payment, creating a strong incentive to control the use of services and costs. Specifically, MA plans receive risk-adjusted capitation rates based on bids the plan submits against a benchmark payment defined by the average Part A and Part B spending by traditional Medicare in each county.

The risk adjustment mechanism represents an effort to ensure that payments adequately offset the incentives that discourage plans from enrolling of the sickest and highest-cost individuals from the eligible population. If the conditions used for risk adjustment in MA and in the Marketplace plans fail to adequately account for differences in health spending for particular health conditions, health plans would find it profitable to discourage the enrollment of people with those conditions.

At least one study determined that behavioral health conditions have inadequate risk adjustment arrangements (Montz et al., 2016). As such, health plan strategies may discourage enrollment by individuals with behavioral health conditions or limit access to treatment. For example, health plans may pay low rates to discourage behavioral health providers from participating in their networks, thereby limiting the supply of clinicians who treat behavioral health conditions. This limited supply may discourage enrollment by people likely to seek behavioral health care and may result in lower plan spending among those enrolled based on selection efforts rather than on effective care management. Another example of strategies aimed at affecting enrollment by those with mental illnesses and SUDs is through administrative processes such as prior authorization that impede care for particular types of treatment, such as intensive outpatient care. Recent evidence shows that 85 percent of MA plans impose prior authorization on psychiatric services compared with 60 percent for other physician specialty services (Freed et al., 2023). The Star Ratings bonus structure also provides substantial financial incentives to plans and is a critical vehicle for benchmark setting, capitated rates, and rebates in the form of benefits provided back to members. Box 6-1 shows the billing complexity that health care providers must navigate when diagnosing an individual, which may be a source of inadequate risk adjustment in behavioral health.

BOX 6-1

Two Separate Diagnostic Classification Systems.

Inadequate risk adjustments for behavioral health conditions are not the only challenge to increasing access to care for MA beneficiaries. Certain MA plans have profited significantly by focusing on specific clinical diagnoses and enhancing coding practices to boost their risk adjustment factors (Adler, 2018; Mandal et al., 2017). These issues are multiplied for beneficiaries with social needs, which are also captured poorly in risk adjustment and which can drive health spending. Thus, health plan incentives lead MA plans to avoid enrolling people with mental illnesses and SUDs through limiting the supply of care, which may help explain the greater supply limitations in areas with greater social needs.

Regulation might partially offset the incentive to restrict the supply of behavioral health care. For example, quality metrics in a plan that create an offsetting incentive to provide access to care, evidence-based treatment, and adequate levels of follow-up might increase the supply of behavioral health care. However, policies such as the MA Quality Bonus Program do not include payments based on any behavioral health care quality metrics. Likewise, the regulation of networks may be an important mechanism to offset the incentives to be overly restrictive of the supply of behavioral health care providers, but outcomes from network adequacy regulations have failed to improve access for behavioral health. The details of these regulations are addressed later in this chapter.

The willingness of health plans to expand behavioral health services depends in part on whether the risk adjustment prevents significant payoffs from avoiding the enrollment of people with behavioral health conditions. MedPAC recently reported evidence of significant problems with biased selection in the MA program, with MA plans enrolling healthier people who yield a payment advantage of 6 percent to 13 percent (Serna, 2023). Thus, there is a meaningful payoff to engaging in actions that affect risk selection, which historically have targeted people with mental illnesses and SUDs (Frank and Glied, 2006). Moreover, CMS has not deployed regulatory strategies that could attenuate incentives for restricting supply and therefore promote access to behavioral health care.

Medicaid Managed Care Incentives

Medicaid MCOs serve over 70 percent of Medicaid enrollees (GAO, 2024). Medicaid MCOs operate in an environment with incentives similar to those found in MA, where Medicaid MCOs receive capitated payments based on a combination of the design of each state’s Medicaid plan and negotiations with the state Medicaid agency. For behavioral health, some states carve out specialized behavioral Medicaid MCOs from a full-service Medicaid MCO to provide behavioral health services. In most states, however, full-service Medicaid MCOs provide care for the behavioral health population, so while all care is combined, the behavioral health population is targeted, with risk adjustment based on acuity.

As in the case of MA, capitation creates incentives to enroll less costly people into the plan. Since people with mental illnesses and SUDs are more costly in terms of both their behavioral health and their general medical care, plans have long taken measures to avoid such groups. In the case of states with behavioral Medicaid MCOs, the specialized and expensive behavioral health services are covered elsewhere, diminishing the incentive to avoid beneficiaries with the behavioral health conditions, though the general incentives that limit clinician availability for Medicaid MCO beneficiaries remain. For example, the median state Medicaid program paid psychiatrists 81 percent of the Medicare fee schedule for psychiatrists (Zhu et al., 2023). In Medicaid, capitation rates are based on historical FFS Medicaid spending levels that typically occurred under a benefit structure that frequently imposed strict limits on behavioral health services. Therefore, the capitation rates are inconsistent with the benefit designs that define the responsibilities of the Medicaid MCOs.

