Introduction

The “California Advancing and Innovating Medi-Cal” (CalAIM) initiative was recently approved by the Centers for Medicare and Medicaid Services (CMS) and launched in January 2022. CalAIM seeks to improve health outcomes and advance health equity for “Medi-Cal” (Medicaid) beneficiaries and other low-income people in the state. As of September 2021, over 13 million people were enrolled in Medi-Cal in California – the largest Medicaid program in the nation. CalAIM will leverage Medicaid as a tool to help address complex challenges, including behavioral health access, the health needs of people experiencing homelessness, and the complex clinical needs of justice-involved populations. Central to the demonstration is a person-centered approach, first authorized as “Whole Person Care” (WPC) pilots implemented through a waiver approved by the federal government under section 1115 of the Medicaid statute (known as “Medi-Cal 2020”). The WPC pilot program aimed to coordinate and integrate physical health, behavioral health, and social services for Medicaid enrollees with complex needs. Under CalAIM, the state will incorporate and transition WPC pilot services statewide through the state’s Medicaid managed care system. The California initiatives are part of a growing movement to better meet behavioral health needs and improve equity and health by addressing underlying social determinants.

As CalAIM launches, it is useful to take a closer look at the WPC pilot program and what has been learned to date from its implementation. This brief draws heavily on findings from the interim evaluation, which assessed WPC infrastructure development and implementation and analyzed whether WPC led to “better care and better health” within the first three years. In this brief, we explore key questions involving the WPC pilots, which may provide insights for other states pursuing or contemplating similar strategies for identifying and addressing the medical and non-medical or social needs of high-risk, high-utilizing Medicaid enrollees.

What Were Whole Person Care Pilots (WPC)?

In 2016, California began its “Whole Person Care” (WPC) pilot program, which aimed to coordinate and integrate physical health, behavioral health, and social services for Medicaid enrollees with complex needs. The WPC pilot program also aimed to increase integration and data sharing among county agencies, health plans, and community-based organizations. The pilot was authorized under the state’s Section 1115 waiver. California’s Department of Health Care Services (DHCS) selected 25 WPC pilots (spanning 26 counties) to participate in the program across the state.¹  (See Appendix for a spotlight on San Francisco County’s WPC pilot.) Each individual WPC pilot program defined its target population and interventions, within state guidelines. The WPC pilots were scheduled to end in December 2020 but were extended through December 2021 under the state’s Section 1115 waiver.

Each WPC pilot included a “lead entity” and key community partners (called “participating entities”). Lead entities coordinated the pilot, collaborated with local community partners, and served as the single point of contact for DHCS. Lead entities included county health and health services agencies, healthcare systems, behavioral health departments, and a city municipality.²  Each WPC pilot included key community partners, including Medi-Cal managed care plans, county health agencies, other public agencies (e.g., county alcohol and substance use disorder programs, human services agencies, public health departments, criminal justice/probation entities, and housing authorities), and community-based providers.³  In their applications, approved WPC pilots enumerated specific roles of each participating entity in service provision, data management, and implementation of WPC programs. In the initial program years, WPC pilots focused heavily on developing data and administrative infrastructure for data sharing, case management, outreach and enrollment, performance, and quality improvement efforts.

WPC pilots targeted a number of high-risk populations with complex needs. Primary WPC target populations included high utilizers of health care (e.g., those with avoidable use of the emergency room and/or hospital admission), individuals experiencing homelessness or at-risk of homelessness, individuals with multiple chronic conditions, individuals with severe mental illness or substance use disorders (SMI/SUD), and justice-involved individuals. Nearly half (46%) of all pilot enrollees were homeless across target populations, and most were men, 50-64 years old, White, English speaking, and enrolled in managed care. Pilots classified 10% of enrollees as at-risk of homelessness and 6% as justice-involved. Beneficiaries had high rates of hypertension, substance use disorders, and mental health conditions, as well as high rates of service use (particularly SUD services and emergency department visits). Participation in the pilot program was voluntary for eligible individuals. Many WPC pilots used street or shelter-based outreach to identify eligible enrollees.⁴  Other strategies used to identify pilot enrollees included referrals from health care facilities, health plan eligibility lists, and predictive modeling based on program criteria.

WPC pilot beneficiaries received care coordination and other services not covered by Medi-Cal to address medical, behavioral health, and social needs. All pilots were required to provide care coordination services, but pilots also offered a variety of other services tailored to local needs and available resources. Services offered by WPC pilots included housing-related services, employment assistance, short-term residential care for individuals experiencing homelessness (referred to as “medical respite”), and sobering centers. Given the high need for housing support, pilots commonly provided housing-related services, including housing navigation (to help identify and secure affordable housing), tenancy supports (e.g., security deposit and first month’s rent), as well as landlord incentives and coaching. Other WPC service examples included providing linkages to health, mental health, and substance use disorder services through medical respite, peer support specialists, mobile teams, or service integration teams.

The state’s Section 1115 waiver authorized up to $3 billion in federal and state Medicaid funds over five years for the WPC pilot. For each demonstration year, $300 million was made available in federal financial participation to fund the WPC pilots. The non-federal share was financed through local funds from participating government agencies (also referred to as “intergovernmental transfers” or IGTs). Each WPC pilot had an annual budget and could draw down funds for eligible activities including infrastructure development, data collection, and services/interventions as well as other outcome-based incentive payments. WPC pilots reimbursed for services on a fee-for-service (FFS) (i.e., per service or encounter) basis or using a per-member-per-month (PMPM) bundle.⁵  In the early stages, pilots were reimbursed for application submission and reporting baseline data, while subsequent years were oriented toward providing services, achieving proposed metrics, and providing incentive payments.

Pilots were required to submit mid-year and annual reports detailing spending and outcome metrics, including NCQA and HEDIS measures as well as county and population-specific measures. The state is also subject to independent evaluation requirements under its Section 1115 waiver. Findings from the state’s interim independent evaluation of the WPC program are discussed in more detail in the section that follows.

What Can We Learn from WPC?

A variety of strategies resulted in the enrollment of more than more than 108,000 unique individuals into WPC between January 2017 and December 2018 (PY 2 and PY 3).⁶  To identify and enroll eligible individuals, WPC pilots used administrative and electronic medical record data, referrals from partners (managed care plans, health care providers, and clinics), and warm hand-offs from social services agencies. To facilitate referrals into the program, some WPC pilots developed formal contracts with community partners. Pilots also used street- and shelter-based outreach to identify people eligible for but not enrolled in Medi-Cal. To identify the appropriate interventions for enrollees, some pilots developed universal risk assessment tools (used across partner agencies) to stratify client acuity (or level of need) across multiple domains (e.g., health and housing status). To address eligibility gaps in Medi-Cal (also referred to as “churn”), WPC pilots proactively monitored eligibility and renewal dates.⁷  Nearly all pilots identified low beneficiary engagement – or willingness to enroll and participate in services – as a barrier and stressed the importance of using a “relationship-based” approach to engage medically and socially complex prospective enrollees, as well as a variety of other strategies to foster engagement (e.g., connecting enrollees to immediate benefits, hiring WPC staff with shared lived experience to build trust).⁸ 

Data sharing presented challenges for the vast majority of WPC pilots, although capacity increased over time. The majority of pilots developed electronic platforms to facilitate and track care coordination activities (e.g., care plans, encounter notes, and service referrals). Multiple pilots reported participating in health information exchanges and reported increased utilization of real-time data platforms across lead entities and partner agencies over time. Challenges – including duplicative data entry and managing multiple disparate data systems across settings – persisted across most pilots. Other data sharing challenges included concerns around legal risks of data sharing and differing interpretations of laws and regulations involving patient privacy and confidentiality. To address data sharing concerns, some pilots developed universal consent forms to facilitate the release of information across WPC partner organizations, while others developed consent forms to allow for the targeted release of specific data, allowing enrollees to opt out of sharing certain information (e.g., mental health history and HIV test results).⁹ 

Pilots reported success with introducing new care coordination processes and using data systems/sharing to support care coordination activities. Care coordination services were frequently provided by non-clinical staff (under supervision of clinical staff) including community health workers (CHWs), and most pilots reported employing care coordinators with shared lived experience. Many pilots reported using a single dedicated care coordinator who followed enrollees across all WPC-participating care settings. Nearly all pilots reported the most common mode of communication between care coordinators and enrollees was in-person. All pilots reported using “active” referral strategies, including helping enrollees schedule appointments, accompanying enrollees to appointments, assisting with transportation, and following up with enrollees after appointments. Challenges common to care coordination included limited availability and/or accessibility of services for referrals. This was especially true for housing needs, where barriers included long wait times for permanent housing, limited housing options available, and poor housing quality.

