Telemedicine has not significantly boosted healthcare costs, study finds

New UCLA-led research finds that use of telemedicine has not significantly increased visits and medical spending across all payer types.

The findings, to be published May 11 in the peer-reviewed journal JAMA Network Open, could ease concerns among lawmakers that the telemedicine expansion that occurred during the COVID pandemic would result in large utilization and spending increases.

With the declaration of the COVID pandemic in 2020, the Centers for Medicare & Medicaid Services (CMS) changed key policies regarding telemedicine flexibility, such as introducing payment parity with in-person visits, waiving geographic restrictions, and eliminating out-of-pocket cost sharing. Once the pandemic was declared over, however, lawmakers extended these changes to analyze how telemedicine impacted healthcare use and spending. Those CMS flexibilities are due to expire in 2027, and lawmakers continue to debate whether to permanently extend or modify them.

While supporters had hoped that the new, wider availability of telemedicine would reach people without easy access to doctors, others were concerned that the new flexibility would increase its use and drive up healthcare costs, said study lead Dr. John N. Mafi, associate professor-in-residence of medicine, division of general internal medicine and health services research at the David Geffen School of Medicine at UCLA.

Our findings suggest neither prediction came true on a national scale," Mafi said. "As telemedicine use grew, visits and spending in heavy users tracked closely with patterns in lighter users. That is reassuring for anyone worried about ballooning costs, but more sobering for anyone hoping telemedicine would close longstanding gaps in access. At least so far, it looks more like a substitute for in-person care than a true expansion of it."

Dr. John N. Mafi, associate professor-in-residence of medicine, division of general internal medicine and health services research, David Geffen School of Medicine, UCLA

Previous research has raised the possibility that telemedicine could worsen health disparities, and others have found geographic differences in telemedicine use. In addition, a previous study published in JAMA and led by UCLA found that Americans' use of common outpatient health services dipped sharply at the outset of the COVID-19 pandemic, then rebounded to near-normal levels by the end of 2020, only to decline again during the second surge in January–February 2021. This study prompted the researchers to examine the effect of telemedicine's adoption among different population groups.

With these questions in mind, the researchers sought to quantify the association between telemedicine use, visits and health spending. They used multi-payer medical claims data from MedInsight's database for more than 3 million US adults who were continuously enrolled in Medicare fee-for-service, Medicare Advantage, dual-eligible Medicaid or commercial insurance from January 1, 2019 through December 31, 2023.

Overall, the researchers found that telemedicine visits fell 2.4% and spending dropped 0.5%. However, the findings "crossed the null," meaning that these changes were not statistically significant over the time period covered. In other words, the authors cannot rule out that the changes observed could simply be due to chance. Importantly, while the confidence intervals could not rule out small changes in either direction, they did rule out large ones, indicating that any true effect on overall visits or spending was modest at most.

They also did not find any significant changes across the subgroups they examined. For example, urban populations had 4.4% fewer visits and 2.3% lower spending, 2.5% lower spending among Medicaid-insured people, 5.3% lower spending among dual-eligible individuals, 3% less spending among those with Medicare Advantage, and 1.5% lower spending among socially vulnerable populations, but none of these changes were statistically significant.

They also found there were 3.4% more visits and 3.8% higher spending in rural areas, 1.1% more spending among commercially insured patients, 1% more expenses among people with Medicare fee-for-service insurance, and 4.5% higher spending among people who were the least socially vulnerable, but again, none of these changes were statistically significant.

Among the study's limitations, aggregated results may not apply to individuals, the results may not apply to the entire country or to people who lack insurance, and the study design could not lead to any causal inferences.

Senior author Dr. Katherine Kahn, distinguished professor of medicine at the David Geffen School of Medicine at UCLA and senior natural scientist at RAND, said the findings should be viewed as an early read on a still-evolving policy.

"Our analysis runs only through late 2023, when telemedicine use was still settling into a new equilibrium," Kahn said. "Much more work is needed to understand telemedicine's longer-term effects on quality of care, health outcomes, and spending, and whether those effects differ across the diverse populations who depend on it. Policymakers should keep monitoring closely as the evidence base matures."

Study co-authors are Sitaram Vangala, Manying Cui, Artem Romanov, Ziyi Li, Chi-Hong Tseng, and Dr. Catherine Sarkisian of UCLA; Jonathan Cantor and Cheryl Damberg of RAND; Melody Craff, Dale Skinner, and Michael Hadfield of MedInsight, Milliman Inc.; Michelle Rockwell of Virginia Tech, and Dr. A. Mark Fendrick of University of Michigan. Mafi and Kahn are also affiliated with RAND

Arnold Ventures funded this study with additional support from the National Institutes of Health and a National Institute on Aging research Career Development award (K76AG064392-01A1).