Skip to main content

Medicaid: HHS's Preliminary Analyses Offer Incomplete Picture of Behavioral Health Demonstration's Effectiveness

GAO-21-394 Published: May 17, 2021. Publicly Released: May 17, 2021.
Jump To:

Fast Facts

Untreated mental health issues or substance use disorders can have ripple effects. In 2014, Congress approved a 2-year program to improve mental health services in up to 8 states, particularly for Medicaid beneficiaries. The program was extended to Sept. 2023.

HHS will continue to have trouble determining how well the program works due to problems with program data.

For example, most clinics hadn't reported quality measures before, so there is no baseline data available for measuring progress.

As another example, HHS didn't tell states which services to provide under the program—a design decision that makes uniform assessments more difficult.

A seated person with a clipboard speaking to another person with their hands clasped.

Skip to Highlights


What GAO Found

The Protecting Access to Medicare Act of 2014 (PAMA) established the Certified Community Behavioral Health Clinics (CCBHC) demonstration and tasked the Department of Health and Human Services (HHS) with its implementation. CCBHCs aim to improve the behavioral health services they provide, particularly for Medicaid beneficiaries. Initially established for a 2-year period, the demonstration has been extended by law a number of times; most recently, it was extended to September 2023. States participating in the demonstration can receive Medicaid payments, consistent with federal requirements, for CCBHC services provided to beneficiaries.

PAMA also required HHS to assess the effect of the demonstration on service access, costs, and quality. HHS's preliminary assessments of the demonstration in eight states, with 66 participating CCBHCs, found the following:

  • Access. CCBHCs commonly added services related to mental and behavioral health, such as medication-assisted treatment, and took actions to provide services outside the clinic setting, such as through telehealth.
  • Costs. States' average payments to CCBHCs typically exceeded CCBHC costs for the first 2 years of the demonstration. CCBHC payments and costs were more closely aligned in the second year for most states, better reflecting the payment methods prescribed under the demonstration.
  • Quality. States and CCBHCs took steps, such as implementing electronic health records systems, to report performance on 21 quality measures.

GAO found data limitations complicated—and will continue to affect—HHS's efforts to assess the effectiveness of the demonstration. For example:

  • Lack of baseline data. PAMA requires HHS to assess the quality of services provided by CCBHCs compared with non-participating areas or states. The demonstration marked the first time these clinics reported performance on quality measures, so no historical baseline data exist. HHS officials noted that with time, additional data may provide insight on the quality of services.
  • Lack of comparison groups. PAMA requires HHS to compare CCBHCs' efforts to increase access and improve quality with non-participating clinics and states. HHS was unable to identify comparable clinics or states due to significant differences among the communities.
  • Lack of detail on Medicaid encounters. PAMA requires HHS to assess the effect of the demonstration on federal and state costs and on Medicaid beneficiaries' access to services. HHS plans to use Medicaid claims and encounter data to assess such changes. However, GAO has previously identified concerns with the accuracy and completeness of Medicaid data and has made numerous recommendations aimed at improving their quality.

HHS's decisions in implementing the demonstration also complicated its assessment efforts. HHS allowed states to identify different program goals and target populations, and to cover different services. HHS also did not require states to use standard billing codes and billing code modifiers it developed. The lack of standardization across states limited HHS's ability to assess changes in a uniform way.

Why GAO Did This Study

Behavioral health conditions—mental health issues and substance use disorders—affect millions of people. HHS estimates that 61 million adults had at least one behavioral health condition in 2019—41 million of whom did not receive any related treatment in the prior year.

Many individuals with behavioral health conditions rely on community mental health centers for treatment, but the scope and quality of these services vary. To improve community-based behavioral health services, PAMA created the CCBHC demonstration and provided HHS with $25 million to support its implementation.

PAMA directed HHS to assess the demonstration and to provide recommendations for its continuation, modification, or termination. To date, HHS has issued three annual reports assessing the initial demonstration period, which ran from 2017 to 2019. HHS plans to issue a fourth annual report and a final report by December 2021.

This report describes HHS's assessment of the demonstration regarding access, costs, and quality. Under the CARES Act, GAO is to issue another report on states' experiences by September 2021.

GAO reviewed federal laws and regulations; HHS guidance; and HHS's assessments of the demonstration, including three issued reports, interim reports, and the analysis plan for future reports. GAO also interviewed HHS officials and officials from organizations familiar with community health clinics.

HHS provided technical comments, which GAO incorporated as appropriate.

For more information, contact Carolyn L. Yocom (202) 512-7114 or

Full Report

GAO Contacts

Office of Public Affairs


Actual costsAdultsAgency evaluationsBeneficiariesCommunity mental health centersCrisisElectronic health recordsHealth careHealth care centersHealth care standardsHealth services administrationManaged health careMedicaidMedicareMental healthMental health servicesQuality of careSchizophreniaSchoolsSubstance abuse