Key Findings and Recommendations

Finding 1: Telehealth policy flexibilities implemented during the pandemic enabled exponential growth of telehealth use, though there were significant differences across states in the level of telehealth utilization.

• Five of the selected states reported delivering 32.5 million services via telehealth to approximately 4.9 million beneficiaries between March 2020 and February 2021, up from 2.1 million services delivered to about 455,000 beneficiaries during the same time period in the previous year.
• Overall services (telehealth and in-person) delivered across the five selected states declined by nearly 64 million during the same time period, suggesting that despite increased access to telehealth, many beneficiaries forwent care during that year.
• The percent of overall services delivered via telehealth followed a similar trajectory across states, rising sharply in April 2020 and then plateauing at a rate higher than before the pandemic by fall 2020. In April 2020, some states reported higher percentages of services delivered via telehealth (e.g., Arizona and Maine reported ~15%) compared to other states (e.g., Mississippi and Missouri reported less than 10%).
• There was notable variation in telehealth utilization across states when comparing time periods before and during the pandemic. For example, Arizona’s Medicaid program updated its telehealth policy in October 2019 and had been actively delivering services. As a result, Arizona reported the highest percentage (11%) of Medicaid beneficiaries who had received at least one service via telehealth prior to the pandemic; other states reported rates between 1% and 3%. Despite the variation in pre-pandemic utilization, Arizona, California and Maine reported similar rates (~40%) of the same metric during the first year of the pandemic; Mississippi and Missouri reported notably lower percentages, 25% and 20%, respectively.
• Four of the selected states with reliable demographic data reported similar trends in telehealth utilization over time across race and ethnicity groups; however, these findings may be obscured due to a high percentage of beneficiaries of unknown race and ethnicity, highlighting the need for accurate and complete race and ethnicity telehealth utilization data.
• Beneficiaries in urban areas utilized telehealth at a higher rate compared to those in rural areas, and older beneficiaries (age 65+) were less likely to receive services via telehealth. Three states reported higher utilization among men as compared to women.

Finding 2: The expansion of telehealth coverage and payment policies improved access to care, and states are now applying lessons learned during the pandemic to the development of post-PHE policies.

• Expanding the use of telehealth supported states’ ability to maintain access to care and mitigate certain barriers, such as reducing COVID-19 exposure, provide alternative care sites, mitigate child care and transportation barriers, and others.
• Limitations of telehealth included less comprehensive well-child visits, limited effectiveness for physical therapy, and technological barriers among both providers and beneficiaries.
• As of November 2021, all selected states were considering revising their telehealth policies to enable continued access to virtual care post-PHE. Modifications states are considering include audio-only, eligible originating sites (including patient’s home) and covered telehealth services.

Finding 3: CMS and states are starting to conduct fraud risk assessment activities related to telehealth.

• CMS identified potential types of risk related to the expansion of telehealth and has implemented some elements of the GAO’s fraud risk framework. However, as of September 2021, agency officials indicated that they had not yet determined fraud risk tolerance for services delivered via telehealth and had no plans to do so.
• All six states conducted program integrity activities to mitigate the risk of fraud, waste and abuse that may result from increased access to telehealth services. Activities included issuing written guidance, offering providers technical assistance and conducting data analysis. Some states discussed investigations related to delivery of telehealth services, and one reported investigating a provider who billed an unusually high number of telehealth services.

Finding 4: There is currently no uniform approach to assessing quality of care delivered via telehealth, and CMS does not have plans to collect telehealth-specific quality data or report on the impact of telehealth on quality of care.

• Despite recent efforts by the National Quality Forum to outline a framework to support telehealth-specific measure development and CMS’ responsibility to oversee states’ efforts to ensure quality of care for Medicaid beneficiaries, CMS does not currently collect, assess or report information related to the impact of telehealth services on quality of care.
• CMS officials indicated that they do not have plans to update existing state reporting requirements, such as Adult and Child Core Sets, to differentiate between services delivered in person or via telehealth. As a result, CMS lacks critical telehealth-specific data and is not positioned to assess quality of care delivered via telehealth.
• Four states reported varying approaches to assessing quality of care delivered via telehealth, including beneficiary surveys, services use analysis and external studies.

Recommendations: Based on these findings, GAO is making two recommendations to CMS:

1. Collect and analyze information about the effect delivering services via telehealth has on the quality of care Medicaid beneficiaries receive.
2. Determine any next steps based on the results of the report.

Implications for State Medicaid Programs

This GAO report shines a light on current gaps in federal and state telehealth quality-monitoring activities. In response to this report, the state should continue to develop infrastructure and processes to enable data collection and reporting related to quality of care delivered via telehealth. In addition, states should enhance data collection related to beneficiary demographics to enable more accurate analysis of the impact of telehealth on health equity and digital inclusion. Finally, states should review existing program integrity activities related to fraud, waste and abuse for all services (in-person and telehealth) and consider adapting existing procedures or adopting new ones that enable better oversight of Medicaid telehealth utilization and quality.

1 CMS selected the six states (Arizona, California, Maine, Mississippi, Missouri, Tennessee) based on variation in geography, Medicaid program size, and percentage of population living in rural areas. CMS determined that data from Tennessee was not sufficiently reliable for analytic purposes, so data from that state was excluded from various analyses in the report.