WPC pilots demonstrated improvements in select health utilization and outcomes, care coordination, and delivery of housing services. The interim evaluation used pilot-reported metrics and Medi-Cal data to determine whether WPC led to better care and better health in the first three years. Analyses included comparison of selected Medi-Cal metrics before and after WPC implementation for WPC enrollees and a control group of Medi-Cal enrollees with similar characteristics. Interim evaluation data showed WPC pilots improved rates of follow-up after hospitalization for mental illness, initiation and engagement in alcohol and drug dependence treatment, and timely provision of comprehensive care plans and suicide risk assessments. The interim evaluation found that emergency department (ED) visit rates from WPC Pilot Year 1 to Pilot Year 2 decreased by 19% for WPC enrollees and 8% for the control group, a significantly larger decrease for the WPC enrollees, although analysis of pre-WPC enrollment compared to post-enrollment ED visit rates found no difference for both groups.¹⁰  While adjusted trends of all-cause readmissions did not show a reduction in either group, rates of all-cause readmissions declined from year 1 to year 2 (this decline was greater among WPC enrollees compared to the control group). The interim evaluation also found improvement in health outcomes metrics, including self-reported overall and emotional health and diabetes and blood pressure control. Nearly half of all pilot enrollees were homeless across target populations. The interim evaluation showed success in delivery of housing services and in the receipt of supportive housing, but challenges related to adequate housing supply and direct funding to provide housing. Recent analyses found the COVID-19 pandemic impacted WPC pilots in a variety of ways, but pilots were able to capitalize on existing WPC infrastructure and leverage/coordinate with other COVID-19 response activities. The final evaluation will assess the role of WPC in reducing costs for pilot enrollees and the Medi-Cal program overall, whether care coordination and partnerships were sustained beyond the demonstration, and any differences in interventions/services by target population and pilot site to help identify which services may improve outcomes.

What is Next for WPC and CalAIM?

Under CalAIM, the state will move beyond county pilots to standardize and incorporate services offered by WPC statewide through Medi-Cal managed care plans. CalAIM, which launched in January 2022, seeks to take the state’s whole person care approach statewide, with a central focus on improving health and reducing health disparities and inequities.¹¹  CalAIM adds new programs and reforms existing programs.¹²  The CalAIM demonstration and its various components are authorized under Section 1115, Section 1915(b), and through state plan amendments. The three stated goals of CalAIM¹³  are as follows:

• Identify and manage member risk and need through whole person care approaches and addressing SDOH.
• Improve quality outcomes, reduce health disparities, and drive delivery system transformation and innovation through value-based initiatives, modernization of systems, and payment reform.
• Move Medi-Cal to a more consistent and seamless system by reducing complexity and increasing flexibility.

Under CalAIM, Medi-Cal managed care plans will provide Enhanced Care Management (ECM) and Community Supports to targeted high-need beneficiaries. These services will be provided through contracts with community-based providers including organizations that participated in WPC.¹⁴  With the launch of ECM and Community Supports, DHCS will sunset the WPC pilots and the Health Homes program.¹⁵  Although the state has indicated many WPC enrollees will automatically transition to receiving ECM and Community Supports (many of which the state says may be equivalent to services received under WPC), some WPC enrollees will not transition to receiving these services, and some individuals may lose access to certain services they received under WPC.¹⁶  ECM and Community Supports (described in detail in the box below) will only be available to Medicaid beneficiaries enrolled in managed care plans, and managed care plans will have discretion over which Community Supports to cover.

Under Section 1115, CMS authorized $1.44 billion in one-time funding to assist with transitioning the WPC pilots and Health Homes programs from Medi-Cal 2020 to CalAIM. “Providing Access and Transforming Health” (PATH) will provide transitional funding to support continuity of services and maintain and support provider and community-based organization capacity. One goal of PATH will be to help ensure Medi-Cal enrollees have continuous access to benefits and services provided previously through WPC pilots as these services and activities are integrated into managed care plans. The funding may not be used to initiate new services but is supposed to ensure continuity for services that correspond to Community Supports managed care plan choose to offer.²²  A portion of PATH funds will be used to support justice-involved individuals by sustaining pre- and post-release services provided through WPC pilots and by supporting certain pre-release application planning and IT investments for justice-involved activities.

In addition to incorporating services offered by WPC pilots statewide, the state is planning to undertake a host of other behavioral health reforms under CalAIM. For example, the state plans to integrate the administration and financing of specialty mental health and SUD services under a single state/county contract. The goal is to improve outcomes for enrollees through more coordinated treatment and to allow providers to respond in a more patient-centered way. The state also plans to improve coordination of mental health care between Medi-Cal managed care plans (managing non-specialty mental health services) and county-operated mental health (managing specialty mental health services and SUD treatment). Distinct from these efforts, CalAIM also includes a plan to test the full integration of physical health, behavioral health, and oral health under one contracted managed care entity.

Looking Ahead

As mental illnesses and substance use disorder have worsened during the COVID-19 pandemic, many states are building on and developing initiatives to address behavioral health issues in Medicaid. States are expanding benefits such as community-based mobile crisis intervention services and screening, expanding the use of telehealth for behavioral health, and using managed care organizations to deliver services. In addition to California, Colorado, Massachusetts, Montana, and Washington are advancing broad behavioral health initiatives spanning many areas. Key lessons from California to date show that such initiatives require integrated data systems, adequate funding streams, and multi-sector provider partnerships to coordinate services. Going forward, California’s experience in trying to scale the WPC pilots statewide and use managed care authority to broadly deliver a range of additional services will be important to understand and learn from as other states work to improve behavioral health services and address social determinants to enhance health and equity.

