Mental Health Outcomes After the Acute Phase of COVID-19: Factors Identified in a Korean Public Mental Health Service

Abstract

Background: Sequelae following coronavirus disease 2019 (COVID-19) frequently include lasting neuropsychiatric symptoms; however, identifying predictors from the acute phase remains challenging due to limitations in collecting prospective data during that time. Methods: Data were obtained from electronic counseling records of 515 COVID-19 patients who received free services from the Korean National Center for Disaster and Trauma. The Clinical Global Impression Severity Scale (CGI-S) was used to repeatedly assess mental health at each counseling session. Baseline predictors included demographic characteristics, psychiatric and medical comorbidities, and psychological response, which was further divided into four sub-factors via factor analysis. Multivariate mixed effect models were used to explore the relationship between these predictors and mental health following the acute phase of COVID-19, with analyses stratified by gender. Results: The most common post-COVID psychological responses were anxiety, depression, and sleep problems, with CGI-S scores dropping from 2.83 initially to 1 by the last observed session and averaging 2.38 at the third follow-up. Four sub-factors were identified through exploratory factor analysis, namely cognitive and physical exhaustion, emotional distress, self-destructive coping, and somatized anxiety. Baseline psychological responses (beta = 0.06, P < 0.001) and pre-existing psychiatric disorders (beta = 0.37, P < 0.001) were significantly associated with higher CGI-S scores over time; among psychological sub-factors, cognitive-physical exhaustion (beta = 0.28, P < 0.001), emotional distress (beta = 0.32, P < 0.001), and self-destructive coping (beta = 0.12, P < 0.001) were significant predictors, with emotional distress significant in men (beta = 0.26, P = 0.001) and both cognitive-physical exhaustion (beta = 0.36, P < 0.001) and emotional distress (beta = 0.36, P < 0.001) significant in women. Conclusion: Baseline psychological responses predict persistent mental health symptoms, and identified profiles may help early identification of high-risk groups during acute COVID-19.