Such payment arrangements are at odds with efforts to achieve parity and broaden behavioral health coverage within Medicaid MCOs. This is in part because FFS Medicaid is not subject to parity regulations under the Mental Health Parity and Addiction Equity Act (MHPAEA). Together, these features, along with health care provider opportunities to earn more for commercial or cash-pay clients, contribute to the low participation rates of behavioral health providers. The result is that only between 30-40 percent of psychiatrists accepted new Medicaid patients in 2015 (MACPAC, 2021; Wen et al., 2019). Most states have some network adequacy requirements, but they usually count the number of in-network health care providers and time and distance in establishing standards (Hinton and Raphael, 2023b). Finally, Medicaid risk adjustment efforts vary widely across states, from minimal age and sex adjustment to the use of well-known risk adjustment systems. As with MA adjusters, these systems are quite imperfect.

Marketplace Incentives

The Marketplaces, with 100 percent of beneficiaries in private plans, are largely subject to the same incentives as MA and Medicaid MCOs. Evidence shows that the risk adjustment shows that the system used in the Marketplaces is weaker than that used in the MA program. One study from 2016 showed that the Marketplace risk adjuster for behavioral health is weak and could be improved with some simple modifications to the diagnoses used to characterize mental illnesses (Montz et al., 2016). However, far fewer data are available for these insurance plans, with a small body of evidence showing that Marketplace plans have especially narrow networks for behavioral health providers (Zhu et al., 2017). Rates of participation for psychiatrists in Marketplace plan networks were lower than for Medicare—nearly 43 percent versus 60 percent, respectively. The consistency of low network breadth for psychiatrists across MA, Medicaid MCOs, and the Marketplace suggests that higher earnings and lower administrative burden from delivering care out of network is an external challenge for constructing robust behavioral health care provider networks.

Finding: The responsibility for health care provider payment structures, health care provider networks, and assorted access arrangements that affect health care provider participation has increasingly fallen on MCOs, which cover the vast majority of the nearly 113 million Americans who participate in Medicare, Medicaid, or the individual market. Behavioral health care provider participation is affected by the incentives created by a per-member, per-month payment system to plans to pay for the care of beneficiaries enrolled in their plans. These payments are frequently risk-adjusted in an attempt to minimize the strong incentive of plans to take measures to avoid the beneficiaries who likely have the highest costs of care.

Conclusion 6-1: Insufficient risk adjustment for those with mental illnesses and SUDs contributes to MA, Medicaid MCO, and Marketplace plan strategies that limit access to behavioral health services. These strategies include creating restrictive behavioral health care provider networks and using administrative mechanisms such as prior authorization. Risk adjustment, oversight of availability of clinicians, and limits on administrative processes such as prior authorization can attenuate such behavior. Improving access to behavioral health care providers and services through managed care could occur through improvements in behavioral health risk adjustment, regulation of access to care, and thoughtful limits on prior authorization.

Marketplace Plans

Enrollees in Marketplace plans are growing at a rapid pace, with signups increasing by 30 percent from 2023 to 2024 (CMS, 2024d). With one-fifth of 2024 Marketplace enrollees having household incomes under 250 percent of federal poverty level, ensuring adequate networks to prevent the need for out-of-network care is important. A 2017 study found that mental health networks in Marketplace plans were narrow compared with primary care networks, with only 42.7 percent of psychiatrists and 19.3 percent of nonphysician behavioral health care providers participating in any Marketplace network in 2015. The study noted that Marketplace plan networks included only 11.3 percent of all behavioral health care providers practicing in a given state-level market (Zhu et al., 2017).

The CMS regulations for Marketplace plans in 2024 require that qualified health plans “must ensure a sufficient choice of care providers,” including behavioral health care providers. By 2022, 30 states had adopted at least one quantitative metric to enforce this regulation. (Giovannelli, 2022). Time and distance criteria apply in 29 states (Giovannelli, 2022; Weber, 2020), wait time standards exist in at least 15 states (Giovannelli, 2022), and some states also have criteria for health care provider–enrollee ratios (Bradley et al., 2021). As a quantitative metric, wait times for new patient appointments focus more on the patient experience of timely access to care than existing time and distance standards. Health care provider-to-enrollee ratios are easier to measure than wait times and attempt to capture a similar concept. Some states have recently adopted quantitative standards that may be more effective at capturing the number of health care providers in a specific geographic area who are willing to contract and who are available to see new patients. For instance, the New Hampshire Insurance Department determines the number of all available health care providers in a county by analyzing claims from the state’s all-payer claims database and then counting the share of available health care providers in each plan’s network. Marketplace consumers can compare qualified health plan hospital networks on the New Hampshire Insurance Department website. This method captures the number of health care providers who are contracting with an insurer, seeing patients, and submitting claims (Bradley et al., 2021; Pollitz, 2022).

