To evaluate the impact on health care access of the change in telemedicine delivery from a clinic-based model, in which patients connect with their healthcare provider from local telemedicine clinics, to a home-based model, in which patients independently connect from their homes.

In this retrospective analysis, we compared relative uptake in telemedicine services in Period 1 (01/01/2019 to 03/15/2020, prepandemic, clinic-based model) vs Period 2 (03/16/2020 to 06/30/2022, home-based model) within a tertiary pediatric hospital system. Using multivariable logistic regression, we investigated the influence of telemedicine delivery model on patient sociodemographic characteristics of completed telemedicine visits.

We analyzed 400 539 patients with 1 406 961 completed outpatient encounters (52% White, 35% Black), of which 62 920 (4.5%) were telemedicine. In the clinic-based model (Period 1), underserved populations had greater likelihoods of accessing telemedicine: Hispanic ethnicity (OR = 1.41, P = .028) vs reference group non-Hispanic, Medicaid (OR = 2.62, P < .001) vs private insurance, and low-income neighborhood (OR = 3.40, P < .001) vs medium-income. In aggregate, telemedicine utilization rapidly increased from Period 1 (1.5 encounters/day) to Period 2 (107.9 encounters/day). However, underserved populations saw less relative increase (Medicaid [OR = 0.28, P < .001], Hispanic [OR = 0.53, P < .001], low-income [OR = 0.23, P < .001]).

We observe that the clinic-based model offers more equitable access, while the home-based model offers more absolute access, suggesting that a hybrid model that offers both home-based and clinic-based services may result in more absolute and equitable access to telemedicine.