The rapid change in mental health among college students after introduction of on-campus quarantine during the 2022 Shanghai COVID-19 lockdown

Dongni Ma, Yifang Kuang, Zhaohui Lan, Suhua Zeng, Yi Li, Mengnan Shang, Ru‐Yuan Zhang, Binglei Zhao, Weidong LiFrontiers in Public Health 2023 · DOI

TL;DR

This study tracks the rapid deterioration of mental health among college students during the 2022 Shanghai COVID-19 lockdown, finding that depressive symptoms nearly quadrupled within just two weeks of strict on-campus quarantine. Crucially, these symptoms did not recover even after restrictions were partially relaxed, highlighting the lasting psychological "scarring" effect of sudden, high-intensity isolation.

What Problem Does This Paper Address?

While the long-term mental health impacts of the COVID-19 pandemic are well-documented, there is less data on the acute phase of sudden, stringent lockdowns—specifically those where students are confined to small dormitory spaces. The researchers sought to understand how quickly psychological resilience erodes under these conditions and which specific factors (e.g., food quality, exercise, social status) mitigate or exacerbate the decline. By studying students at Shanghai Jiao Tong University (SJTU) during the 2022 lockdown, the authors provide a "real-time" look at the transition from normal campus life to total isolation.

Methods

The researchers employed a retrospective cohort trial design using an online survey conducted in early April 2022. They divided the timeline into three distinct phases:

  1. Period 1 (Pre-quarantine): Normal activities before March 9.
  2. Period 2 (Strict Quarantine): March 9–23, where students were confined to dorm rooms (~30 $m^2$ suites).
  3. Period 3 (Relaxed Quarantine): March 24–early April, where students could engage in essential campus activities but could not leave the university.

To quantify mental health, they used the Beck Depression Inventory, second edition (BDI-II), a 21-question self-reporting tool. They analyzed the data using repeated-measures ANOVAs and linear regression models to identify predictors of change in depression severity.

Datasets

The study utilized a primary dataset collected from 274 valid survey responses from SJTU students (58.39% undergraduate, 41.61% graduate).

  • Demographics: 40.51% male, 59.49% female; age range 18–42.
  • Context: The dataset represents approximately 1% of the total student population at the university during the lockdown.

Key Insights and Ideas

  • The "No Reversal" Effect: One of the most significant findings was that depression levels did not drop when restrictions were eased in Period 3. This suggests that once a psychological threshold of distress is crossed during a strict lockdown, simply "opening up" isn't enough to trigger immediate recovery.
  • The Relationship Paradox: Contrary to what one might expect, students in romantic relationships experienced more psychological distress than single students. The authors hypothesize that the sudden deprivation of physical contact and joint activities with a partner created a sharper sense of loss compared to single students who were already more accustomed to solitary time.
  • Video Games as a Buffer: In the linear regression model, time spent playing video games was a negative predictor of increased depression ($\beta = -0.23, p < 0.001$). This suggests that gaming served as a vital cognitive distraction and a social outlet during total isolation.
  • Physical Infrastructure of Mental Health: Satisfaction with food and the ability to exercise were the strongest positive predictors of resilience. This shifts the conversation from purely "psychological" interventions to the importance of logistical and nutritional support during crises.

Key Results

  • Prevalence Spike: The proportion of students exhibiting depressive symptoms (BDI-II score $\ge$ 14) jumped from 9.1% in Period 1 to 36.1% in Period 2.
  • Persistence: In Period 3, the prevalence remained high at 34.67%, showing no statistically significant improvement from the strict lockdown phase ($p = 0.954$).
  • Severity Distribution: Major depression (scores > 28) increased from a baseline of 1.46% to 10.95% during the lockdown.
  • Regression Analysis: Smaller quarantine areas, less exercise, and lower food satisfaction were significantly associated with worsening BDI-II scores ($p < 0.001$).

Limitations and Future Work

The authors acknowledge several limitations:

  • Recall Bias: Because the survey was administered once in April, students had to recall their feelings from March and earlier, which may introduce "retrospective bias."
  • Sample Size: The study only captured ~1% of the university population, which may not be fully representative.
  • Narrow Focus: The study focused exclusively on depressive symptoms. Future work should include anxiety, PTSD, and suicidal ideation to provide a more holistic view of the crisis.
  • Longitudinal Gaps: The study ends shortly after the lockdown eased; a longer-term follow-up is needed to see if and when students eventually return to their baseline mental health.

Summary updated

Abstract

Depressive symptoms in university students rapidly increased after 2 weeks of quarantine and no perceptible reversal was observed over time. Concerning students in a relationship, ways to take physical exercise and to relax should be provided and the food supplied should be improved when young people are quarantined.
BibTeX
@inproceedings{ma2023rapid,
  author = {Dongni Ma and Yifang Kuang and Zhaohui Lan and Suhua Zeng and Yi Li and Mengnan Shang and Ru‐Yuan Zhang and Binglei Zhao and Weidong Li},
  title = {{The rapid change in mental health among college students after introduction of on-campus quarantine during the 2022 Shanghai COVID-19 lockdown}},
  booktitle = {Frontiers in Public Health},
  year = {2023},
  doi = {10.3389/fpubh.2023.1132575},
}
Figure