Impact of COVID-19 on Mental Health: A Longitudinal Study Using Penalized Logistic Regression

Cui, J.; Lu, J.; Weng, Y.; Yi, G. Y.; He, W.

The COVID-19 pandemic has posed significant influence on the public mental health in a stealthy manner. Current efforts focus on alleviating the impacts of the disease on public health and economy, with the psychological effects due to COVID-19 largely ignored. In this paper, we analyze a mental health related dataset from the US to enhance our understanding of human reactions to the pandemic. We are particularly interested in providing quantitative characterization of the pandemic impact on the public mental health, on top of qualitative explorations. We employ the multiple imputation by chained equations (MICE) method to deal with missing values and take the logistic regression with least absolute shrinkage and selection operator (Lasso) method to identify risk factors for mental health. The analyses are conducted to a large scale of online survey data from 12 consecutive weeks, so that the longitudinal trend of the risk factors can be investigated. Our analysis results unveil evidence-based findings to identify the groups who are psychologically vulnerable to COVID-19. This study is useful to assist healthcare providers and policy makers to take steps for mitigating the pandemic effects on public mental health.