Study design and participants
To prevent the spread of SARS-CoV-2 (Severe Acute Respiratory Syndrome Coronavirus 2) through droplets or contact, we used a web-based cross-sectional survey based on the National Internet Survey on Emotional and Mental Health (NISEMH) to collected data, NISEMH is an ongoing, online health-related behavior survey of Chinese population. This web-based survey of the COVID-19 was broadcasted on the Internet through the WeChat public platform and the mainstream media. All Chinese people using WeChat or other social tools may see this survey, and answered the questionnaire by scanning the two-dimensional barcodes of the questionnaire address or clicking the relevant link. To encourage the recruitment of potential participants, all participants in the survey would receive a report on their mental health after completing the evaluation. This web-based questionnaire was completely voluntary and non-commercial.
Data Collection
Participants answered the questionnaires anonymously on the Internet from February 3, 2020 to February 17, 2020. All subjects reported their demographic data, COVID-19 related information, and three standardized questionnaires, which assessed their generalized anxiety disorder (GAD), depressive symptoms, and sleep quality. In order to ensure the quality of survey, we have set the response range of some items (e.g., the age range was limited to 6–80 years old, some items needed to be answered in reverse) and encouraged participants to answer carefully through questionnaire explanations. In addition, questionnaires that were completed < 1 minute or > 60 minutes would be excluded from analysis. Finally, a total of 7,236 participants who completed the questionnaires (response rate of 85.3%) were included in the analysis.
Measures
Demographic information
Demographic variables included gender (male or female), age, and occupation. Occupation included the following four types: (1) Healthcare workers, which included doctors, nurses, and health-related administrators; (2) Enterprise or institution workers, which consisted of enterprise employees, national/provincial/municipal institution workers, and other relevant staff; (3) Teachers or students, which included teachers or students from universities, middle schools, or elementary schools; and (4) Others, which consisted of freelancers, retiree, social worker, and other relevant staff.
COVID-19 related knowledge
This section was evaluated by two items: (1) Times to focus on the COVID-19, which measured the average time spent focusing on the COVID-19 epidemic information every day; (2) Knowledge of the COVID-19, which assessed based on the following six judgment questions about COVID-19 related knowledge: a. Inhalation of droplets from sneezing, coughing, or talking of an infected person could cause infection; b. Contact with something contaminated by an infected person could lead to infection; c. The incubation period of the virus does not exceed 14 days; d. Contact with an asymptomatic person might also lead to infection; e. There are already targeted drugs that could cure the disease; f. Taking “Shuanghuanglian Oral Liquid” could prevent infection of this disease. Of the above six questions, one point was given for correct answers, and no points were given for incorrect or uncertain answers. Participants with scores ≥ 5 points, equal to 4 points, and ≤ 3 points were considered to be quite understand, generally understand, and do not understand.
Generalized anxiety disorder
We used Chinese version of GAD-7 (Generalized Anxiety Disorder-7) scale to assess subject’s anxiety symptoms. The GAD-7 has been previously used in Chinese populations, and found to have good reliability (Cronbach’s alpha = 0.90) [15, 16]. Seven items assess the frequency of anxiety symptoms over the past two weeks on a 4-point Liker-scale ranging from 0 (never) to 3 (nearly every day). The total score of GAD-7 ranged from 0 to 21, with increasing scores indicated a more severe functional impairments as a result of anxiety [17]. For the purpose of this study, we defined a GAD-total score of 9 points or greater as the presence of anxiety symptoms [15].
Depressive symptoms
The Center for Epidemiology Scale for Depression (CES-D) in Chinese version was used to identify whether participants had depressive symptoms [18], and the Chinese version of this scale has been validated and extensively utilized in Chinese population [18, 19]. Twenty items assess the frequency of depressive symptoms over the past two weeks on a 4-point Liker-scale ranging from 0 (rarely or none of the time) to 3 (most or all of the time). The score range of the CES-D is 0–60 points, and higher scores indicated more severe depressive symptomatology [20]. In our study, CES-D scores greater than 28 points indicated depressive symptoms.
Sleep quality
The Chinese version of the PSQI (Pittsburgh Sleep Quality Index) scale was used to assess the subject’s sleep quality over the past two weeks [21]. The PSQI scale contains seven components (subjective sleep quality, sleep duration, sleep latency, habitual sleep efficiency, use of sleep medications, sleep disturbance, and daytime dysfunction), and the score for each component ranging from 0 to 3 points. The global PSQI score ranges from 0 to 21, with higher scores indicated more severe sleep disorder [22]. The Chinese version of PSQI has been demonstrated to be reliable and valid in Chinese population [21], a global PSQI score greater than 7 points indicated poor sleep quality.
Statistical analysis
First, descriptive analyses were conducted to describe the demographic characteristics and COVID-19 related knowledge in Chinese population. Second, the prevalence of GAD, depressive symptoms, and sleep quality stratified by gender, age, and occupation were reported, and Chi-square test (χ2) was used to compare the differences between groups. Third, univariate and multivariate logistic regression models were performed to explore potential influence factors for GAD, depressive symptoms, and sleep quality during COVID-19 epidemic. Odds ratio (OR), adjusted odds ratio (AOR), and 95% confidence interval (95% CI) were obtained from logistic regression models. All data were analyzed using Statistical Package for Social Sciences (SPSS) version 24.0. P-values of less than 0.05 were considered statistically significant (2-sided tests).