Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 Sep 17:16:2111-2118.
doi: 10.2147/NDT.S262004. eCollection 2020.

Predicting Psychological State Among Chinese Undergraduate Students in the COVID-19 Epidemic: A Longitudinal Study Using a Machine Learning

Fenfen Ge #  1 , Di Zhang #  2 , Lianhai Wu  2 , Hongwei Mu  2
Affiliations

Predicting Psychological State Among Chinese Undergraduate Students in the COVID-19 Epidemic: A Longitudinal Study Using a Machine Learning

Fenfen Ge et al. Neuropsychiatr Dis Treat. .

Abstract

Background: The outbreak of the 2019 novel coronavirus disease (COVID-19) not only caused physical abnormalities, but also caused psychological distress, especially for undergraduate students who are facing the pressure of academic study and work. We aimed to explore the prevalence rate of probable anxiety and probable insomnia and to find the risk factors among a longitudinal study of undergraduate students using the approach of machine learning.

Methods: The baseline data (T1) were collected from freshmen who underwent psychological evaluation at two months after entering the university. At T2 stage (February 10th to 13th, 2020), we used a convenience cluster sampling to assess psychological state (probable anxiety was assessed by general anxiety disorder-7 and probable insomnia was assessed by insomnia severity index-7) based on a web survey. We integrated information attained at T1 stage to predict probable anxiety and probable insomnia at T2 stage using a machine learning algorithm (XGBoost).

Results: Finally, we included 2009 students (response rate: 80.36%). The prevalence rate of probable anxiety and probable insomnia was 12.49% and 16.87%, respectively. The XGBoost algorithm predicted 1954 out of 2009 students (translated into 97.3% accuracy) and 1932 out of 2009 students (translated into 96.2% accuracy) who suffered anxiety and insomnia symptoms, respectively. The most relevant variables in predicting probable anxiety included romantic relationship, suicidal ideation, sleep symptoms, and a history of anxiety symptoms. The most relevant variables in predicting probable insomnia included aggression, psychotic experiences, suicidal ideation, and romantic relationship.

Conclusion: Risks for probable anxiety and probable insomnia among undergraduate students can be identified at an individual level by baseline data. Thus, timely psychological intervention for anxiety and insomnia symptoms among undergraduate students is needed considering the above factors.

Keywords: COVID-19; anxiety; cohort; insomnia; machine learning.

PubMed Disclaimer

Conflict of interest statement

Fenfen Ge and Di Zhang are co-first authors. The authors report no conflicts of interest in this work.

Figures

Figure 1
Figure 1
The AUCs for the prediction of probable anxiety A and insomnia B.
Figure 2
Figure 2
Confusion matrix (A) anxiety; (B) insomnia.
Figure 3
Figure 3
Bar graph showing “feature importance” in predicting probable anxiety and insomnia, respectively (A) anxiety; (B) insomnia.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Mahase E, Humphreys K, Cullen MR, Basu S. China coronavirus: WHO declares international emergency as death toll exceeds 200. BMJ. 2020;368:m408. doi:10.1136/bmj.l6968 - DOI - PubMed
    1. Wang C, Horby PW, Hayden FG, Gao GF. A novel coronavirus outbreak of global health concern. Lancet. 2020;395(10223):470–473. doi:10.1016/S0140-6736(20)30185-9 - DOI - PMC - PubMed
    1. Gallagher TH, Schleyer AM. “We signed up for this!” - student and trainee responses to the Covid-19 pandemic. N Engl J Med. 2020;382:e96. doi:10.1056/NEJMp2005234 - DOI - PubMed
    1. GOV.UK. Coronavirus (COVID-19): what you need to do. 2020. Available from: https://www.gov.uk/government/publications/wuhan-novel-coronavirus-backg.... Accessed August22, 2020.
    1. Brooks SK, Webster RK, Smith LE, et al. The psychological impact of quarantine and how to reduce it: rapid review of the evidence. Lancet. 2020;395:912–920. doi:10.1016/S0140-6736(20)30460-8 - DOI - PMC - PubMed