Depressão infantil em contextos de catástrofe: avaliação de um programa de intervenção com crianças afetadas pela pandemia de covid-19

Abstract

The COVID-19 pandemic constituted a global health catastrophe, whose psychosocial impacts disproportionately affected vulnerable populations, including children. Among the mental disorders that gained prominence during the pandemic period, childhood depression holds a relevant place, with consequences that can extend throughout development. This doctoral thesis aimed to investigate the effects of a psychosocial intervention in reducing depression symptoms among children exposed to catastrophe contexts, taking the COVID-19 pandemic as a reference. To this end, three interrelated studies were conducted. The first study evaluated the psychometric properties of the Children's Depression Inventory (CDI) in its 20-item version. The results indicated high internal consistency for the total score (α = 0.88) and for the affective (α = 0.79) and cognitive (α = 0.74) subscales, but unsatisfactory consistency in the behavioral (α = 0.52) and somatic (α = 0.49) subscales. Confirmatory factor analysis demonstrated adequate fit of the four-factor model (CFI = 0.99; TLI = 0.99; RMSEA = 0.000). The second study, with a quasi-experimental design including control and experimental groups and pre- and post-intervention measures, examined the efficacy of the ConViver program in reducing depressive symptoms. The results showed a significant reduction in symptoms over time for the entire sample (F(1,114) = 11.83; p = 0.0008), especially in the affective and cognitive dimensions, but without significant differences between the experimental and control groups. Stratified analyses revealed that girls in both the experimental and control groups showed greater symptom reduction and that non-Black children in the experimental group benefited more from the intervention than Black children. The third study, whose aim was to conduct a cluster analysis based on the change score identified in depression symptoms before and after the intervention (ΔCDI = T2 − T1), identified three trajectory profiles: Improved (n = 47), Stable (n = 46), and Worsened (n = 23). There was no significant association between allocation group and cluster membership (p = 0.641; Cramer's V = 0.00), indicating that the program did not significantly alter the normative trajectory of depressive symptoms. Multinomial logistic regression revealed that male gender was the only significant predictor of a lower chance of improvement (OR = 0.29; 95% CI: 0.12–0.73). Taken together, the findings indicate that the CDI is an adequate instrument for screening depressive symptoms in Brazilian children exposed to catastrophes, although its behavioral and somatic subscales require revision. The ConViver program, while not proving effective for reducing depressive symptoms, fulfills a relevant role as a mental health promotion and prevention intervention, acting on protective factors such as positive affect and support networks. The 13 improvement observed in some children appears to be explained by the normative course of the disorder and by the beneficial effects of school reopening and the resumption of social bonds in the post-pandemic period. The results reinforce the need for specialized interventions for established depressive conditions and point to the importance of an intersectional perspective in the implementation of psychosocial programs.