Introduction: In recent decades, depression and anxiety disorders have increased among adolescents. The recent COVID-19 pandemic had an impact on the total population either directly through infections and the associated threat of infection or indirectly through the countermeasures taken to limit the spread of the disease. Reviews indicate a deterioration of mental health during the COVID-19 pandemic, specifically among adolescents, yet fail to link this to pre-existing trends. Building on these findings, this meta-analysis aims to investigate if and how the COVID-19 pandemic has impacted the long-term prevalence of depressive and anxiety symptoms among adolescents. Methods: A systematic search was conducted using PubMed/Medline, PsycINFO, and Embase. A multilevel meta-analysis used continuous and dichotomous data extracted from the studies to examine longitudinal trends. Results: The search yielded 4,750 unique articles. After full-text screening, nine articles were included. Most studies found increasing rates of anxiety and/or depression symptoms among the youth before the pandemic (7 out of 9). The patterns found during the pandemic were diverse: in five studies, pre-existing trends in depressive and anxiety symptoms intensified, two studies reported a similar increase in symptoms (in line with pre-existing trends), and two found no increase during the pandemic. The meta-analyses demonstrated that depression and anxiety symptoms during the pandemic were not significantly different from (already rising) levels before the COVID-19 pandemic. Substantial variance was also observed at the measurement and study levels, highlighting extensive heterogeneity behind this estimate, both across and within studies. Conclusion: Although findings show that a worsening of existing trends during the pandemic was not universally visible, the results of most studies do support this hypothesis. More longitudinal studies based on random general population samples are needed to effectively monitor youth mental health trends and to ascertain the true impact of crises.
Bosmans et al. (Sun,) studied this question.