LAUSR

BACKGROUND The burden of mental health disorders is high, especially in developing countries. We assessed the prevalence of depressive and anxiety symptoms and their relationship with life-threatening events, tobacco dependence, and hazardous alcohol drinking in the Brazilian Amazon. METHODS Cross-sectional population-based study conducted in Manaus in 2019 with adults selected by probabilistic sampling. Depressive symptoms were measured by the PHQ-9 instrument (cut-off ≥9) and anxiety symptoms by the GAD-7 scale (cut-off ≥10). Prevalence ratios (PRs) of depressive and anxiety symptoms were calculated by Poisson regression with robust variance with 95% confidence intervals (CI) following a hierarchical model. Partial least squares structural equation modeling was used to investigate the relationship between the outcomes and risk behaviors. RESULTS Out of the 2,321 participants, 24.3% (95%CI 22.2-26.5%) had depressive and 21.6% (95%CI 19.6-23.7%) had anxiety symptoms. Depressive symptoms were associated with women (PR=1.32; 95%CI1.08-1.61), lower social class (PR=1.59; 95%CI1.11-2.27), life-threatening events (PR=2.66; 95%CI2.00-3.54), tobacco dependence (PR=1.84; 95%CI1.37-2.47), worse health statuses (p<0.001), and chronic diseases (PR=1.63; 95%CI1.33-2.00), but were lower in older adults (p=0.014). Anxiety symptoms were associated with women (PR=1.74; 95%CI1.42-2.14), lower educational levels (PR=2.19; 95%CI1.38-3.47), evangelical individuals (PR=1.28; 95%CI1.05-1.57), having no religion (PR=1.72; 95%CI1.24-2.38), life-threatening events (PR=3.26; 95%CI2.41-4.41), tobacco dependence (PR=1.53; 95%CI1.09-2.16), worse health statuses (p<0.001), and chronic diseases (PR=1.77; 95%CI1.40-2.25). Depressive symptoms, anxiety symptoms, and life-threatening events were directly correlated with one another, while tobacco dependence and hazardous alcohol drinking were significantly intercorrelated (p<0.05). LIMITATIONS Cross-sectional design limits the assessment of causality. Recall bias was possible as responses were self-reported. GAD-7 scale was not validated in the Brazilian population. CONCLUSION Nearly a quarter of the population had depressive symptoms and one-fifth presented anxiety symptoms, which were associated with socioeconomic, behavioral, and health-related factors. Implementation of social well-being policies is required to minimize the burden of mental health disorders in the Amazonian population.