Appendix

1. Two pilots (Small County and Solano County) discontinued their WPC programs as of 12/31/2020. ↩︎
2. Organizations eligible to apply as lead agencies as defined by the state included county or city agencies, certain designated public hospitals, federally recognized tribes or tribal health programs, or a consortium of these entities. ↩︎
3. Letter of extension approval for California Medi-Cal 2020 Demonstration. December 2020. Centers for Medicare and Medicaid Expenditure Authority. Accessed via: https://www.medicaid.gov/Medicaid-CHIP-Program-Information/By-Topics/Waivers/1115/downloads/ca/ca-medi-cal-2020-ca.pdf ↩︎
4. Pourat P, O’Masta, B, Haley LA, and Chuang E. A Snapshot of California’s Whole Person Care Pilot Program: Implementation Strategies and Enrollees. May 2021. UCLA Center for Health Policy Research. Accessed via: https://healthpolicy.ucla.edu/publications/Documents/PDF/2021/wholepersoncare-policybrief-may2021.pdf ↩︎
5. For example, FFS payments might be made for discrete services including mobile clinic visits, housing transition services, medical respite, transportation, or care coordination. PMPM bundled payments might be made for one or more services and/or activities including comprehensive case management, housing support services, mobile outreach and engagement, or long-term care diversion. DHCS may allow pilots to rollover unspent funds from one year to the next. ↩︎
6. Pourat N, Chuang E, Chen X, O’Masta B, Haley LA, Lu C, Huynh MP, Albertson E, and Huerta DM. Interim Evaluation of California’s Whole Person Care (WPC) Program. Los Angeles, CA: UCLA Center for Health Policy Research, September 2019. Accessed via: https://healthpolicy.ucla.edu/publications/Documents/PDF/2020/wholepersoncare-report-jan2020.pdf ↩︎
7. Ibid 6. ↩︎
8. Chuang E, Pourat N, Haley LA, O'Masta B, Albertson E, and Lu C. Integrating Health And Human Services In California's Whole Person Care Medicaid 1115 Waiver Demonstration. Health Aff (Millwood). 2020 Apr;39(4):639-648. doi: 10.1377/hlthaff.2019.01617. PMID: 32250689. Accessed via: https://www.healthaffairs.org/doi/10.1377/hlthaff.2019.01617 ↩︎
9. Ibid 8. ↩︎
10. The interim evaluation used difference-in-difference methodology to compare WPC enrollees to the Medi-Cal enrollee control group between a baseline period (pre-WPC enrollment) and an enrollment period (including data from one year after WPC Year 1 and two years after WPC enrollment). The analysis found no significant change for either group for adjusted trends in emergency department (ED) visit rates and a significant increase in hospitalizations for WPC enrollees compared to the control group. However, the analysis found a decrease in ED and hospitalization rates for both groups from the first to the second years of WPC enrollment. Emergency department (ED) visits, hospitalizations, and all-cause readmission for WPC enrollees were rising prior to WPC enrollment and continued to increase in the first year of pilot enrollment (reflecting successful enrollment of high utilizers into the pilot), before declining overall in the second program year. ↩︎
11. CalAIM builds on existing infrastructure for the 25 WPC pilots, and counties that did not already have WPC pilots in place were able to apply for additional funding in 2020 to initiate WPC-like pilot capacity. All of the 33 counties that applied for the $20 million one-time funding were approved. Accessed via: https://www.dhcs.ca.gov/services/Documents/MCQMD/WPC_Documents/20M-One-Time-Funding-10-18-19.pdf ↩︎
12. CalAIM Explained: Overview of New Programs and Key Changes. October 2021. California Health Care Foundation. Accessed via: https://www.chcf.org/publication/calaim-explained-overview-new-programs-key-changes/ ↩︎
13. CalAIM: Our Journey to a Healthier California for All. Department of Health Care Services. Accessed via: https://www.dhcs.ca.gov/CalAIM/Pages/calaim.aspx ↩︎
14. Ibid 4. ↩︎
15. In 2018, DHCS launched the Health Homes Program (HHP). The HHP serves eligible Medi-Cal managed care plan Members with complex medical needs and chronic conditions who may benefit from intensive care management and coordination. Health plans are required to automatically authorize all HHP Members for ECM and reassess within six months to determine if they still benefit from ECM-level care coordination. Health plans must also appropriately transition eligible HHP Members to applicable Community Supports that the health plan elects to offer. Accessed via: https://www.dhcs.ca.gov/Documents/MCQMD/WPC-and-HHP-Transition-Public-Webinar-9-21-21.pdf ↩︎
16. Ibid 15. ↩︎
17. Ibid 16. ↩︎
18. California Advancing and Innovating Medi-Cal (CalAIM) Enhanced Care Management. Department of Health Care Services. Accessed via: https://www.dhcs.ca.gov/CalAIM/Documents/CalAIM-ECM-a11y.pdf ↩︎
19. “In-lieu-of” services are a substitute for covered services and may qualify as a covered service for the purposes of capitation rate setting, as long as the state determines in-lieu-of services are medically appropriate and cost effective. Enrollees are not required to use the alternative service or setting but may choose to do so. Approved in-lieu-of services are authorized and identified in MCO contracts. Costs associated with in-lieu-of services may be included in the numerator of the medical loss ratio (MLR) and are taken into account when developing capitation rates. ↩︎
20. Ibid 18. ↩︎
21. The Section 1115 demonstration provides separate authority for short-term post hospitalization housing and recuperative care services (delivered by managed care plans). ↩︎
22.  Funding will be available where there is a current commitment from the managed care plan that it will elect to offer the service by January 1, 2024. ↩︎

The COVID-19 pandemic has coincided with worsening mental health across the country, and California is no exception. For this data note, we analyzed data from the California Health Interview Survey (CHIS), which asks noninstitutionalized adults detailed questions on mental health and substance use, to further examine the impact of the pandemic on mental health in the state.

The pandemic comes on the heels of an already worsening mental health and substance use crisis in California. As shown in Figure 1, a growing share of nonelderly adults have reported past year serious psychological distress over time in California, up from 9% in 2016 to 15% in 2020. Substance use issues have also worsened – drug overdose deaths increased over 40% from 2019 to 2020 in California.

During the pandemic, serious psychological distress was more pronounced among some communities, including women and low-income adults in California. As shown in Figure 2, women in California (16%) were more likely than men (13%) to experience serious psychological distress in 2020. In general, women have more often reported poor mental health compared to their male counterparts both before and during the pandemic. In California, lower-income adults were more likely to experience serious psychological distress (19% for incomes between 0 and 138% FPL; 17% for incomes between 139 and 249% FPL; and 15% for incomes between 250 and 399% FPL) than high-income adults (11%) in 2020 (Figure 2). A recent survey of nonelderly adult Californians who saw a health care provider in the past year also found that low-income individuals were more likely than higher-income individuals to report worsening mental health during the pandemic.¹  Separately, compared to White adults in California (15%), some communities of color – including Asian (12%) and Black (11%) adults – were less likely to report experiencing serious psychological distress in 2020 (Figure 2). Serious psychological distress among Hispanic adults (15%) was similar to White adults (Figure 2). Other KFF analyses found that, nationally, mental health and substance use concerns have been exacerbated among many people of color throughout the pandemic. Research has also shown that stigma – including stereotypes and discrimination – around mental health concerns is more pronounced among communities of color, perhaps affecting self-reporting of mental health and access to care. Additionally, a survey of adults in California found that among nonelderly individuals with mental health needs, people of color were less likely than White people to receive care.

In 2020, poor mental health outcomes among nonelderly adults varied greatly across counties in California. As shown in the map below, the share of nonelderly adults with past year serious psychological distress ranged from 9% to 28% across counties in 2020; highest in some Central Valley counties (Colusa/Glenn/Tehama, 28.3%; Butte, 26.9%; Merced, 24.2%) and lowest in Bay Area counties (San Francisco, 9.1%; Marin, 9.5%; Contra Costa, 9.6%). The share of nonelderly adults who ever had thoughts of suicide ranged from 8% to 34% across counties in 2020; highest in northern counties (Del Norte/Lassen/Modoc/Plumas/Sierra/Siskiyou, 34.3%; Butte, 29.7%; Mendocino, 29.2%) and lowest in some southern and eastern counties (El Dorado, 8.1%; Imperial, 8.8%; Placer, 9.3%).

Nearly nine in ten nonelderly adults in California have insurance coverage to help them access and cover the cost of health care, including mental health services. In recent years and in response to the pandemic, the state has introduced several policy measures to increase access to mental health and substance use services; many of these efforts leverage insurance coverage through Medi-Cal (the state’s Medicaid program), which covers 20% of nonelderly adults, or private coverage, which covers 67%.

Across insurance types, large shares of nonelderly adults reported poor mental health outcomes in 2020. Medi-Cal enrollees were significantly more likely than private insurance enrollees to experience past year serious psychological distress in 2020 (18% vs. 11%, respectively) (Figure 4), perhaps reflecting Medi-Cal coverage of many people with disabling health conditions including serious mental illness and/or many low-income individuals who may be more likely to report poor mental health. Many nonelderly adults also reported thoughts of suicide in 2020: 14% among Medi-Cal enrollees and 13% among private insurance enrollees (data was not statistically different).

Many insured adults in California reported barriers to care for mental health and substance use services in 2020. Twenty four percent of nonelderly adults in California reported needing help for mental health or substance use issues in the past year (Table 1).²  Among nonelderly adults with mental health or substance use needs, the share who received treatment in the past year was similar across plan type (60% of Medi-Cal enrollees and 55% of private insurance enrollees) (Table 1). Medi-Cal enrollees who received treatment reported a higher average number of visits to a professional in the past year compared to their privately insured counterparts (18 visits vs. 11 visits).