In another state example, webinar 3 panelist Cara Cheevers, the behavioral health program director at the Colorado Division of Insurance, explained her agency’s approach:

(Colorado) Insurance Regulation 4-2-53 also sets standards related to network adequacy, and the idea here is really to understand how hard or how easy it is for a consumer to be able to get the care that they need. This regulation in particular pertains to behavioral health because we set standards in data around both mental health and substance use disorder treatment separately. While we understand integrated care is incredibly important, we also understand that somebody with generalized anxiety disorder has a different set of needs as someone with opioid use disorder. And while those two might combine and those two might pertain or be relevant to one consumer, we want to make sure that when we’re counting and measuring access, we’re doing so almost by condition. And that requirement also indicates that a consumer must be able to get care within 7 calendar days from the moment they attempt to make an appointment to actually being seen. So those time and distance standards in that regulation are really important.

Evaluations of the effectiveness of these various network adequacy regulation approaches are limited, but multiple years of trial and error in practice has shown that network adequacy policies have generally failed to realize their goals. The current time and distance measures used most frequently to assess network adequacy may be ineffective if they do not account for telehealth availability and its use in treating behavioral health conditions. Other measures such as appointment wait times may be more patient-centered and capture access more effectively. In addition, the effectiveness of how well network adequacy regulation works in one state might not apply in a different context.

Our network adequacy statute currently prohibits the use of telehealth for network adequacy purposes, so while we do understand the crucial role that telehealth plays in behavioral health services and medical-surgical services, insurers are actually not allowed to use telehealth as a way to meet network adequacy requirements currently. And so we are looking at whether or not we should be utilizing telehealth for network adequacy purposes, but in general we do like to see primarily in-person, person-centered care and so that’s what our statute reflects.

—Brooke Hall, webinar 3 panelist

Innovations to Improve Mental Health and Substance Use Disorder Access in Medicare, Medicaid, and Marketplace Insurance Plans

Regulatory approaches will continue to evolve. In 2023 the Department of Labor and CMS proposed new MHPAEA rules that require plans to take affirmative steps to collect, evaluate, and analyze specific types of outcome data. In this context, several state insurance departments have begun to review reimbursement parity data from insurers in their state to assess MHPAEA compliance and to assist with enforcing network adequacy standards. New Hampshire, Massachusetts, Maryland, Oregon, New Mexico, and other states are beginning to scrutinize insurers’ approaches to setting reimbursement rates for behavioral health care providers (Volk, 2023). The “data outcome” approach outlined in the proposed MHPAEA regulations may hold promise, but evaluations are needed to identify and understand best practices in this important area of regulation which is intended to ensure sufficient behavioral health care provider networks. Separate from network adequacy regulation, improving network transparency for consumers may be another approach to hold plans accountable for the adequacy of their networks.

State Medicaid and Medicaid Managed Care

Federal law requires Medicaid managed care plans to have the capacity to serve the expected enrollment in their service area and to maintain a sufficient number, mix, and geographic distribution of health care providers. These plans must make covered services accessible to their enrollees to the same extent that such services are accessible to other state residents with Medicaid who are not enrolled in a managed care plan. MHPAEA applies to Medicaid managed care and Children’s Health Insurance Program (CHIP), and about 72 percent of all Medicaid enrollees are covered by those programs (Hinton and Raphael, 2023b).

The 2020 CMS Medicaid managed care final rule removed the requirement that states use time and distance standards to ensure health care provider network adequacy and instead lets states choose any quantitative standard, such as minimum health care provider-to-enrollee ratios, maximum travel time or distance to health care providers, minimum percentage of contracting health care providers accepting new patients, maximum wait times for an appointment, or hours of operation requirements (Hinton and Raphael, 2023b). The 2024 CMS Ensuring Access to Medicaid Services Final Rule, effective as of April 2024, further strengthens network adequacy regulations through the establishment of appointment wait time standards and enforcement through secret shopper surveys and audits (Federal Register, 2023a).

There may, however, be a tension between allowing greater flexibility on network adequacy standards and increasing plan accountability for developing an adequate behavioral health network. Moreover, existing standards have not been associated with improved access to behavioral health care providers (Hu et al., 2023; Ndumele et al., 2017; Zhu et al., 2021). Given that these standards in Medicaid are relatively new and rapidly evolving, there is scant evidence on the degree to which adopting these standards increases beneficiary access to behavioral health care providers. The committee expects challenges to increasing access to behavioral health care that are similar to those described above for the Marketplace (Hinton and Raphael, 2023b; Zhu et al., 2021) and encourages assessment of these regulatory changes and their effects.

