PC2089: Predictors of Adolescent Outpatient Mental Health Service Utilization During the COVID-19 Pandemic

Background: During the COVID-19 pandemic, depression and anxiety symptoms rose for adolescents and delivery capacity of mental health services for adolescents experienced added hurdles with social isolation and school cancellations. Less clear is the predictive role of sociodemographic factors, including family structure, play in mental health service (MHS) utilization during the COVID-19 pandemic. Therefore, our aim is to determine the relationship between family structure and other sociodemographic factors and adolescent MHS utilization.

Methods: Our retrospective cohort study included adolescents aged 12-17 years with a mental health diagnosis as identified by ICD-10 codes in the electronic medical record (anxiety, mood disorder/depression, attention-deficit/hyperactivity, anxiety and mood disorder/depression, and psychosis and multiple diagnoses) and enrolled in Kaiser Permanente Mid-Atlantic States (KPMAS), a comprehensive integrated health system. We analyzed differences in KPMAS MHS utilization before (March 1, 2019 – February 28, 2020) and during (March 1, 2020 – February 28, 2021) the COVID-19 pandemic in Maryland, D.C., and Virginia. We used logistic regression models with an interaction term for the COVID-19 pandemic year to determine the relationship between family structure type (two-parent, single-parent, extended family, and no parent households) and adolescent MHS utilization defined as at least one outpatient behavioral health visit within the measurement year, while adjusting for age, gender, race, state of residence, chronic medical condition (a physical illness lasting >12 months), and mental health diagnosis. Generalized estimating equations were used to analyze repeated measurement data.

Results: Among the 7035 adolescents included in our analysis (3811 female [54.2%]; median age of 14.6 years), 4563 (64.9%) lived in a two-parent household. Odds of adolescents accessing behavioral health visits were 23% higher during the COVID-19 pandemic compared to the pre-COVID-19 timeframe (OR 1.23, 95% CI: 1.14 -1.33, p< 0.01). Adolescents from two-parent and single-parent households increased MHS utilization during the COVID-19 pandemic compared to the pre-pandemic year (McNemar’s χ2=24.8, p < 0.01; χ2=5.3, p = 0.02, respectively); however, in the fully adjusted model family structure was not a significant predictor of MHS utilization. Having a chronic medical condition was associated with higher odds of utilizing MHS (aOR=1.16; 95% CI: 1.07 – 1.25, p < 0.01). White/Caucasian adolescents had greater odds of MHS utilization compared to all racial/ethnic minorities. Adolescents with psychosis or multiple mental health diagnoses (aOR=4.17, 95%CI: 3.62 - 4.81, p < 0.01) were more likely to utilize MHS compared to adolescents with anxiety.

Conclusion: Two-parent and single parent households experienced a significant rise in MHS utilization during the COVID-19 pandemic with varying levels of increase for extended family and no parent households. Racial identity, chronic medical condition status, and mental health diagnoses served as predictors of adolescent MHS utilization. Knowledge of subpopulations with higher odds of utilization can inform prevention programming for adolescent well-being.