Many Medi-Cal (40%) and private insurance (45%) enrollees with mental health or substance use needs did not receive treatment. Large shares of insured adults who did not receive treatment cited concerns about costs (54% of Medi-Cal enrollees and 49% of private insurance enrollees) and difficulties scheduling an appointment (30% of Medi-Cal enrollees and 20% of private insurance enrollees) as barriers. With the pandemic and accompanying economic recession, many adults have faced financial hardship and have had difficulty paying for usual expenses. This may be linked to the high shares of adults citing concerns over costs as a barrier for accessing mental health or substance use treatment – even for Medi-Cal enrollees, who have very low or no out-of-pocket costs for these services. Scheduling difficulties as a barrier to accessing mental health or substance use services may reflect an insufficient availability of providers.

 The pandemic highlights issues with accessing mental health and substance use care across California. California continues to face shortages of mental health providers. In light of pandemic-related precautions and restrictions, accessing services for health care, including mental health or substance use care, has been challenging nationally. In response, measures, including the temporary public health emergency, have allowed many providers to offer care through telehealth. Across the state of California, access to mental health care can also vary greatly. Specifically, for Medi-Cal enrollees, access can vary widely from county to county and navigating care can be particularly challenging for enrollees with co-occurring serious mental health and physical health needs.

In response to mental health and substance use needs in California during the pandemic, several policy actions were taken to increase and standardize access to care. Senate Bill 855 addresses gaps in existing mental health parity laws by requiring insurers to expand coverage to a broader range of mental health and substance use disorders (apart from Medi-Cal). Assembly Bill 2265 will allow for funds from the Mental Health Services Act, typically reserved for mental health services only, to be used for substance use services as well among people with co-occurring mental health and substance use issues. Additionally, through Senate Bill 803, Medi-Cal mental health and substance use services will expand to include peer support specialists. In California’s recent state budget proposal, funds were allocated for dyadic care – allowing for behavioral health services for parents to be incorporated into pediatric health care visits among Medi-Cal beneficiaries. California’s renewal applications for Section 1115 and 1915(b) waivers, known as California Advancing and Innovating Medi-Cal (CalAIM), were also recently approved. CalAIM builds on the state’s Medi-Cal 2020 Whole Person Care pilot program to coordinate and integrate physical health, behavioral health, and social services for Medicaid enrollees with complex needs. In addition to transitioning these services to the state’s managed care system, CalAIM includes behavioral health initiatives to address the division of services between county-level mental health plans and managed care plans, and to standardize access to services for enrollees with serious mental health needs. Private insurers have also taken steps to address mental health and substance use care. Nationally, some plans expanded coverage to include mental health or substance use services accessed through telehealth and many employer sponsored plans made changes such as waiving or reducing cost-sharing and increasing coverage for out-of-network services. Separately, California has taken steps to fund the statewide launch of 988, a mental health crisis hotline.

Looking ahead, further research on access to care and utilization of mental health services will be informative as state policymakers, administrators, and health care providers continue to discuss how to address mental health and substance use needs among Californians.

1. This survey from California Health Care Foundation used both a probability-based sample and nonprobability sample of nonelderly adults (age 18–64) in California. ↩︎
2. All respondents of the California Health Interview Survey were asked whether they felt they needed help for a mental or emotional or alcohol or drug problem in the past year. ↩︎

The No Surprises Act is a new federal law that establishes protections from unexpected bills for consumers with private health insurance.

Test your understanding of the law and share your results!

As states plan for the end of the COVID-19 public health emergency, the resumption of eligibility redeterminations and disenrollments when the continuous Medicaid enrollment requirement is lifted could lead to coverage disruptions and losses, according to a new KFF 50-state survey.

The requirement, a condition of states receiving enhanced federal Medicaid funding during the pandemic, has helped drive Medicaid and CHIP enrollment to a record 85 million people, an increase of 19.1 percent since the start of the pandemic. It prohibits states from disenrolling individuals while the public health emergency is in effect, to cushion low-income people against the economic and health disruptions of the pandemic.

But with the public health emergency expected to expire in the coming months, the new survey of state Medicaid and CHIP officials finds that many state Medicaid programs still have not made key decisions about how to handle the resumption of eligibility redeterminations that will lead to the disenrollment of people who are no longer eligible and those who are eligible but do not complete renewal paperwork.

Only about half of states (27) said they had a plan for how they will prioritize outstanding eligibility and renewal actions when the continuous enrollment requirement is lifted, the survey finds. The states with plans in place have different strategies and approaches, with 11 states indicating that they will target enrollees who appear to be no longer eligible first, while nine plan to conduct fresh renewals on a rolling basis based on the month in which the individual’s annual renewal is due. Seven states have adopted a hybrid approach. States that prioritize re-checking the status of people who no longer appear to be eligible will likely generate coverage disruptions sooner than states that wait until enrollees’ standard renewal date arrives.

Current guidance from the Centers for Medicare and Medicaid Services gives states up to 12 months to initiate and 14 months to complete all redeterminations. The survey – fielded prior to the release of the guidance – finds that of the 48 states that have established a timeframe for processing redeterminations, 41 states plan to take 9-12 months, four states plan to take 6-9 months, and three states plan to take 3-6 months. States that take less than a full year to complete all redeterminations could trigger greater coverage losses among eligible individuals, because there will be less time to conduct outreach and boost administrative staff capacity.

Across the 20 states that were able to report such data, about 13 percent of Medicaid enrollees are expected to be disenrolled when the continuous enrollment requirement ends. The estimates range widely, from about 8 percent of enrollees in West Virginia to over 30 percent of enrollees in Hawaii.

Such estimates suggest that millions of people will lose Medicaid coverage in the months following the end of the public health emergency. Many children will be eligible for CHIP and many adults will be eligible for Affordable Care Act (ACA) Marketplace or other coverage. Successfully transitioning such individuals into other coverage options could reduce the number who become uninsured.

Among other key findings from the survey:

• While states cannot disenroll people while the public health emergency is in place, as of January 2022, 42 states report conducting so- called “ex parte” renewals, which are processed automatically using data available to the agency, such as state wage databases. Continuing to process these renewals now can help to maintain coverage and reduce backlogs at the end of the public health emergency. However, a majority of states indicate that less than half of renewals can be completed using available data sources without having to follow up with the enrollee for more information about their income and circumstances.
• Forty-one states plan to follow up with enrollees when the enrollee must take action to avoid a loss of coverage. States are not required to follow up with enrollees who do not respond to a renewal request; however, sending reminder notices via mail – and also by phone, text, and/or email – can reduce the number of people who remain eligible but are disenrolled at the end of the public health emergency because they did not respond to a request for information.
• Almost all states (46) are planning to try to update enrollees’ mailing addresses, including through conducting data matches with the U.S. Postal Service, working with managed care organizations, and launching outreach campaigns. Also, 35 states say they will follow-up on returned mail via telephone, email, and text before terminating an enrollee’s coverage.
• Thirty states plan to boost staff capacity, including through approving overtime, hiring new eligibility workers, or borrowing staff from other agencies.
• Forty-one states have the data systems necessary to report the share of people who are disenrolled because they are determined to be ineligible versus the share who are disenrolled for procedural reasons. CMS has indicated it will require states to report monthly data to monitor their progress on unwinding and compliance with current rules.

These and other findings from the survey will be discussed today at a public web briefing. An archived video recording of the briefing will be available on kff.org later today. The public health emergency has been extended to mid-April, but the Biden Administration has indicated that states will have 60 days’ notice before it ends, so it is expected that the declaration will be extended again.

KFF conducted the 20th annual survey of states Medicaid and CHIP eligibility levels and enrollment and renewal policies in January 2022, with help from the Georgetown University Center for Children and Families. The survey presents a snapshot of actions states are taking to prepare for the end of the continuous enrollment requirement, as well as key state Medicaid enrollment and renewal procedures in place during the pandemic.