Medicare and Medicare Advantage

The traditional Medicare program, based on FFS, does not manage a health care provider network. Physicians, nonphysician practitioners, and other health care suppliers must enroll in the Medicare program to be eligible to receive Medicare payment for covered services provided to Medicare beneficiaries. CMS regulates MA plans that use health care provider networks, and therefore they are subject to federal network adequacy regulations. These regulations are quantitative. Plans must maintain a network of appropriate health care providers that is sufficient to provide adequate access to covered services to meet the needs of the population served. However, in 2022 and 2023, CMS finalized regulations that added requirements for maximum time and distance standards in rural areas and rules governing the use of telehealth. Beginning in 2023, for contract years 2024, CMS may deny an application for a new or expanding service area based on an evaluation of the applicant’s network (CMS, 2023d).

The non-interference provision in section 1854(a)(6) of the Social Security Act¹ prohibits CMS from requiring MA plans to contract with specific health care providers or require a specific reimbursement for contracted services. However, CMS can enforce the network adequacy regulations that apply to MA plans and may refuse to approve MA plans that do not meet those requirements. The MA plans themselves can decide what strategies are successful in attracting more behavioral health and SUD care providers to join their networks, allowing the plan to comply with network adequacy standards (CMS, 2023d). In 2024, CMS finalized new Medicare Advantage rules, establishing enhanced network adequacy standards for outpatient behavioral health, including diverse providers like MFTs, MHCs, OTPS, addiction medicine physicians, and other behavioral health care providers who provide addiction medicine and counseling (CMS, 2024c). New categories for LCSWs and clinical psychologists were introduced, alongside wait time standards for primary care and behavioral health services (CMS, 2023a).

Finding: Past experience demonstrates that “qualitative” network adequacy regulations do not improve access to behavioral health care.

Finding: Over the past few years, CMS and many states have adopted quantitative approaches to network adequacy regulation and guidelines, such as those based on time and distance network parameters, in efforts to strengthen networks and improve access to behavioral health care providers. Some of these metrics are inconsistent with advances in telehealth approaches to treatment. For metrics such as waiting times, which may better reflect patient experience, there is insufficient evidence to demonstrate whether they motivate plans to ensure adequate supply or offer guidance on best practices.

Conclusion 6-2: Various approaches to network adequacy regulations have not been shown to be effective in expanding behavioral health care provider participation or patient access. Nevertheless, these adequacy regulations are tools that regulators currently rely on to prevent insurers from selling health plans that are overly restrictive in the supply of behavioral health services offered. Thus, while network adequacy regulation remains a key tool for regulators, current approaches are unlikely to be the avenue for improving health care provider participation in Medicare, Medicaid, and the Marketplace. Strengthening plan accountability for providing adequate supply of behavioral health services based on outcome data would improve regulatory oversight.

Conclusion 6-3: Studies should explore the role of outcome-based approaches for expanding health care provider participation, the results of which may lead to a recommended regulatory approach.

Conclusion 6-4: Approaches to measuring access for the purposes of regulating plan networks have largely been health care provider–focused, measuring the availability of health care providers. Patient-focused measures, including ease of finding and receiving quality treatment from a culturally appropriate health care provider, are likely to require investments in new and alternative data sources, including patient surveys.

Coverage and Payment Policies that Facilitate Telehealth Uptake

A number of state and federal policies around telehealth coverage and payment have facilitated its uptake in behavioral health (CCHP, 2024b; Chu et al., 2021; McBain et al., 2023). In March 2020, during the COVID-19 public health emergency, the federal government and individual state governments provided temporary, specific telehealth flexibilities for coverage and reimbursement to meet the demand for virtual care. These temporary flexibilities included payment parity for specific telehealth clinical services such as behavioral health and expanded the definition of telehealth to include audio only (AMA, 2023). While some of these changes remain, many of the flexibilities concerning licensure are gone.

Coverage parity requires payers to cover a telehealth service if it is also covered in person and can be delivered remotely while meeting the standard of care. Payment parity requires payers to reimburse for telehealth visits at the same rate as the equivalent in-person visit. States and payer program have adopted coverage parity more widely than payment parity. In states with payment parity for commercial plans, for example, there may exist caveats such as a sunset date or only being available for specific services, such as behavioral health. The statutory language of these provisions in the various states is not uniform, which makes practicing across state lines more difficult (AMA, 2023; Augenstein and Marks Smith, 2024).

