Participants
From February 23 to 29, 2020, the current study used a snowball sampling approach to distribute questionnaires online in the mainland of China. The questionnaires were distributed to WeChat circles of friends, Tencent QQ and other public platforms. When participants completed the questionnaire, they forwarded it to their own WeChat circle of friends or other public platforms to expand the sample size. Each IP address could only be filled in once. The inclusion criteria were as follows: (1) 18–65-years-old, (2) native Chinese residents and could complete the questionnaires on the cellphone or computer, (3) informed consent. The exclusion criteria were: (1) unable to read correctly or use computer or cellphone to complete the questionnaires, (2) refused to participate in this research.
In order to control bias, the questionnaires were initially distributed in the same number in each provincial capital city. The daily sample size is based on the experience of a national survey and the sample size is set to be more than 1500 [22]. A total of 66152 questionnaires were returned. On February 26 and 27, only 5 individuals filled out the questionnaire. Therefore, the data of these two days were excluded. And 5948 questionnaires such as missing data, incomplete information or extreme data were excluded. After deleting these substandard responses, a total of 60199 valid questionnaires were analyzed in this study, with an effective rate of 91%.
The current study was approved by the Ethics Committee of the First Affiliated Hospital of Chongqing Medical University, China. All participants confirmed the informed consent before answering the questionnaires.
Measures
Self-compiled descriptive characteristics questionnaire: Demographic data, work environment, family conditions, and social activities were extracted by 13 items, including gender, age, education level, occupation, marital status, family structure in childhood, whether the only child in the family, parenting style in childhood, whether living with parents before 10-years-old, number of close friends, the collective atmosphere in work/school, the management style of work/school, and social activities of last 2 weeks. According to Baumrind’s research [23], parenting style in childhood was divided into authoritarian, neglectful, permissive and democratic styles. The management style of work/school was divided into three most common types of leadership styles as defined by Kurt Lewin—autocratic, laissez-faire and democratic [24].
Depressive symptoms: The self-rating depression scale (SDS) [25] contains 20 items, and the design was based on the diagnostic criteria for depression. The subjects rated each item using a 4-point Likert scale with respect to how they have felt during the past several days. The raw sum score of SDS is 20–80; however, the results are usually presented as the SDS index, which is obtained by expressing the raw score converted to 100-points scale. The cut-off value of SDS standard score is 53, 53–62 with mild depressive symptoms, 63–72 with moderate depressive symptoms, and >73 with severe depressive symptoms according to the Chinese norm [26]. The Chinese version had a good internal consistency reliability of the total scale (α = 0.86) [27]. The Cronbach’s α of current study is 0.63.
Anxiety symptoms: The state anxiety inventory (SAI) is a scale from the state-trait anxiety inventory [28], containing 20 items to evaluate state anxiety under stress, using a 4-point Likert scale. The total score ranged from 20–80; according to the score boundaries, the total score was 20–39 without anxiety symptoms, 40–47 with mild anxiety symptoms, 48–54 with moderate anxiety symptoms, and 55~80 with severe anxiety symptoms. The Chinese version had a good internal consistency reliability of the total scale (α = 0.91) [29]. The Cronbach’s α of current study was 0.68.
Negative cognitive processing bias questionnaire (NCPBQ): It [16] is a 16-item self-report measure in Chinese, assessing the negative attention bias, negative memory bias, and rumination, using a 4-point Likert scale (1 = not match; 4 = perfect match). An example of an item is “I always remember my mistakes clearly.” Higher total scores indicate negative cognitive processing bias. It had a good internal consistency reliability of the total scale (α = 0.89) in a normal population of college students. The Cronbach’s α of current study was 0.84, and that of negative attention bias, negative memory bias, and rumination was 0.78, 0.68, and 0.72, respectively.
Confirmed growth rate and death rate
Confirmed growth rate is the ratio of the difference between the cumulative number of confirmed cases announced on the day and on the previous day to the cumulative number of confirmed cases announced on the previous day. The death rate is the ratio of the difference between the cumulative number of deaths announced on the day and on the previous day to the cumulative number of deaths announced on the previous day. The number of confirmed cases and deaths was provided by the website of National Health Commission of China [3].
Depressed group and anxiety group
The participants were divided into non-depressed, mild depressive symptom, moderate depressive symptom, and severe depressive symptom groups according to the SDS scoring criteria. Also, they were divided into non-anxiety, mild anxiety symptom, moderate anxiety symptom, and severe anxiety symptom groups according to the SAI scoring criteria.
Data analysis
The data were analyzed using software SPSS 23.0 and SAS 9.4. Measurement data were expressed as mean ± standard deviation ( ± sd). The age data were divided down by the maximum age, with every ten years as an age group. Enumeration data were expressed by the number of people (%). The correlation between the severities of depressive and anxiety symptoms was evaluated by Pearson’s correlation coefficient, and P < 0.05 on double sides was statistically significant. The analysis of the correlations between characteristics (gender, age, education level, occupation, family structure in childhood, the only child in the family, parenting style in childhood, living with parents before 10-years-old, number of close friends, the collective atmosphere in work/school, the management style of work/school, marital status, and social activities of last 2 weeks) and anxiety or depressive symptoms initially were assessed by the chi-square test. The correlations between negative cognitive processing bias (negative attention bias, negative memory bias, and rumination), confirmed growth rate, death rate, and anxiety or depressive symptoms initially were assessed by one-way analysis of variance (ANOVA). The variables with P < 0.05 were entered in the ordered multiple logistic regression analysis models by stepwise method.