Background China was the first country affected by the COVID-19 virus, and it is a very important case to study the effects of the virus and the consequent restrictions. However, national representative studies of how the COVID-19 pandemic affects mental health in China are still limited.

Methods Using two waves of the China Family Panel Studies, we follow the same individuals before and during the pandemic. We compare weighted means using 95% CIs to explore mental health deterioration, and we and perform several linear regressions with the Ordinaly Least Square (OLS) estimator to identify individuals most affected by the COVID-19 pandemic.

Results The prevalence of severe cases of depression, measured using an eight-item version of the common Centre for Epidemiologic Studies Depression Scale (CES-D), increased from 6.68% in 2018 to 7.86% in 2020; quantifiable as around a 18% increase. This deterioration is higher for individuals subject to strict lockdowns, about 0.4 symptoms more on average, and it is stronger among those who already reported symptoms of depression in the 2018 wave of data. Individuals with more open personalities tend to experience more severe deterioration: a 1 SD change in the openness trait corresponds to 0.05 more symptoms. On the other hand, more neurotic individuals seem less negatively affected.

Conclusion We find clear evidence of a moderate level of mental health deterioration between 2018 and 2020. These effects are larger for individuals subject to stricter lockdowns and for individuals with more open personalities.

Data are available on reasonable request. This paper uses data from China Family Panel Studies (wave 2018 and wave 2020), which are available to researchers on request. The datasets generated and/or analysed during the current study are available in the (CFPS Data Platform) repository, (http://www.isss.pku.edu.cn/cfps/download). Please contact the China Family Panel Studies project team for data access. Their mailing address is Office of China Family Panel Studies, Science Building #5, Peking University, 5 Yiheyuan Road, Haidian District, Beijing, P.R. China; Postal Code：100871; email: isss.cfps@pku.edu.cn.The authors will make replication files (Stata do file) available.