Telehealth is included within the scope of Medicaid, where it is considered a mode of service delivery rather than a distinct service under federal law (Medicaid.gov, 2024). States have broad authority to design their own Medicaid telehealth policies, including which health care provider and service types may use telehealth, the mode of telehealth delivery, such as audio only or audio–visual, and reimbursement. As a result, there is wide variation regarding telehealth coverage and payment parity, although most states are allowing telehealth services for behavioral health, home as the originating site for telehealth, and both synchronous and asynchronous services (Telehealth.HHS.gov, 2023b). As of 2023, 24 state Medicaid programs offered payment parity for telehealth, including for behavioral health services (CCHP, 2023). Figure 6-1 shows some of the variations states can have in their telehealth policies. Box 6-2 highlights the specifics of coverage for audio-only payments.

FIGURE 6-1

Medicaid telehealth flexibilities. SOURCE: Committee generated using data from CCHP, 2024a.

BOX 6-2

Audio-Only Coverage and Payment.

For Medicare, the Consolidated Appropriations Act of 2023² and the 2024 Physician Fee Schedule extended many of the COVID-19 public health emergency telehealth flexibilities through December 31, 2024. As of this writing, Medicare telehealth flexibilities include permanent coverage of both audio and visual visits for a broad swath of behavioral health services, as well as broad health care provider eligibility for telehealth billing (Telehealth.HHS.gov, 2023a). In addition, there are no geographic restrictions for the originating site for behavioral telehealth services, and federally qualified health centers, rural health centers, and rural emergency hospitals are all permanently eligible originating sites for telehealth services (Telehealth.HHS.gov, 2023c). For commercially insured populations, including by MA and Marketplace plans, 43 states and the District of Columbia have enacted legislation requiring insurers to reimburse for telehealth services. Of these states, 24 have additional requirements for payment parity.

Telehealth Uptake in Behavioral Health Care

Telehealth is one tool that can leverage existing health care provider supply to address access needs for diverse populations, including those in rural areas. Telehealth modalities have several potential advantages over in-person care for those living in areas with few available behavioral health providers, including connecting patients with specialized providers who may not be available locally; eliminating the need for and hassle of travel; offering flexible scheduling and allowing patients to receive care from the comfort of their own homes; reducing stigma; integrating with other digital tools like electronic health records to facilitate care coordination and treatment planning; and facilitating real-time interventions in a timely manner.

However, there has been differential uptake of telehealth for behavioral health services across modalities, population, and geographies. A 2020 national survey examining trends in telehealth found significant disparities among subgroups in terms of audio versus video telehealth use, with video telehealth rates lowest among those without a high school diploma, older adults, and people of color (Karimi et al., 2022). Another study found that people of color had substantially fewer telemedicine visits for behavioral health conditions than White individuals (Marcondes et al., 2024). Thus, while telehealth has had widespread and sustained uptake, less clear is whether telehealth has reduced pre-existing disparities in access. These gaps in telehealth use may be driven in part by structural issues like broadband internet access; rural areas often have significantly less access to broadband internet at FCC minimum broadband speeds than urban areas (Kohli et al., 2024). Disparities in digital literacy have also affected adoption of and access to telehealth services, as patients need digital skills to work with the technology needed to engage in telehealth.

We . . . need access to digital literacy and training and support that is funded or reimbursable because sometimes people may not have access to the technology and when they get it, they don’t know how to use it.

—Keris Myrick, webinar 1 panelist Lived Experiences in Accessing Behavioral Health Care Services through Public Insurance Programs

Telehealth uptake also has been lower in the public sector than in the private sector, with recent data suggesting that even states with the largest uptake of telehealth did not increase access to behavioral health services more than states with lower telehealth uptake (Cantor et al., 2024). Many questions also remain about the quality of telehealth care, the best applications for video versus phone-based telehealth, appropriate payment models, and the role of new asynchronous models of telehealth using, for example, smartphone apps (Bartelt et al., 2023; Samson et al., 2021), all of which are widely applicable to behavioral health.

Unknowns about Telehealth’s Effects on Clinicians

Due to its rapid and relatively recent rise, it is still unknown the extent to which telehealth modalities may help or hinder behavioral health clinicians’ ability to care for enrollees in Medicare, Medicaid, and ACA Marketplaces. There have been few studies to date that examine the net effects of telehealth delivery on behavioral health provider satisfaction, capacity, and retention.

On the one hand, telehealth may help to alleviate care burdens among clinicians, for instance, by reducing existing geographic barriers in care delivery. Proponents argue that the convenience and flexibility of telehealth can be a protective tool that can streamline clerical tasks, improve workplace autonomy, introduce greater flexibility in work schedules, reduce time constraints, and improve job satisfaction. On the other hand, while the digital delivery of care has many potential advantages for access to care, workplace flexibility, and clinician capacity, it may also increase clinician burden and burnout. At least one study has shown that health care providers with higher telehealth use have more after-hours EHR-based work, which could lead to higher levels of burnout (Lawrence et al., 2022). Telehealth may also increase depersonalization and dissatisfaction among clinicians by increasing required documentation, duplicating or slowing work processes because of interoperability, and increasing total work time through greater asynchronous patient messaging in EHR systems. One study, for example, found that clinician time spent in the EHR increased during 2020, with asynchronous messaging from patients increasing to 157 percent of the prepandemic average. Each additional patient message was associated with a more than 2-minute increase in EHR time per day (Holmgren et al., 2022).

