2.1 Study design
A web-based questionnaire was delivered between November and December 2020 as a link via a communication application (WeChat) to those HCWs, who worked at several COVID-19 units (including the emergency department, outpatient, fever clinic, intensive care unit, infection ward, isolation ward, mobile cabin hospital) in Wuhan, from the onset of the COVID-19 outbreak. This study was approved by the Ethics Committee of the First Affiliated Hospital of Wannan Medical College (Yijishan Hospital) (approval number 20201025). All participants provided written informed consent.
HCWs having cared for COVID-19 patients in Wuhan were recruited by snowball sampling17. Inclusion criteria were HCWs who worked at a COVID-19 designated institution, had direct contact with COVID-19 patients, or had close contact with those provided health care service to COVID-19 patients. The post-hoc analysis indicated that all participants had worked in the designated institutions for more than 7 days, hence we did not set an exclusion criterion to a minimal working time in the designated institutions. Exclusion criteria were non-HCWs, and provision of invalid answers of the questionnaire.
2.3.1 Social-demographic, job and COVID-19 related variables
Social-demographic variables in the questionnaire include age, gender, marital status, education level (college diploma, bachelor, or master and above), occupation (doctor, nurse or non-clinical; which includes nurses and doctors carrying out administrative and logistics tasks during the outbreak of COVID-19 due to lack of human resource in these departments), whether suffering from chronic diseases, and whether being the only child. Job-related variables include work experience (≤ 2 years, 2–5 years, 6–10 years, 11–20 years and > 20 years), position level (unrated, elementary, intermediate, advanced) and department (emergency department, outpatient, fever clinic, intensive care unit, infection ward, isolation ward, mobile cabin hospital). COVID-19 related variables include number of working days during the epidemic, experience in caring for patients suffering from other respiratory infectious diseases (e.g. SARS, MERS, influenza), direct care for COVID-19 patients, occupational exposure without protection, getting infected, relatives or close friends getting infected or died of COVID-19 (diagnosed by qPCR or clinical diagnostic criteria), having been in quarantine and time of quarantine due to COVID-19, experience of social isolation, received psychological assistance, having received an award for fighting COVID-19, frequency of obtaining COVID-19 related information (rarely, sometimes, always), sleeping time (< 5 h, 6-8 h, > 8 h), job satisfaction (dissatisfied, neutral, satisfied), intention to resign, feeling of loneliness (rarely, sometimes, often).
2.3.2 PCL-5, family care index questionnaire, and quality of life scale
The presence and severity of PTSD symptoms were measured by the post-traumatic stress disorder checklist-5 (PCL-5), containing 20 self-report items, with high reliability and validity18,19.Each item of PCL-5 has five-point Likert scare (0 = not at all, 1 = a little bit, 2 = moderately, 3 = quite a bit and 4 = extremely). Individual’s score of PCL-5 ranges from 0 to 80, and the cut-off point score for diagnosis PTSD was set as more than or equal to 33, with a higher score indicating more serious PTSD symptoms20.
Family function was assessed with the family care index questionnaire (Adaptation, Partnership, Growth, Affection and Resolve, APGAR) designed by the University of Washington21. The questionnaire contains five dimensions: adaptation, partnership, growth, affection and resolve. The questionnaire uses a Likert 3-level scoring response model, 0 points (almost never), 1 point (sometimes), 2 points (often like this). The full score of the scale is 10 points, and 0 to 3 point indicates a severe family dysfunction, 4 to 6 point indicates a moderately impaired family function, and a score of 7 to 10 indicates a good family function.
Quality of life was measured with the Chinese QOL questionnaire22, with a total of 6 items, assessing the physical health, mental health, economic status, work or study status, family relationship and non-family peers’ relationship, respectively. The answers are divided into 5 levels (1=very poor, 2=poor, 3=fair, 4=good, and 5=very good).
2.4 Statistical analysis
Continuous variables were summarized as either means and standard deviations or medians and interquartile range as appropriate. Categorical variables were described as frequencies and percentages. The differences between groups with or without PTSD were analyzed by Fisher’s exact test and the Mann – Whitney U test for continuous variables. Factors associated with PTSD were identified using logistic regression Adjustment for the following variables (determined a priori) was conducted in a forward stepwise manner: age, suffered from chronic diseases, being an only child, occupational exposure without protection, relatives or friends be infected, relatives or friends died of COVID-19, the experience of social-isolation, job satisfaction, frequency of obtaining COVID-19 related information, working department, and variables with p ≤ 0.2 in the univariable analysis. Odds ratios (OR) and 95% confidence intervals (CI) were reported. Two-tailed p < 0.05 was considered statistically significant. All analyses were performed using R software version 3.6.2 (R Foundation for Statistical Computing).