The full survey report, Medicaid and CHIP Eligibility and Enrollment Policies as of January 2022: Findings from a 50-State Survey, includes state-level data about Medicaid and CHIP eligibility in every state. Also available are other recent KFF analyses related to the end of the public health emergency, including Unwinding of the PHE: Maintaining Medicaid for People with Limited English Proficiency and Without Build Back Better, Will the End of the Public Health Emergency Leave Even More People Uninsured?

In the face of rising prescription drug costs, a large majority of the public supports federal efforts to lower drug spending. In his 2022 State of the Union address, President Biden urged Congress to pass legislation to rein in drug costs. In November 2021, the House of Representatives passed the Build Back Better Act (BBBA), which includes several provisions that would lower prescription drug costs, but the Senate has yet to take action on this legislation. Proposals included in the BBBA would allow the federal government to negotiate the price of some drugs covered under Medicare Part B (drugs administered by physicians and other health care providers) and Part D (retail prescription drugs); require drug companies to pay rebates to the federal government when annual increases in drug prices for Medicare and private insurance exceed the rate of inflation; cap monthly insulin costs for people with Medicare and private insurance; and cap Medicare beneficiaries’ out-of-pocket drug spending under Part D (but not Part B).

To better understand the potential out-of-pocket cost exposure that Medicare beneficiaries may face for Part B drugs, in this brief we analyze cost-sharing liability for these drugs in traditional Medicare and cost-sharing requirements in Medicare Advantage plans. Data limitations preclude us from analyzing actual out-of-pocket costs paid by beneficiaries who used Part B drugs. For traditional Medicare beneficiaries, claims data do not report separately cost-sharing liability paid directly by beneficiaries versus supplemental insurance (where applicable). For Medicare Advantage enrollees, there are no data available on actual out-of-pocket costs paid for Part B drugs.

Beneficiaries in traditional Medicare are charged 20% of the cost of Part B drugs, with no annual limit on their out-of-pocket costs. Beneficiaries enrolled in Medicare Advantage plans – which account for a growing share of the Medicare population and currently cover close to half of all beneficiaries – also typically face cost-sharing requirements for Part B drugs up to their plan’s out-of-pocket limit ($7,550 for in-network cost sharing and $11,300 for in-network and out-of-network cost sharing combined in 2022). Most but not all beneficiaries in traditional Medicare have some form of supplemental coverage to help with their Medicare cost-sharing requirements, while most Medicare Advantage enrollees do not. For example, most traditional beneficiaries who have Medigap have a policy that covers the 20% coinsurance for Part B drugs and services, while Medicaid and some of the Medicare Savings Programs cover Medicare cost sharing for eligible low-income beneficiaries in both traditional Medicare and Medicare Advantage. But even those with supplemental insurance may face some out-of-pocket costs for their Part B drugs, depending on the generosity of their coverage. In addition, roughly 6 million Medicare beneficiaries have no supplemental coverage and would be responsible for the full 20% coinsurance.

In the first part of this analysis, we examine Medicare claims data for 2019 to assess cost-sharing liability for Part B drugs for beneficiaries in traditional Medicare (excluding Part B vaccines since these are provided at no cost to Medicare beneficiaries). In the second part of our analysis, we use Medicare Advantage benefit design data for 2022 to examine the range in cost-sharing amounts at or below 20% coinsurance charged by Medicare Advantage plans for in-network Part B drugs. We also analyze variation in cost-sharing amounts for out-of-network Part B drugs charged by plans that provide out-of-network coverage. (See Methods for additional details on both parts of our analysis.)

Findings

One-fourth of the 4.1 million traditional Medicare beneficiaries who used one or more Part B drugs in 2019 had average annual cost-sharing liability of at least $1,000

Of the 4.1 million beneficiaries in traditional Medicare who received one or more Medicare Part B drugs included in this analysis, 1 in 4 (1.0 million beneficiaries) faced cost-sharing liability of at least $1,000 and nearly 1 in 5 (0.7 million) faced cost-sharing liability of at least $2,000 in 2019 (Figure 1). About 0.4 million traditional Medicare beneficiaries – or 1 in 10 of those who used Part B drugs – had at least $5,000 in cost-sharing liability for these drugs in 2019. As previously noted, we are unable to analyze how many beneficiaries had supplemental insurance to cover some or all of these costs and how many were responsible for paying the full amount out-of-pocket.

For more than half of all Part B drugs included in this analysis, average annual cost-sharing liability was $1,000 or more in 2019

Of the 287 Part B drugs included in this analysis, more than half (54% or 155 drugs) had average annual cost-sharing liability of at least $1,000 in 2019, and more than 4 in 10 (43% or 123 drugs) had cost-sharing liability of at least $2,000.

• Eight of the top 10 Part B drugs with the highest total spending, and 18 of the top 20, had average annual cost-sharing liability of at least $1,000 in 2019 (Figure 2, Table 1). For example, average beneficiary liability for Eylea, the top-spending Part B drug in 2019 used by 270,300 traditional Medicare beneficiaries to treat macular degeneration, was $2,100. For the cancer drug Keytruda, second in terms of total Medicare Part B spending, average cost-sharing liability was $9,100.
• Conversely, average annual cost-sharing liability for most of the Part B drugs with the highest number of users was below $1,000. Only 2 of the top 10 Part B drugs based on the number of users, and 4 of the top 20, had average annual cost-sharing liability of at least $1,000 in 2019 (Table 2). For example, average beneficiary liability for the most-commonly used (non-vaccine) Part B drug, Prolia (and the equivalent brand Xgeva), an osteoporosis treatment used by nearly 600,000 traditional Medicare beneficiaries in 2019, was $600.

Average annual cost-sharing liability exceeded $10,000 for more than 1 in 10 Part B drugs in this analysis (13% or 36 drugs). While some of the highest-liability drugs were used by relatively few beneficiaries to treat rare conditions, two of these drugs were used by more than 10,000 beneficiaries in 2019: Opdivo, a treatment for several types of cancer used by 30,300 beneficiaries, with average annual cost-sharing liability of $10,200; and Darzalex, a treatment for multiple myeloma used by nearly 12,000 beneficiaries, with average annual cost-sharing liability of $12,900.

Like beneficiaries in traditional Medicare, Medicare Advantage enrollees typically face 20% coinsurance for Part B drugs, but can be exposed to higher cost-sharing requirements for these drugs when administered by an out-of-network provider

Medicare Advantage plans have flexibility to determine cost-sharing amounts for Part B covered drugs, subject to certain limits, and can differentiate cost sharing for chemotherapy from other Part B drugs. In 2022, Medicare Advantage plans are prohibited from charging more than 20% coinsurance, or the equivalent copay amount, for both chemotherapy and other Part B drugs from in-network providers. There are no similar restrictions on out-of-network cost-sharing amounts.

Based on our analysis of Medicare Advantage plan cost-sharing requirements and enrollment as of January 2022:

• Consistent with current requirements, all 22.9 million Medicare Advantage enrollees included in this analysis (excluding enrollees in employer group plans) face coinsurance of 20% or less (or equivalent cost sharing) for chemotherapy and other Part B drugs when provided in-network (Figure 3). Some Medicare Advantage plans always charge less than 20% coinsurance for both types of drugs; for example, 25% of HMO enrollees and 9% of PPO enrollees and other plans that offer out-of-network coverage charge less than 20% coinsurance for Part B drugs administered in network. Other plans vary cost-sharing amounts for Part B drugs (e.g., a $0 copayment or 20% coinsurance) depending on the type of drug (e.g., chemotherapy vs. other Part B drugs) and/or where it is furnished (e.g., a pharmacy or other provider setting).

• Among the 8.3 million Medicare Advantage enrollees in PPOs and other types of plans with out-of-network coverage, one-fourth of these enrollees (2.1 million) face 20% coinsurance for Part B drugs received from an out-of-network provider, while close to half (44% or 3.7 million) would be charged more than 20% for these drugs if administered by an out-of-network provider (Figure 4). Among these enrollees, 1.1 million are in plans that charge 50% for out-of-network Part B drugs, 1.1 million are in plans that charge 40% coinsurance, and 0.9 million enrollees are in plans that charge a 30% coinsurance (0.9 million). Another 31% of enrollees with out-of-network coverage (2.5 million) may face coinsurance higher than 20% depending on the type of drug and/or where the drug is administered.