Other workforce effects are likely to be seen over time and across markets, as telehealth delivery organizations rise in prominence. An increasing number of retail chains now offer telehealth offerings, including via behavioral health–specific companies as well as tele-mental health programs or partnerships with primary care–focused retail companies (Ashwood et al., 2017). While there are limited studies on the effects of telehealth retail companies on care access and quality, early evidence suggests that digital health and retail companies have attracted patients with lower-acuity behavioral health needs, opening new avenues for access that may be less stigmatized than traditional behavioral health delivery settings. Interim COVID-19-era flexibilities that allowed patients with opiate use disorder (OUD) to initiate or continue receiving buprenorphine treatment via audio or video telehealth have also ushered in many retail telehealth companies focused on telehealth-based OUD treatment, though there is uncertainty whether this flexibility will be extended after November 2024 (Federal Register, 2023b).

While there is optimism about retail telehealth improving access to services for some populations, there are also concerns that retail telehealth may shift resources, including the behavioral health workforce, away from community-based health settings that disproportionately care for Medicare and Medicaid populations. It is also important to consider whether retail telehealth may target less acute populations, shifting more complex patients to community health settings. Finally, there are also concerns that telehealth companies may be disconnected from traditional delivery systems, requiring individuals to re-establish care and creating care navigation challenges should more acute or complex behavioral health needs arise. In the absence of evidence about the potential effects of these alternative delivery models, more evaluation is needed as alternative behavioral health delivery systems emerge out of telehealth platforms.

Recent Innovations in Telehealth

In addition, recent innovations, including the use of asynchronous telehealth, where communication between the individual and health care provider is not live, may help with both with patient care and clinician satisfaction. For example, new digital innovations around chatbots and behavioral health apps are not reaching Medicare and Medicaid populations (Miller-Rosales et al., 2023) and are largely untested, but some of these applications may increase access to care by offering wellness and non-clinical support to people with lower-acuity needs. Technological tools may also make clinicians more efficient by offering this augmented support to more complex patients.

Asynchronous tools such as apps, virtual reality, and self-help computer programs are important to explore but are not yet ready for routine clinical payment. However, there is a need to support rigorous research to understand their true role and value in care. These assessments need to be conducted in a real-world clinical setting, with diverse patients, and by objective teams without a conflict of interest regarding the outcome. Recent reports indicate that actionable use cases are emerging but still limited and that the technology solutions are unstable (SAMHSA, 2023).

The uptake of any digital tools into care requires a concomitant focus on patients as the recipients of these tools’ actions. Tasking clinicians with digital literacy, education, and support regarding these new tools will quickly nullify any promised efficiencies. Therefore, it will be important to support a new workforce of digital navigators who help both patients and clinicians with digital literacy, digital tool selection, patient engagement, and clinical workflow in order to realize the potential of these tools (Perret et al., 2023). This new role must be considered in the costs and savings associated with technology.

Studying hybrid care or blended models of care that combine synchronous and asynchronous telehealth offers the potential to discover where actual efficiencies in telehealth lie. The industry is already actively experimenting, but the results are not public or shared on how this new model of care could best serve Medicare and Medicaid patients.

Finding: Telehealth is a feasible, acceptable, and effective modality for providing behavioral health treatment across the lifespan and for a range of disorders. The COVID-19 pandemic-associated jump in telehealth use has been unique for behavioral health, given the sustained share of visits conducted via telehealth following the pandemic.

Finding: The telehealth flexibilities introduced during the COVID-19 public health emergency, particularly the expanded coverage and payment policies for telehealth services, enabled expanded adoption of telehealth and have supported the sustained use of telehealth for behavioral health. Coverage and payment for telehealth varies across Medicare, Medicaid, and Marketplace plans, with uncertainty as to how it will change following the expiration of the extended public health emergency flexibilities.

Finding: There has been differential uptake of telehealth across modalities, populations, and geographies. Research has found that variations in broadband access and digital skills are among the most critical barriers to equitable access to telehealth services.

Finding: Audio-only telehealth increases access to behavioral health services, decreases health inequities, and expands access to underserved communities with inadequate access to broadband internet services at adequate speeds. However, it may also carry some disadvantages for the clinician–patient interactions for some behavioral health conditions.

