Learning Modalities During COVID-19 and Adolescent Mental Health in North Carolina

The mental health of youth represents a critical public health issue and has worsened over the course of the pandemic, partially attributed to social isolation and school closures.

This analysis leveraged data on school learning modality (COVID-19 School Data Hub), emergency department (ED) visits (North Carolina SHEPS data), and school-district characteristics (National Center for Education Statistics) to investigate how school learning modality predicts mental health-related ED visitation in the state of North Carolina. Generalized linear mixed models were employed to quantify how the odds of mental health ED visitation varied with school learning modality. Models adjust for school-level covariates (e.g., broadband, student:teacher ratio) and patient-level characteristics (age, sex). Patient and school data are nested within counties, which were included as a random effect to address potential clustering within the data.

In-person (odds ratios [OR]: 0.92, 95% CI: 0.85-0.99) and hybrid (odds ratios: 0.95, 95% CI: 0.93-0.98) modalities were associated with lower ORs for ED visits due to mental health compared to a virtual learning modality.

School districts should consider increasing the provision of mental health services for students during circumstances in which a virtual learning modality is utilized and consider adopting hybrid learning rather than virtual learning when feasible.

School administrators should seek to balance students' mental health and learning needs in times of crisis.