• The 14.7 million Medicare Advantage enrollees in HMOs and other plans with no out-of-network coverage would pay 100% of the cost for Part B drugs administered out-of-network unless they received prior approval from their plan.

Unlike traditional Medicare, Medicare Advantage plans have an out-of-pocket limit, but Medicare Advantage enrollees can still face substantial cost-sharing amounts for some of the higher-priced Part B drugs, especially if administered out-of-network.

• For Eylea, a treatment for macular degeneration, Medicare Advantage enrollees would face average cost-sharing liability of $2,100 at a 20% coinsurance rate and up to $5,200 at a 50% coinsurance rate, assuming the plan pays the same price as traditional Medicare.
• For Rituxan, a cancer and autoimmune treatment, Medicare Advantage enrollees would face average cost-sharing liability of $4,600, based on a 20% coinsurance rate for in-network chemotherapy, but could face costs up to their plan’s maximum out-of-limit for this one drug alone based on higher out-of-network coinsurance rates (up to $11,300 in 2022).
• For Prolia (and the equivalent brand Xgeva), an osteoporosis treatment, Medicare Advantage enrollees would face average cost-sharing liability of $600 at a 20% coinsurance rate and up to $1,400 at a 50% coinsurance rate, assuming the plan pays the same price for these drugs as traditional Medicare.

Discussion

Beneficiaries in both traditional Medicare and Medicare Advantage can be exposed to potentially high out-of-pocket costs for Part B drugs – though data limitations preclude us from analyzing out-of-pocket costs paid directly by beneficiaries who used Part B drugs in either traditional Medicare or Medicare Advantage plans and the share of costs covered by private supplemental insurance or Medicaid. We find that, overall, 1 in 4 beneficiaries in traditional Medicare who used Part B drugs in 2019 faced cost-sharing liability of at least $1,000, and more than half of Part B drugs included in this analysis had average cost-sharing liability of $1,000 or more. Given the absence of an out-of-pocket limit for beneficiaries in traditional Medicare, the costs of chemotherapy and other Part B drugs could represent a substantial financial burden for beneficiaries with no supplemental coverage, or with supplemental coverage that does not cover all cost-sharing requirements for these drugs.

While Medicare Advantage plans are required to have a maximum out-of-pocket limit and can charge no more than 20% for Part B drugs administered by an in-network provider, most Medicare Advantage enrollees would face higher costs for Part B drugs furnished by an out-of-network provider, including close to 4 million enrollees in plans with out-of-network coverage and nearly 15 million enrollees in HMOs and other plans with no out-of-network coverage. Plans with out-of-network coverage typically charge higher cost sharing for Part B drugs and other services received out-of-network to encourage enrollees to receive care from in-network providers where plans have negotiated lower prices. These network arrangements and cost-sharing differences can have potentially large cost implications for Medicare beneficiaries.

Among the set of proposals that policymakers have recently considered to lower prescription drug costs, allowing the federal government to negotiate prices for some drugs covered under Part B and Part D and requiring inflation rebates for Medicare-covered drugs to limit annual increases in drug prices could help to address the spending burden that Medicare beneficiaries could face if they need high-cost drugs, whether covered under Part B or Part D.

Tables

A new analysis from KFF and Epic Research finds that telehealth visits for outpatient mental health and substance use services went from virtually zero percent in 2019 prior to the COVID-19 pandemic to a peak of 40% in mid-2020 – and continued to account for more than a third (36%) of such visits in the six months ending in August 2021.

The telehealth boom for mental health and substance use services far exceeds the increase recorded over the same period for other outpatient services. As a result, mental health and substance use services accounted for 39% of all telehealth outpatient visits in the six-month period between March and August 2021, about five times the share of all other outpatient visits. This reflects the tremendous increase in need for mental health services as a result of the pandemic, social distancing and ensuing economic turmoil, as well as a return to in-person visits for other outpatient care. 

The analysis examines data from Cosmos, Epic’s HIPAA-defined limited data set of more than 126 million patients from hospitals and clinics across the country. It also examines variations in telehealth use by age, gender and diagnosis.

There has been a rapid increase in the use of telehealth thus far into the COVID-19 pandemic for both mental and physical health concerns. While many employer health plans covered telehealth prior to the pandemic, utilization of these services was relatively low, accounting for less than 1% of outpatient visits. At its pandemic peak, telehealth represented 13% of outpatient visits between March and August of 2020. As in-person care resumed, telehealth began to represent a smaller share of outpatient care (8% between March and August 2021). While many continue to envision an expanded role for telehealth in the delivery of care following the pandemic, there remains considerable uncertainty in what services will be available, where and how providers will be able to practice, how benefits will be structured, and how providers will be paid.

The COVID-19 pandemic has taken a toll on the nation’s mental health, with 3 in 10 adults in the U.S. reporting symptoms consistent with depression or anxiety disorder since April 2020. Over 20% of adults reporting poor mental health also report not receiving counseling or therapy during the pandemic. Telehealth has played a particularly significant role in meeting the need for mental health services. Thus far into the pandemic, some private payers have improved coverage for mental health and substance use, removing pre-pandemic restrictions on coverage for these services via telehealth. Similarly, Medicaid expanded coverage of telehealth services during the pandemic for mental health and substance use services. Many state Medicaid programs reported that telehealth has had particular value in maintaining or improving access to such services during the pandemic.

This analysis looks at outpatient visits during five six-month periods between March 2019 and August 2021, using data from Cosmos, a HIPAA-defined Limited Data Set of more than 126 million patients from over 156 Epic organizations, including 889 hospitals and 19,420 clinics across all 50 states. Mental health and substance use visits were identified based on the primary ICD-10-CM diagnosis code listed for the visit. We find that a year and a half into the pandemic, telehealth continues to play a significant role in providing services, particularly for mental health and substance use services.

Mental health and substance use services by telehealth has remained elevated whereas other outpatient care use by telehealth has declined

Telehealth represented less than 1% of outpatient care before the pandemic (rounding to zero) for both mental health and substance use and other concerns. However, at its pandemic peak, telehealth represented 40% of mental health and substance use outpatient visits and 11% of other visits (during the March- August 2020 period). Since then, in-person care has returned and telehealth visits have dropped off to represent 5% of other outpatient care visits, those without a mental health or substance use claim in the March-August 2021 period. However, telehealth use has remained strong for mental health and substance use treatment, still representing 36% of these outpatient visits.

Telehealth use for mental health or substance use continues to grow as a share of all telehealth visits

Mental health and substance use visits represent a growing share of both telehealth visits and outpatient visits overall, but the trend is much more pronounced for telehealth. By the period of March-August 2021, 39% of telehealth outpatient visits were primarily for a mental health or substance use diagnosis compared to 24% a year earlier and 11% two years earlier. Among all outpatient visits (in-person and over telehealth), the share with a mental or substance use diagnosis grew from 4% in March-August 2019 to 8% during the pandemic, and has remained at 8% in March-August 2021.

This is a product of several factors, including the tremendous increase in need for mental health services as a result of the pandemic, social distancing and ensuing economic turmoil, as well as a return to in-person visits for other outpatient care.

Rural residents are more likely to use telehealth for mental and substance use disorder visits

The adoption of telehealth varies based on the health needs of patients, their readiness and ability to adopt the technology, and the restrictions or incentives they face from payers and providers. Looking at the most recent study period (March-August 2021), a relatively high share of patients in rural areas relied on telehealth to receive outpatient mental health and substance use services (55%) compared to those in urban areas (35%). This pattern is especially pronounced in comparison to other outpatient health care services, where we observed a similar rate across urban and rural (5% vs. 6%) areas. This may be impacted by the number of rural areas which have a shortage of mental health providers.

