Depressive symptoms in students during school closure due to COVID-19 in Shanghai.

Since the first identification of a patient with COVID‐19, this infection has spread rapidly worldwide, with a high case fatality rate among the elderly and individuals with chronic diseases. While this virus is less likely to cause serious illness in the younger generation, many affected countries have adopted school closure as a measure to prevent its spread among the community.

Concerns may be raised about the adverse effects of school closure not only on education but also on the health of children. The drastic changes in children's lifestyles during school closure may deteriorate their mental health. The present paper reports the changes in depressive symptoms among students in Shanghai caused by school closure due to the COVID‐19 epidemic.

The present study was originally designed to assess health and health‐associated factors using a web‐based survey among schoolchildren (aged 6–17 years) and their parents from five schools across Shanghai, China. Details of the study procedure are described in Appendix S1. The first survey was conducted from 3 to 21 January 2020 (before school closure) and the second survey was conducted from 13 to 23 March 2020 (during school closure). Depression was assessed using the Children's Depression Inventory – Short Form (CDI‐S) ; the total score ranges from 0 to 20, with a higher score indicating more depressive symptomatology. We also asked about the positive aspects of lifestyle change during school closure. We calculated the mean and SD of CDI‐S scores before and during school closure and tested the difference using the paired t‐test. A within‐subjects analysis of variance was conducted to determine the interactions of Time × Subgroup (sex, school levels, and annual household income levels) in terms of CDI‐S score. We declared statistical significance when P < 0.05 for the paired t‐test and P < 0.1 for the interaction test.

Of 3042 students in the target schools, 2641 (87%) participated in the first survey. Of these, 2427 participated in the second survey. Overall, the mean CDI‐S score significantly decreased between the two surveys: 4.19 before school closure and 3.90 during school closure (P < 0.01). In stratified analyses, children in middle school showed a greater decrease in CDI‐S scores than those in primary school (P for interaction = 0.09), whereas a lower CDI‐S score during school closure was consistently observed across sexes and household income categories (Table 1). As regards positive aspects of lifestyle change during school closure, 71%, 80%, and 83% of children were satisfied with having more time ‘at home,’ ‘with their parents,’ and ‘doing their own things,’ respectively.

Table 1

Depression scores before and during school closure due to the COVID‐19 pandemic among students in Shanghai

	Before school closure	During school closure	Mean differences	P‐value †	P for interaction ‡
Total (n = 2427)	4.19 (2.82)	3.90 (2.56)	−0.29	<0.01
Sex					0.21
Male (n = 1242)	4.03 (2.63)	3.83 (2.57)	−0.20	<0.01
Female (n = 1185)	4.37 (3.00)	3.98 (2.55)	−0.39	<0.01
School grade					0.09
Primary school (n = 875)	3.89 (2.28)	3.77 (2.07)	−0.12	0.13
Grade 1 (n = 207)	3.43 (1.75)	3.53 (1.82)	0.11	0.44
Grade 2 (n = 198)	3.63 (1.77)	3.62 (1.84)	−0.01	0.95
Grade 3 (n = 165)	4.08 (2.36)	3.77 (1.82)	−0.32	0.10
Grade 4 (n = 171)	4.25 (2.70)	4.13 (2.56)	−0.11	0.57
Grade 5 (n = 134)	4.31 (2.79)	3.90 (2.26)	−0.41	0.08
Middle school (n = 1549)	4.36 (3.06)	3.97 (2.78)	−0.39	<0.01
Grade 6 (n = 558)	4.05 (2.79)	3.82 (2.72)	−0.23	0.07
Grade 7 (n = 417)	4.44 (3.02)	4.06 (2.66)	−0.38	<0.01
Grade 8 (n = 344)	4.56 (3.28)	3.92 (2.86)	−0.65	<0.01
Grade 9 (n = 230)	4.66 (3.37)	4.22 (3.01)	−0.43	0.02
Annual household income (CNY)					0.19
<100 000 (n = 277)	5.12 (3.50)	4.68 (3.33)	−0.44	0.02
100 000–200 000 (n = 677)	4.23 (2.76)	3.92 (2.56)	−0.31	<0.01
>200 000–400 000 (n = 767)	3.93 (2.65)	3.78 (2.43)	−0.15	0.09
>400 000 (n = 367)	3.92 (2.52)	3.82 (2.40)	−0.10	0.49

Data are presented as mean (SD).

CDI‐S, Children's Depression Inventory – Short Form; CNY, Chinese Yuan.

Calculated using paired t‐test.

Calculated using within‐subjects analysis of variance by the interaction effect of Time × Subgroup (sex, school levels [primary or middle school], or annual household income levels [<100 000, 100 000–200 000, >200 000–400 000, or >400,000 CNY]).

School closure with a stay‐at‐home policy should drastically decrease the opportunity for direct contact with persons other than family members, decrease physical activity, and increase the time of screen viewing, all of which may deteriorate the mental health of school children. However, the present result is at variance with the prior hypothesis; rather, we observed fewer depressive symptoms during school closure compared to those measured before the closure.

The present finding based on pre‐ and post‐observation without a comparison group should be interpreted with caution. Nevertheless, this unexpected finding could be attributed to relatively good educational environments at home. In Shanghai, nearly all families own a personal computer or smartphone, which would have been used for communication with schools before the pandemic. Soon after school closure, each school started providing an online and TV program for education, and children received interactive education at home from their teacher and communicated with classmates on a daily basis. Such a remote educational system might have contributed to better mental health during school closure.

Children, especially those under high academic pressure, might have less stress in the home environment than at school. Analysis by school type showed that middle school students, who had a higher depressive score before school closure than primary school students, showed a greater decrease in depressive symptoms during school closure. This may partly reflect greater academic stress preparing for entrance examinations in middle school. As regards the positive aspects of staying at home, most children reported being satisfied with having more time ‘at home,’ ‘with their parents,’ and ‘doing their own things.’ Taken together, a well‐equipped home‐schooling system, together with fewer stressors and the positive aspects of staying at home, may have jointly outweighed the potential harm of school closure to children's mental health.

In conclusion, the present longitudinal study in Shanghai during the COVID‐19 pandemic does not show any evidence of increased depressive symptoms among students after a 2‐month school closure. Additional study is required to assess the long‐term effect of school closure on the mental health and educational attainment of children.

Disclosure statement

All authors have indicated they have no financial relationships relevant to this article to disclose.

Appendix S1. Methods of study sampling.

Click here for additional data file.