Telehealth use stabilized after its pandemic peak, with lower visit payments

Telehealth did not vanish when the acute phase of the COVID-19 pandemic receded. A new analysis of nationally representative United States data shows that virtual office-based visits settled into a stable, persistent role in American ambulatory care between 2020 and 2023, with behavioral health emerging as the modality’s clear stronghold. The study, published in Cost Effectiveness and Resource Allocation, also finds that telehealth encounters carried lower adjusted expected payments than in-person visits, while cautioning that this gap cannot be read as proof of genuine cost savings.

The researchers drew on the Medical Expenditure Panel Survey, a large federal survey designed to capture how Americans use and pay for health care. Their pooled cross-sectional analysis covered 543,537 office-based visit events from 2020 through 2023, with all statistical work accounting for the survey’s complex design so that estimates reflect the national population rather than the raw sample. Insurance status was sorted into seven groups: employer-sponsored insurance, Medicare only, Medicaid or SCHIP only, Medicare-Medicaid dual coverage, other private coverage, uninsured, and other or unknown coverage.

The headline utilization finding is one of stabilization rather than collapse. Survey-weighted telehealth prevalence stood at 10.87 percent of office-based visits in 2020, dipped to 8.82 percent in 2021, and then held nearly flat at 9.15 percent in 2022 and 9.20 percent in 2023. In other words, after an early pandemic surge and a modest retreat, roughly one in eleven office-based encounters continued to happen remotely through the end of the study window.

Where telehealth persisted is as telling as how often. The modality was heavily concentrated in psychotherapy and mental-health counseling encounters, which accounted for 48.78 percent of telehealth visits, and in behavioral-health provider encounters more broadly, at 46.80 percent. That concentration suggests virtual care has become embedded in a specific clinical niche where remote delivery appears to fit the nature of the service, rather than diffusing evenly across all of office-based medicine.

Utilization patterns also diverged sharply by insurance type. In survey-weighted logistic regression models with a year-by-insurance interaction, telehealth trends differed significantly across coverage groups. Compared with people holding employer-sponsored insurance, those with Medicare-only coverage had a 2.69-percentage-point lower adjusted probability of using telehealth, with a 95 percent confidence interval of -4.50 to -0.89 and a p-value of .004.

On the economic side, the team treated visit-level payments as the primary outcome and analyzed them with a two-part model, a standard approach for health care cost data in which many encounters generate no payment and spending among users is highly skewed. Billed charges were retained only as a secondary descriptive measure. In the corrected two-part model, adjusted expected payments came to $208.44 per telehealth visit, with a 95 percent confidence interval of $189.73 to $227.16, compared with $293.68 for in-person visits, with a 95 percent confidence interval of $285.32 to $302.05.

The authors are explicit that this payment differential should not be interpreted as definitive evidence of cost savings. The types and intensity of services delivered may differ systematically between the two modalities, and a cheaper average visit could simply reflect a different case mix rather than true efficiency.

That caveat shapes the study’s forward-looking conclusion. The researchers call for future work incorporating service characteristics, subsequent health care utilization, and patient outcomes to determine the longer-term economic and policy implications of telehealth. Such studies would need to track what happens after a virtual visit, whether it substitutes for an in-person encounter or adds to total utilization, and whether clinical outcomes are preserved.

Its core contributions are descriptive and associative: telehealth remained a meaningful component of office-based care after its 2020 peak, sustained use clustered in behavioral health, utilization patterns differed across insurance groups, and telehealth visits were associated with lower adjusted expected payments. As the authors note, whether those patterns translate into durable value for patients, payers, and the health system remains an open question that this observational analysis was not designed to answer.

Subject of Research: Trends in telehealth utilization and visit-level payments in US office-based care from 2020 to 2023

Article Title: Trends in telehealth utilization and cost of office-based care in the United States: evidence from the Medical Expenditure Panel Survey, 2020–2023

Article References: Olatunde, K. (2026). Trends in telehealth utilization and cost of office-based care in the United States: evidence from the Medical Expenditure Panel Survey, 2020–2023. Cost Effectiveness and Resource Allocation. https://doi.org/10.1186/s12962-026-00856-4

Image Credits: AI Generated

DOI: 10.1186/s12962-026-00856-4

Keywords: telehealth, office-based care, Medical Expenditure Panel Survey, behavioral health, health care costs, insurance coverage, Medicare, Medicaid, health services research, United States, payment differences, COVID-19