S1 Data -

Aims
This study aims to assess the dynamics of in-school adolescents’ mental health problems in Harari regional state, eastern Ethiopia for a year.
Materials and methods
Using multistage sampling technique, we conducted a year-long longitudinal study at three public high schools between March 2020 and 2021. Three hundred fifty-eight in-school adolescents were chosen by systematic random sampling for the baseline assessment, and 328 completed the follow-up assessment. We used self-administered, adolescent version of SDQ-25 Questionnaire to collect the data. Wilcoxon matched-pairs signed-rank test and McNemara’s Chi-squared tests were used to examine the median difference and distribution of mental health problems between times one and two. Random-effects logistic regressions on panel data was used to identify factors associated with mental health problems. A p-value < 0.05 was considered as statistically significant.
Results
The magnitude of overall mental health problems at baseline assessment was 20.11% (95% CI: 16–25), with internalizing problems accounting for 27.14% (95% CI: 23–32) and externalizing problems accounting for 7.01% (95% CI: 4.6–10.3). At the follow-up assessment, these proportions rose to 22.56% (95% CI, 18–27) for overall problems and 10.3% (95% CI, 7.7–14.45) for externalizing problems. On other hand, internalizing problems decreased unexpectedly to 22.86% (95% CI, 18.6–27.7) at follow-up assessment. Internalizing problem scores at time two were significantly lower than baseline among older adolescents, girls and those with average wealth index in our study cohort.
Conclusions
The prevalence of mental health problems were high among the study cohort. The proportion of overall problems and externalizing problems has increased over time, indicating a deterioration in the mental health of the study cohort. However, the decrease in internalizing problems among older adolescents, girls, and those with an average wealth index is a positive sign. The findings highlight that tailored interventions are required to reduce externalizing problems and maintain the decrease in internalizing problems. These interventions should target middle-aged and male adolescents from low-income families.