Self-reported mental distress in the United States:a Bayesian analysis of the spatial structure over the COVID-19 pandemic across age groups

Background The COVID-19 had an outstanding impact on well-being and mental health, which might have elicited geographical variations over time. This study examines the eventual impact of COVID-19 on self-reported mental distress in the mainland USA. Aims There were two main aims. First, to evaluate the pre-pandemic (2019) and post-pandemic (2021) mental distress spatial distribution. Second, to contrast spatial data across three age groups, young (18-44 years), middle-aged (45-65 years), and old (older than 65 years). Method We considered a the Bayesian modified Besag–York–Molli`e (BYM2) model, which is a Bayesian hierarchical model. Mental distress was the response variable function of age group, year and spatially structured and unstructured effects. Results The main findings indicate a positive spatial dependence between states of general mental distress before and after the COVID-19 and across age groups with substantial unstructured component. Moreover, younger individuals reported higher levels of mental distress and suffered the major worsening due to the pandemic. Conclusions COVID-19 had a detrimental impact on mental health across the population, with consistent evidence of positive spatial dependence across states. Notably, young adults emerged as particularly vulnerable, exhibiting concerning levels of mental distress problems and being more sensitive to the effects of the pandemic.