Non-elderly adults consistently used telehealth to access mental health and substance use services

Early on in the pandemic (March-August 2020), outpatient services for mental health and substance use were delivered by telehealth at a similar rate among children and the elderly, and a slightly lower rate among non-elderly adults. A year later (March-August 2021), non-elderly adults were somewhat more likely to be treated via telehealth compared to kids and seniors. In the most recent period, 58% of all mental health or substance use outpatient visits, and 62% of these visits performed via telehealth, were among people in the 19-64 age group.

Telehealth use is significant across major mental health and substance use disorder conditions

The primary ICD-10-CM diagnosis code listed on an outpatient visit was categorized into conditions based on the Clinical Classifications Software Refined (CCSR). Major conditions were selected if there were more than 5,000 total visits (in-person and telehealth) during the March-August 2021 period. A large share of outpatient visits for major mental and substance use conditions were delivered over telehealth during this period, including for substance use disorders such as opioid-related disorders (29%) and alcohol-related disorders (29%). For mental health needs, over 1 in 3 outpatient visits were delivered by telehealth (for example, 35% and 38% of outpatient visits were over telehealth for depression or anxiety, respectively).

Discussion

How the mass adoption of telehealth affects access, cost, and quality of mental health and substance use disorder services remains to be seen. Telehealth may provide a way to improve access to mental health and substance use disorder care, particularly for people living in areas with fewer providers. Though differences in comfort with technology, digital literacy and lack of internet at home may hinder access to telehealth for some.

Payers’ and regulators’ policies governing telehealth services continue to evolve. As required under the Consolidated Appropriations Act of 2021 (CAA), and as implemented under the CY 2022 Medicare Physician Fee Schedule Final Rule, Medicare has permanently removed geographic restrictions for mental health and substance use services and permanently allows beneficiaries to receive those services at home. Also under the Physician Fee Schedule final rule, CMS will now permanently cover audio-only visits for mental health and substance use services, though only when the beneficiary is not capable of, or does not consent to, the use of two-way, audio/video technology.

States have broad flexibility to determine Medicaid coverage of telehealth services as well as providers’ licensure requirements to practice and prescribe during a telehealth consultation. States expanded Medicaid telehealth coverage in response to the pandemic, with nearly all states covering and paying parity for audio-visual and audio-only mental health and substance use disorder visits in their fee-for-service Medicaid programs as of July 2021. Many states plan to maintain all or some of these expanded telehealth policies post-pandemic, especially flexibilities for behavioral health visits. However, some states have moved to reinstate certain rules that limit telehealth use, providers’ ability to practice across state lines, and providers’ ability to prescribe medication without an in-person visit. In some other states, telehealth coverage expansions for Medicaid are tied to the COVID-19 public health emergency (PHE) and will expire with the PHE unless extended. State and federal laws regarding providers’ abilities to prescribe medication via telehealth and practice across state lines will also affect telehealth use in the future. Temporary federal policies that gave qualified providers more flexibility to prescribe controlled substances over telehealth are set to expire at the end of the PHE.

Data during the pandemic suggest there is a concerning increase in the number of individuals reporting symptoms of depression or anxiety and showing signs of substance use disorder. Many employers have expressed concern about the availability of mental health providers in their plan network. Given the increased need for services and concerns about the availability of mental health and substance use care providers, some payers may look to bolster access to these services through telehealth.

As policy makers continue to look at how to regulate and pay for telehealth services, it is important to consider opportunities for patient choice so that telehealth is not necessarily given as the only option for those looking for care. Additionally, the Substance Abuse and Mental Health Services Administration (SAMHSA) has called for refining clinical guidelines for tele-mental health services, including broader adoption of telehealth for mental and substance use disorders. In particular, how laws affecting prescribing patterns for mental health and substance use therapies are devised after the COVID-19 public health emergency ends may influence the long-term use of telehealth for mental health and substance use disorders.

Rising health care prices have led premiums and deductibles for employer-sponsored coverage to grow faster than wages and general inflation, creating affordability challenges for employers and employees. One proposal to address the high prices paid by private insurers is to cap these prices at a multiple of Medicare rates. States, including Montana and Oregon, have adopted this approach for certain providers under their public employee health plan. In Washington state, the public option in the individual market ties payments to a percent of Medicare rates. This brief considers the potential implications for inpatient admissions and spending of applying a price cap to all private insurance hospital payments in the large employer group market.

In this analysis, we look at in-network payment rates for inpatient hospital stays, other than maternity/newborn admissions, among large employer plans relative to Medicare payment rates. To do so, we analyzed data from the 2018 IBM MarketScan Commercial Claims and Encounters Database that includes health claims from a sample of 18 million non-elderly people, representing about 82 million covered lives in large employer plans. Specifically, we examined the share of non-maternity inpatient hospital admissions and associated spending among large employer plans for in-network inpatient admissions paid above various ratios of private-to-Medicare rates. We then focus on 15 common types of admissions (classified using diagnosis related groups, or DRGs), representing more than a quarter of all non-maternity inpatient spending in the large group market (see methods for additional information). This analysis is intended to be illustrative and does not assess potential spillover effects on volume, access, or quality of a policy that caps prices for the privately insured.

Findings

About half of non-maternity inpatient hospital admissions in the large group market would be affected by a cap on prices set to 150% of Medicare rates. Based on our analysis, 52% of inpatient admissions were paid above 150% of Medicare rates, meaning just over half of all admissions would be affected by a cap on payments set at this level. Moving from 150% to 200% of Medicare rates, the share drops to just about one-third of admissions (32%). If the payment rate was capped at 300% of Medicare rates, 13% of admissions would be affected (Figure 1).

A cap set at a multiple of Medicare rates would affect a sizeable amount of inpatient spending in the large group market, even if set at 300% of Medicare rates. Just over one-third of non-maternity inpatient spending in the large group market is for spending associated with the portion of prices above 150% of Medicare rates. In other words, if no admission was paid more than 150% of Medicare, all else equal, spending would be 36% lower. Capping prices at 300% of Medicare rates would affect 13% of spending covered by employer plans (Figure 1).

Across 15 common types of admissions, the share of in-network admissions paid above 150% of Medicare varies substantially. For example, among patients covered by large group plans, 71% of admissions for hip and knee joint replacements (DRG 470) are paid more than 150% of Medicare rates, whereas only 15% of admissions for psychosis (DRG 885) are paid above 150% of Medicare rates.

Among these same 15 DRGs, between 3% and 19% of large group plan admissions have payments above 300% of the Medicare rates. For 6 of these 15 DRGs, at least 15% of inpatient stays in the large group market are paid above 300% of Medicare rates (DRG 247, DRG 460, DRG 219, DRG 025, DRG 871 and DRG 853). Figure 2 shows the share of admissions paid within different ranges of multiples of Medicare rates.

In other words, because the distribution of current prices varies widely across DRGS, the choice of where to set a cap would matter more for some types of admissions than others (Figure 2). It would also have different affects across individual hospitals depending on their current prices and distribution of admissions.

The average price for an in-network large group admission would fall more for admissions that currently have higher prices relative to Medicare. To illustrate how a cap set at a percent of Medicare rates might work, we repriced admissions currently paid above each multiple of the Medicare payment rate and recalculated the average price for the DRG in the large employer market. If a cap is set at 200% of Medicare rates, we assumed all large group admissions that are currently paid a at a higher rate would be paid the lower price of 200% of Medicare. For example, for hip and knee replacements, the average price paid by large employer plans would decrease from $30,506 to $25,366. The decrease would be smaller for psychoses under the same scenario, falling from $9,425 to $8,420 (Figure 3). In these illustrative calculations, we made no assumptions about spillover effects on the volume of admissions or prices paid underneath the cap. In practice, volume could increase to compensate for lower average prices with a cap, and the price for admissions below the cap could rise for the same reason.

Conclusions

High and rising health care costs, driven by high and rising prices, and exacerbated by increasing provider consolidation, are contributing to affordability challenges for people with employer-sponsored coverage. To counteract these effects, some have considered capping prices as a multiple of Medicare rates, following the lead of states, such as Montana, that have adopted this policy for its state employees’ health insurance plan.

