Abstract
PURPOSE: The impact of the COVID-19 pandemic on adolescent mental health is a global concern; however, most research is cross-sectional or started after the pandemic response began and thus unable to evaluate within-individual change. The purpose of this prospective study was to evaluate the effect of the initial COVID-19 response on adolescent mental health and ill-health as a natural experiment. METHODS: We used 3-year linked data from the COMPASS study, including 7,653 Canadian (Quebec, Ontario) adolescents from which 2,099 completed surveys in all three waves (pre-COVID-19 [2018 and 2019] and online [May-July 2020], 2-3 months into the pandemic). A structural equation modeling approach to fixed effects and a difference-in-differences design were used to estimate pre-COVID-19-to-early lockdown change in mental health (psychosocial well-being [flourishing-reverse scored]) and ill-health (depression and anxiety symptoms), compared with 2018-to-2019 change. Models were adjusted for self-selection, age of entry into the cohort, and sociodemographics. RESULTS: Depression, anxiety, and reverse-flourishing scores increased across all waves; however, the mental health changes from the pre-COVID-19 wave (2019) to 2020 were not greater relative to the changes seen across the 2018-to-2019 waves. CONCLUSIONS: Our results do not support a detrimental effect of the initial stages of the COVID-19 lockdown measures on adolescent mental health. The deterioration in mental health in the early COVID-19 response was less than the decline found over a prepandemic period. Further prospective research is needed to explore the impact of the prolonged pandemic and related measures on adolescents and inequitable effects in population subgroups.