Use of Telehealth Across Pediatric Subspecialties Before and During the COVID-19 Pandemic.

Importance: The identification of variation in health care is important for quality improvement. Little is known about how different pediatric subspecialties are using telehealth and what is driving variation. Objective: To characterize trends in telehealth use before and during the COVID-19 pandemic across pediatric subspecialties and the association of delivery change with no-show rates and access disparities. Design, Setting, and Participants: In this cohort study, 8 large pediatric medical groups in California collaborated to share aggregate data on telehealth use for 11 pediatric subspecialties from January 1, 2019, to December 31, 2021. Main Outcomes and Measures: Monthly in-person and telehealth visits for 11 subspecialties, characteristics of patients participating in in-person and telehealth visits, and no-show rates. Monthly use rates per 1000 unique patients were calculated. To assess changes in no-show rates, a series of linear regression models that included fixed effects for medical groups and calendar month were used. The demographic characteristics of patients served in person during the prepandemic period were compared with those of patients who received in-person and telehealth care during the pandemic period. Results: In 2019, participating medical groups conducted 1.8 million visits with 549 306 unique patients younger than 18 years (228 120 [41.5%] White and 277 167 [50.5%] not Hispanic). A total of 72 928 patients (13.3%) preferred a language other than English, and 250 329 (45.6%) had Medicaid. In specialties with lower telehealth use (cardiology, orthopedics, urology, nephrology, and dermatology), telehealth visits ranged from 6% to 29% of total visits from May 1, 2020, to April 30, 2021. In specialties with higher telehealth use (genetics, behavioral health, pulmonology, endocrinology, gastroenterology, and neurology), telehealth constituted 38.8% to 73.0% of total visits. From the prepandemic to the pandemic periods, no-show rates slightly increased for lower-telehealth-use subspecialties (9.2% to 9.4%) and higher-telehealth-use subspecialties (13.0% to 15.3%), but adjusted differences (comparing lower-use and higher-use subspecialties) in changes were not statistically significant (difference, 2.5 percentage points; 95% CI, -1.2 to 6.3 percentage points; P = .15). Patients who preferred a language other than English constituted 6140 in-person visits (22.2%) vs 2707 telehealth visits (11.4%) in neurology (P < .001). Conclusions and Relevance: There was high variability in adoption of telehealth across subspecialties and in patterns of use over time. The documentation of variation in telehealth adoption can inform evolving telehealth policy for pediatric patients, including the appropriateness of telehealth for different patient needs and areas where additional tools are needed to promote appropriate use.