The Psychological Effects of Quarantine During COVID-19 Outbreak: Sentiment Analysis of Social Media Data

Background: We rely on social distancing measures such as quarantine and isolation to contain the novel coronavirus (COVID-19). However, the negative psychological effects of these measures are non-negligible. Whilst we have well delved into the psychological effects after quarantine in previous research, we know little about its effects amid COVID-19. Methods: We adopt a sentiment analysis approach to analyze the psychological state changes of 1,278 quarantined persons’ 214,874 tweets over four weeks spanning the period before, during, and after quarantine. We formed another group of 1,278 unquarantined persons with 250,198 tweets as the control group. The tweets of both groups are analyzed by matching with a tailor-made lexicon to measure the anxious depression level changes over time. Findings: We discovered a clear pattern of psychological changes for quarantined persons. Anxious depression levels significantly increased right after quarantine, but gradually diminished as the quarantine progresses. However, anxious depression levels resurged after 14 days’ quarantine, as people are still locked down indoor. Interpretation: Whilst quarantine is deemed necessary, proper interventions should be introduced to mitigate its adverse psychological impacts. Funding Statement: No funding source for this research. Declaration of Interests: There is no conflict of interests in this research. Ethics Approval Statement: The ethics committee of the University of Hong Kong has approved the ethics application of this research (reference number: EA2005004).