Abstracts

Rationale: Following the declaration of the COVID-19 pandemic state of emergency in March 2020, the NYU Langone Health system quickly transitioned to telehealth services for all but essential outpatient visits. Telehealth visits were performed using a HIPAA-compliant smartphone-based video application or telephone. However, it is unknown if all patients had equal access to telehealth services. We examined differences between patients with epilepsy or seizures who completed their visits via telephone and video visits over a 2-month period at the start of the pandemic (April and May 2020) and patients who completed in-person office visits over the same 2-month period the year prior (April and May 2019).

Methods: We identified all pediatric and adult outpatient visits for epilepsy or seizures (ICD10 codes G40.* or R56.9) at all NYU Comprehensive Epilepsy Center locations via query of the electronic medical record in April/May 2019 and April/May 2020. We extracted demographic information (age, sex, race, ethnicity), ZIP code, insurance type, and visit type (office, telephone and video). Using patient ZIP code of residence, we also identified community information for each patient from the American Community Survey 2014-2019 (socialexplorer.com) including proportion of population with broadband internet access and median household income quintile for the New York City metropolitan area. We performed a multinomial logistic regression to examine independent factors associated with a visit type (office, phone, video). This study was approved by the NYUGSOM IRB.

Results: There were 1975 unique patients who had outpatient office visits in April/May 2019. There were 1918 outpatient visits in April/May 2020; 215 (11%) occurred in the office, 1557 (81.2%) were video visits, and 146 (7.6%) were telephone visits. Patients seen in 2020 were younger (median 31, IQR 19-48 years) compared to 2019 (34, IQR 22-53 years, p< .001, Mann-Whitney U-test). During April/May 2020, patients seen via video were younger than those who had telephone visits (median 29, IQR 17-44, years vs 49, IQR 30-61, years, p< .001). Multinomial logistic regression demonstrated that, when compared to office visits the prior year, younger age (relative risk: 0.98, 95% CI 0.97 – 0.99) and higher community rates of broadband access (relative risk: 14.3 , 95% CI 1.1-190) were associated with a video visit. Older age (relative risk: 1.02, 95% CI 1.01-1.03) and lower levels of community broadband access (relative risk: 0.002, 95% CI 0.00016 – 0.033) were associated with a telephone visit.