Finding: The net effect of telehealth on the behavioral health workforce, and particularly on recruitment, retention, and insurance participation, is yet unknown. On the one hand, telehealth may help to alleviate care burdens for providers by introducing new workplace flexibilities, and on the other hand, new technologies may increase depersonalization and burnout among providers. Other new trends, including the rise of commercial telemental health companies, may have additional workforce effects that should be monitored.

Conclusion 6-5: To maintain health care equity, audio-only behavioral health and SUD telehealth services are essential for serving individuals without adequate internet video access. There is not enough evidence on the relative effectiveness of audio-only telehealth, but until the digital divide is addressed, the access to audio-only telehealth for those facing disparities in access may outweigh the uncertainly regarding its relative effectiveness compared with video telehealth for behavioral health services.

Conclusion 6-6: Telehealth is innovating rapidly with many new models coming on board with little evidence concerning the quality of care across these new modalities. This uncertainly makes it unclear whether future modalities within existing regulatory and payment frameworks will be effective in promoting health care provider access in Medicare, Medicaid, and Marketplace plans. Developing agile and flexible payment and regulatory structures may be needed. For example, hybrid care models that blend synchronous and asynchronous telehealth may increase access to care, but developing best practices and regulations to protect consumers and ensure integrity of clinical services would be necessary. In addition, payment for these models must balance access with the potential for overuse of low-value care. It is important to explore new regulatory pathways for novel asynchronous telehealth tools that can quickly assess value, build public trust, and increase transparency.

Conclusion 6-7: To improve access to behavioral health care amidst broadband gaps, targeted efforts should identify regions needing both services and broadband. Collaborating with federal agencies such as the Department of Commerce, Treasury, Agriculture, and the Federal Communications Commission can strategically allocate broadband funds. Effective distribution of these resources to underserved areas is crucial for enhancing connectivity and equitable access to essential behavioral health services nationwide.

Quality Measurement Infrastructure to Support Value-Based Payment and Quality Improvement

A measurement of performance linked to a system’s consequences for strong and weak performance lies at the core of accountability systems. Quality measurement has become widely established in medical care and increasingly includes more behavioral health measures. A broad range of state and federal programs include quality measures, with about a third of Medicare core quality measures specific to behavioral health (CMS, 2024a,b). These measures can facilitate quality improvement by tracking trends over time and providing a basis for comparison across health care providers. However, the effort required to collect and report these measures when participating in federal programs may reduce health care provider participation. The National Committee for Quality Assurance (NCQA) reports that federal programs use over 100 behavioral health performance measures, yet only four are commonly used, with most measures used by only one federal program. This puts an undue burden on health care providers participating in federal programs to adhere to the varied measurement requirements (NCQA, 2021).

Tying measures to payment is critical if these measures are to ultimately play a role in improving the quality of care. The mantra of “what gets measured, gets done” is particularly true when those measures are tied to payment, yet few, if any, behavioral health quality measures are tied to payment. For payment models to produce true benefit to society, CMS needs to ensure that payment incentivizes “value” in ways that drive behavioral health access and the outcomes that are important to the people who experience illness. They must also carry sufficient revenue effects to reward health care providers that excel and penalize those that provide sub-par care.

In considering value-based payment and alternative payment models as an opportunity for addressing health care provider participation and meaningful beneficiary access in Medicare, Medicaid, and the Marketplace, the committee considers the following opportunities and challenges for quality measurement in behavioral health:

BHI and Care Provider Participation in Public Insurance Programs

BHI creates the opportunity for enhanced collaboration and communication that can promote the adoption of integrated care practices by primary care providers, knowing they have support from behavioral health specialists (Rybak et al., 2023). Some integrated care models offer opportunities for enhanced reimbursements, usually involving value-based payment arrangements that encourage investing time and resources in delivering comprehensive care to patients. BHI can enhance professional satisfaction for care providers’ holistic health needs and allow them to see the effects of their care more directly. By sharing the workload with behavioral health professionals, primary care physicians may experience reduced burnout as well, making them more likely to participate in these kinds of models.

Finding: A broad collection of bi-directional integrated care models have been tested and found to be cost-effective in care for some behavioral health conditions. These models can reduce treatment times and improve patient outcomes compared with usual care.

Finding: The biggest barriers to integrated care are organizational and cultural, including restrictions on licensing and coding.

Finding: Although successful models of integrated care exist, the development of payment, coverage, and infrastructure investments that promote the widespread adoption of such models in Medicare, Medicaid, and the Marketplace environments remains elusive.

Persons with Behavioral Health Conditions and Complex Social Risk Factors

For individuals with serious mental illnesses, dual diagnoses, or intricate behavioral health issues, addressing social and economic factors alongside health behaviors is critical in determining overall health. For instance, research has shown that individuals with higher levels of education and social support experience better health outcomes (Magnan, 2017). Consequently, addressing these wide-ranging drivers of health requires involving professionals with diverse expertise, particularly those skilled in engagement strategies. Medicaid has recognized the new opportunities for states to address social drivers, and CMS recently issued guidance aimed at expanding state capabilities to use Medicaid in tackling issues such as housing and food insecurity (HHS, 2023).

