(PS5-82) The Effect of the COVID-19 Pandemic on Length of Care in a Diverse Population of Adolescents

Title: The Effect of the COVID-19 Pandemic on Service Utilization in a Diverse Population of Adolescents

Rates of mental health adversities in adolescents increased during the COVID-19 pandemic, with a widespread shift to telehealth service delivery leading to concerns that access to mental health services may have inequitably decreased for racial and socioeconomic minority groups. As such, the current investigation explored how the COVID-19 pandemic affected rates of service utilization in nineteen School-Based Health Clinics (SBHC’s) across NYU Langone’s Family Health Center system in Brooklyn.

A secondary data analysis was performed on a deidentified data set including a sample of n=2022 students (41.2% female; M age = 14; 39.6% Black, 3.51% Asian, 5.5% Native American, 2.1% Native Hawaiian/Pacific Islander; 40.5% Hispanic) who used behavioral health services at a SBHC location between 01/03/2017 and 08/29/2022.Variables included in analyses were self-reported demographics (gender, age, ethnicity, race, preferred language), length of care (measured in days from intake date to last behavioral health visit date), insurance type (proxy for socioeconomic status (SES)), intake (before vs after start of pandemic), and treatment location. Multi-way analyses of variance (ANOVAs) were conducted to evaluate differences in length of care between demographic variables. A backwards stepwise regression was also performed with each sociodemographic variable as a predictor of length of care.

Black patients were found to have significantly shorter average length of care (301.9 days) from their White counterparts (451.5 d.) (p < 0.001). Hispanic individuals had a significantly longer length of care (475.9 d.) than non-Hispanic individuals (343.9 d.)(p < 0.001). Those with an intake before the mandated shutdown date in NYC schools of 3/6/2020 had significantly longer average length in care than those with an intake after the shutdown. Significant differences in length of care across locations and age were found, as well. When considered as predictors in a regression, 21 variables were ultimately included in the model (13 excluded, 8 reference), which had an adjusted R² value of 0.34. Variables that predicted a shorter Length in Care included: “Asian” race (p < 0.052), “Native Hawaiian/Pacific Islander” race (p=0.017), “Other” race (p < 0.001); intake after pandemic start (p < 0.001); “Other” preferred language (p < 0.061) and 3 treatment locations (all p≤0.05, 2 with p< 0.001). Variables that predicted a longer Length in Care included: “Unknown” race (p < 0.052); “Hispanic” ethnicity (p < 0.001); “Self-Pay” insurance type (p < 0.001); Age ranges 5-10, 11-16 (p < 0.001) and 8 Locations (all p≤0.05, 3 p < 0.001).

Significant differences in length of care between Racial, Ethnic and intake categories were found, though location was ultimately found to be the best predictor of length of care. With a longer length being viewed as greater engagement with services, this may indicate differences in effectiveness of clinicians or programs between SBHC’s or varying levels of need across sites. Historical redlining and environmental racism can also be considered when interpreting this result.