Our analysis finds that a cap of 150% of Medicare rates would affect 52% of in-network admissions and 36% of in-network spending, while a cap of 300% of Medicare rates would affect 13% of in-network admissions and 13% of in-network spending, with variation across types of admissions. A lower cap would affect a larger share of admissions than of spending, because at lower levels more admissions are currently paid just above the potential cap. A cap on prices could potentially apply to both in- and out-of-network services, as has been proposed in other contexts. Our analysis examines in-network admissions, which account for the vast majority of spending in the group market. Therefore, our findings are illustrative of the range of admissions that could be affected and would not be substantially different if we included out-of-network admissions.

A cap on prices paid in the group market could be disruptive depending on the level at which the cap is set and the number of services to which it applies. At the same time, if a cap achieved meaningful savings, it could make health care more affordable - tradeoffs that warrant careful attention. In our analysis, we do not attempt to model any changes that may follow from a cap on prices, such as an increase in volume or an increase in prices for admissions that are below the cap. However, in Montana, capping prices at a multiple of Medicare rates resulted in net savings and utilization did not increase. There could also be effects on networks. In Washington state, some hospitals are not participating in the network for the public option in the individual market because the cap on prices of 160% of Medicare is too low. While it is unlikely that hospitals could forgo the large group market entirely, it is possible that some hospitals would choose to contract with fewer plans, creating access concerns.

While this analysis does not consider the potential impact on quality, it is possible that some high-priced inpatient admissions are of high quality and restricting prices could have a negative impact in those cases. There is some evidence this may be the case in unconcentrated (more competitive) markets. Additionally, a recent report by the Congressional Budget Office notes a correlation between prices and quality, though states there is not evidence of causality and it is not clear whether higher prices lead to higher quality or vice versa.

Capping the prices employer-sponsored plans pay for inpatient admissions would likely reduce hospital revenue. The magnitude of the impact would depend on the level of the cap, and whether volume changed in response. While affected providers could respond by operating more efficiently, it is also possible that decreases in revenue could lead to lower pay for hospital staff, fewer capital investments, and efforts to shift admissions away from payers who pay lower prices. These changes could be especially unpalatable during the ongoing COVID-19 pandemic, which has prompted billions of dollars in federal funding to help support and stabilize hospital finances.

Methods

To calculate private insurance payment rates, we analyzed a sample of medical claims obtained from the 2018 IBM Health Analytics MarketScan Commercial Claims and Encounters Database. We only included claims for people under the age of 65. This analysis used claims for 18 million people representing about 22% of the 82 million people in the large group market in 2018. Weights were applied to match counts in the Current Population Survey for enrollees at firms of a thousand or more workers by sex, age, and state. Weights were trimmed at eight times the interquartile range. We exclude admissions accounting for 12% of total spending in Marketscan, namely DRGs related to childbirth. When selecting common DRGs, we did not include DRGS with fewer than 800 observations, regardless of the amount of spending attributed to those cases.

Averages represent the amounts paid to the hospitals for an admission. Across all the DRGs, hospital spending represented about 87% of the total cost of the admission. Costs include both amounts paid by enrollees in the form of cost sharing and spending by the plan. Hospital costs are trimmed to exclude the highest 0.5% of hospital costs within a DRG and admissions below 5% of the median. This is intended to exclude admissions in which the claims do not capture all the spending on the admission. These data reflect cost sharing incurred under the benefit plan, but do not include balance-billing payments that beneficiaries may make to health care providers for out-of-network services delivered during the admission or out-of-pocket payments for non-covered services. Only admissions with in-network room and board charges are included. Limiting to in-network admissions does not qualitatively affect our findings. In-network spending is about 90% of total spending after trimming outliers.

To calculate Medicare payment rates, we analyzed average payments to hospitals for admissions identified through the DRG, as reported in Inpatient Charge Data FY 2018. According to CMS documentation, the DRG reimbursements are "the average total payments to all providers for the DRG including the MS-DRG amount, teaching, disproportionate share, capital, and outlier payments for all cases. Also included in average total payments are co-payment and deductible amounts that the patient is responsible for and any additional payments by third parties for coordination of benefits." These files are prepared by the Centers for Medicare and Medicaid Services (CMS) using Medicare Provider Analysis and Review (MEDPAR) data.  For more information see: https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Medicare-Provider-Charge-Data/Downloads/Inpatient_Methodology.pdf.

IBM assigns a DRG to each admission using the Centers for Medicare & Medicaid Services (CMS) Grouper 37. This method selects a DRG for the admission based on the diagnosis and procedures a patient received during the case. The total payments to hospitals in the Marketscan data reflect the payments made to the hospitals. Some variation in the payment rates of admissions is accounted for by differences in the intensity or types of services that a patient receives, and not differences in the rates paid for those services. The rates Medicare uses to reimburse DRGs are designed to account for this variation in the intensity of cases and services. This analysis compares the average of DRG payment rates in Medicare to admissions in Marketscan. CMS suppresses DRGs with low counts; in order to use the most DRGs, we use national averages rather than the price in particular States or MSAs. Therefore, this analysis does not account for regional variation in Medicare reimbursement.

In Medicare, hospital admissions are reimbursed based on DRGs that reflects a patient’s clinical conditions and treatment. In contrast, private insurers pay for hospital admissions using different approaches that may vary with the procedures performed during the stay, including per diem payments, discounted fee-for-service payments, DRGs or other combinations of payments and performance incentives. Therefore, the variation in payment within DRG among large group plans, may reflect both variation in the intensity of services and the length of stay, as well as the prices being paid for those services.

Americans Likely Owe Hundreds of Billions of Dollars in Total Medical Debt

A new KFF analysis of government data estimates that nearly 1 in 10 adults (9%) – or roughly 23 million people – owe medical debt. This includes 11 million who owe more than $2,000 and 3 million people who owe more than $10,000.

The analysis is based on data from the 2020 Survey of Income and Program Participation, a nationally representative survey that asks every adult in a household whether they owed money for medical bills in 2019 and how much they owe. It looks at people with medical debt of more than $250.

The 2020 survey suggests Americans’ collective medical debt totaled at least $195 billion in 2019, though with quite a bit of uncertainty. A small share of adults account for a huge share of the total, with considerable variation from year to year. The estimate is significantly higher than other commonly cited estimates, which generally rely on data from credit reports that may not capture medical debts charged to credit cards or included in other debts rather than being directly owed to a provider.

Other findings include:

• People ages 35-49 (11%) and 50-64 (12%) are more likely than other adults to report medical debt. They have greater health needs than younger people on average and aren’t yet old enough to qualify for Medicare coverage, which may protect them from high costs.
• Larger shares of people in poor health (21%) and living with a disability (15%) report medical debt. People in these groups are more likely to need and receive care than people in better health and without disabilities.
• Among racial and ethnic groups, a larger share of Black adults (16%) report having medical debt compared to White (9%), Hispanic (9%), and Asian American (4%) adults.
• Adults who were uninsured for more than half of the year are more likely to report medical debt (13%) than those who were insured for all or most of the year (9%).

It’s not yet clear how the pandemic and resulting recession affected medical debt. Many people lost jobs and income early in the pandemic, which could have led to more difficulty affording medical care. At the same time, many people delayed or went without care, so fewer people may have been exposed to costly medical care. Shifts in insurance coverage and COVID-related cost-sharing waivers could also affect what people had to pay out-of-pocket.

Two related KFF analyses look at other challenges related to health costs:

The first finds that many households do not have enough money available to cover the cost of a typical deductible in a private health plan. For example, about a third (32%) of single-person households with private insurance in 2019 could not pay a $2,000 bill, and half (51%) could not pay a $6,000 bill.

The second finds that lower-income families (less than twice the federal poverty level) who have employer health coverage spend a tenth of their income on health care on average, including both their share of the premium and their out-of-pocket costs.

The new analyses are available through the KFF-Peterson Health System Tracker, an online information hub that monitors and assesses the performance of the U.S. health system.