The study team reported a measurable decline in telehealth utilization among US adults between 2022 and 2024. Population-weighted prevalence fell from 39.3% in 2022 to 34.5% in 2024. After adjusting for sociodemographic and health-related covariates, the odds of any telehealth use were lower in 2024 compared with 2022 (adjusted odds ratio, 0.81; 95% CI, 0.69–0.95), the investigators noted. Simple and direct: fewer adults reported using telehealth after the public health emergency ended. The data indicate a contraction from pandemic-era volumes, rather than a persistent ceiling at the higher levels seen during the emergency.
Use was not uniform across demographic or clinical subgroups. The study team observed higher adjusted odds of telehealth use among adults aged 35 to 49 years and among those with a college degree or higher. Insured adults also had greater odds of use than uninsured adults. Clinical history mattered: respondents reporting diagnosed depression or anxiety had substantially higher odds of telehealth use, and the study team reported elevated odds among adults with a history of cancer. These patterns suggest that telehealth continued to reach some groups with ongoing need, while other groups used it less often. The preprint did not report additional subgroup breakdowns beyond the specified characteristics.
Respondents who had used telehealth were asked their reasons for doing so; the reported motivations shifted markedly between 2022 and 2024. The proportion citing infection avoidance dropped steeply, from 50.2% to 19.1%. Provider recommendation or requirement decreased as a stated reason (72.8% to 47.2%). By contrast, convenience showed a smaller decline—from 65.6% to 56.2%—and became the most frequently reported reason for telehealth use in 2024, the authors reported. In short, pandemic-driven drivers diminished rapidly; convenience persisted as a durable rationale for integration into routine care.
The investigators used nationally representative data from the Health Information National Trends Survey (HINTS) collected in 2022 and 2024. The combined analytic sample comprised 13,141 US adults across the two survey waves. The team employed survey-weighted analyses to estimate population prevalence and proportions for reasons for telehealth use. For multivariable modeling, survey-weighted logistic regression was used to compare telehealth odds between cycles and to assess sociodemographic and health-related correlates of use. The authors applied multiple imputation techniques to address missing data. The preprint states that all data are available online via the HINTS website. The source did not report additional details about the imputation model specifications or variable-by-variable missingness in the text of the abstract.
The report is a preprint and has not undergone peer review, and the authors explicitly caution against using these findings to guide clinical practice until the work is evaluated. The analysis derives from serial cross-sectional survey data; the design supports population-level trend estimates but does not permit inference about individual-level longitudinal change. The abstract does not list whether mode-specific telehealth types (video vs telephone) were disaggregated in the reported results, nor does it provide granular timing or frequency of telehealth encounters. The preprint describes use of multiple imputation and survey-weighted regression but does not include, in the abstract, imputation model specifics or sensitivity analyses.
Key unanswered questions remain in the source material: how will payer policy and clinician workflow changes after the public health emergency affect future utilization; which implementation strategies most effectively reduce the observed inequities in access; and whether changes in telehealth modality or quality accompany the shifts in reported reasons for use. The study team noted that the data are publicly available through HINTS, which enables researchers to examine these and other open questions further, but concrete next regulatory milestones or policy changes were not detailed in the preprint.
Data availability: the authors reported that all study data are available on the HINTS website. The authors declared no competing interests. The abstract cautions readers that the manuscript has not been peer reviewed and should not yet be used to guide clinical practice.