One innovative pathway that CMS outlined uses managed care to address social drivers of health. In early 2023, CMS introduced guidance that enables states to permit Medicaid managed care plans to offer “in lieu of” services and “value-added benefits,” such as housing, transportation, and nutrition supports, as alternatives to standard Medicaid benefits (Hinton and Diana, 2024). These services must be medically appropriate, cost-effective, and voluntary for both the Medicaid MCOs to offer and the beneficiaries to receive. Ensuring the supply of the “right services” to people with severe illnesses and complex needs is part and parcel of promoting traditional clinician participation in Medicare, Medicaid, and the Marketplaces.

In the longer term, adopting value-based payment models or alternative payment models may present an important opportunity for the flexibility to address the range of needs of people with mental illnesses and SUDs. These models encourage a shift from a rigid, prescriptive approach to a more dynamic, outcome-focused system that would enable health care providers to develop and implement care models that are tailored to the unique needs of beneficiaries (Johnson and Rittenhouse, 2023). However, as discussed above, the success of alternative payment models and value-based payment models hinges on implementing robust performance measures. In particular, it will be important for these measures to capture the degree to which health plans and care providers are addressing these social needs by measuring social outcomes and improving beneficiaries’ lived experiences in terms of quality of life, loneliness, and related patient-reported outcomes.

Effectively addressing the social factors that influence care and improve health outcomes among those with complex behavioral health needs will require new approaches that expand the set of health care providers supporting behavioral health needs. For example, community health workers have emerged as a critical bridge between health care systems and marginalized communities, demonstrating their value in improving population health and reducing health disparities (Phillips et al., 2023). Their roles, which vary from case managers to health navigators, were particularly highlighted during the COVID-19 pandemic, where they played a vital role in facilitating access to testing, vaccination, and treatment for marginalized groups.

Persons with Co-Occurring Mental Health Issues and SUD

Co-occurrence of a mental illness and SUD is prevalent among those presenting with either condition. In 2021, over one-third of adults aged 18 or older who had any mental illness also had an SUD in the past year. Approximately 20 percent of adolescents aged 12 to 17 had a major depressive episode in the past year and 20 percent of adolescents with a major depressive episode had a co-occurring SUD (SAMHSA, 2022). Around 25 percent of individuals with a serious mental illness also had an SUD (SAMHSA, 2024). Treating two or more co-occurring conditions effectively requires coordination, collaboration, and integrated treatment by multiple health professionals. Multiple barriers exist to developing an adequate network for integrated treatment for co-occurring mental illness and SUD.

The lack of specialized services and integrated treatment settings for adults and youth, including residential or rehabilitation programs and intensive inpatient care, presents a barrier to integration (Priester et al., 2016). Only slightly over half of U.S. substance abuse treatment facilities report offering programs for clients with co-occurring mental illness and SUD (SAMHSA, 2020).

Another barrier cited by both mental health and SUD clinicians is the need for additional training and for staff that specialize in co-occurring disorders within both systems, as well as staffing and technology to facilitate coordination among health care providers (Priester et al., 2016). There are limited staff and faculty with expertise in integrated treatment, increasing the difficulty of improving access through education and training (Yule, 2019).

Coverage and reimbursement are identified repeatedly as barriers to integrating behavioral health and SUD treatment (Yule, 2019). Inconsistent coverage of services in Medicare, Medicaid, and Marketplace plans creates a checkerboard of service availability for those with co-occurring disorders. Services may not be covered, may be covered through FFS, or may be covered through an alternative payment mechanism (Priester et al., 2016). In some states, payment for integrated treatment delivery is limited by diagnostic and billing criteria that do not recognize the need for treatment of two or more disorders, the clinical complexity added when a co-occurring condition is present, or the additional time and staffing needed to coordinate care among practitioners. Reimbursement inequities for each type of care can also disincentivize integrated care. Historically, insurance benefits for behavioral health treatment have been greater than the benefits for substance use treatment (Yule, 2019).

Statewide efforts to implement integrated behavioral health, SUD, and physical health with Medicaid as the cornerstone payer are showing results but require an intentional focus to address multiple barriers. Washington State’s multi-year plan and implementation effort is an example of a systematic effort to identify and address barriers in order to ensure an adequate network for integrated behavioral health and physical care (CHCS, 2020).

Box 6-4 illustrates the complexity of treating behavioral health conditions for specialized populations by focusing on the particular case of women who experience a behavioral health issue during or after pregnancy.

BOX 6-4

Pregnancy and Postpartum Care.