key: cord-0978153-gef2b6am
authors: Geng, Shiyu; Zhou, Yaoguang; Zhang, Wenyong; Lou, Anni; Cai, Yuhan; Xie, Jieying; Sun, Jing; Zhou, Wangmei; Liu, Weizhi; Li, Xu
title: The influence of risk perception for COVID‐19 pandemic on posttraumatic stress disorder in healthcare workers: A survey from four designated hospitals
date: 2021-02-16
journal: Clin Psychol Psychother
DOI: 10.1002/cpp.2564
sha: 467d4f587d2211d88df33c98489d004a2599fd51
doc_id: 978153
cord_uid: gef2b6am

The aim of current study was to investigate risk perception of COVID‐19 pandemic, sleep quality and time change of leisure activity and their correlations with posttraumatic stress disorder (PTSD) in healthcare workers (HCWs) from four designated hospitals in China. Medical staffs (n = 317) from three designated hospitals in Guangdong Province and one designated hospital in Guangxi Province were surveyed on their demographic information, sleep quality and time change of leisure activity, risk perception of pandemic and PTSD symptoms (by using PTSD checklist for DSM‐5 (PCL‐5)). Hierarchical regression and structural equation model (SEM) were used to examine the correlated factors of PTSD. The prevalence of high level of PTSD symptoms (PCL‐5 > =33, a probable diagnosis of PTSD) was 10.7%. Regression analysis found that risk perception (dread: β = 0.142, p < 0.01; familiarity: β = 0.203, p < 0.01), sleep quality (β = 0.250, p < 0.001), time change of leisure activity (β = −0.179, p < 0.01), were independently correlated with PTSD severity, which was further confirmed by SEM. Locations of COVID‐19‐related hazards were significant different in cognitive map of risk perception between groups with high and low levels of PTSD symptoms. Risk perception of COVID‐19 pandemic influenced PTSD symptoms in HCWs. Adequate time for leisure activity and good sleep quality protected some HCWs against PTSD symptoms under the influence of pandemic. More researches were warranted to understand the path from pre‐factors of risk perception to its psychological consequences among HCWs.

followed by cough (Lai, Shih, Ko, Tang, & Hsueh, 2020) . According to the diagnosis and treatment of novel coronavirus pneumonia from National Health Committee (NHC), patients with severe symptoms and underlying disease would have life-threatening situations (NHC, 2020) . On 11 March, Tedros Adhamon Ghebreyesus, directorgeneral of the WHO, stated that COVID-19 had become a global pandemic (Director-General, 2020) . This pandemic now has brought updating challenge to the whole world, reaching over 200 countries, areas or territories with 11 million confirmed cases and more than 204,967 deaths (WHO situation report updated to 9 July 2020). As of 7 July, there have been a 25% increase in cases and 18% increase in deaths reported from the African continent over the past week (WHO, 2020).

Such widespread outbreaks are associated with adverse mental health consequences. Fear of falling sick or dying themselves, feelings of stigma and helplessness, worry about contacting the virus, closure of business were negative events that potentially put negative impact on individual mental health . Healthcare workers (HCWs), especially those who were working at frontline in designated hospitals for COVID-19, have been facing kinds of enormous pressure, including a high risk of infection and inadequate protection from contamination, frustration, overwork, isolation, discrimination, negative emotions, worry about the shortage of protective equipment, feelings of incapability when facing critically ill patients, a lack of contact with their families, and exhaustion (Kang et al., 2020) . It was also an alarming concern that the psychological intervention services for HCWs constantly encountered obstacles, as medical staff hesitated to participate in the group or individual psychology interventions provided to them. For example, a proportion of HCWs refused to commit they had psychological distress even they showed some signs of symptoms. And some stated that they did not want their families to worry . Accordingly, the mental health status of HCWs deserves more attention and investigation (McAlonan et al., 2007) .

Posttraumatic stress disorder (PTSD) occurs in a subset of individuals exposed to traumatic events and consists of four clusters of symptoms: intrusive re-experiencing of the trauma, avoidance of trauma-related stimuli, negative alternation of cognition and mood, and excessive arousal or reactivity (Shalev, Liberzon, & Marmar, 2017) . PTSD was seen as one of the major psychological impacts caused by previous massive outbreak of infectious disease in studies (Jung, Jung, Lee, & Kim, 2020; Lancee et al., 2008; Vyas, Delaney, Webb-Murphy, & Johnston, 2016) . During the ongoing COVID-19 infections, Liu et al reported that the prevalence of posttraumatic stress symptoms in hardest-hit area 1 month after outbreak was 7%, with women showing higher symptom severity . Another study investigated psychological impact of COVID-19 pandemic (e.g., anxiety, depression and stress) during the initial stage of the COVID-19 outbreak found that 8.1% of participants reported moderate to severe PTSD symptoms . Recently, published studies showed 3.8% of HCWs involved in the care for COVID-19 patients in Singapore and India showed the prominent self-reported PTSD symptoms (Chew et al., 2020) , and general HCWs investigated online in China had prevalence rate of 2.2% (Yin et al., 2020) . During the SARS outbreak, PTSD severity is significant different from those who worked at frontline and those who did not (Lin et al., 2007) . The characteristics of PTSD in HCWs from designated hospital in China under influence of COVID-19 needs further exploration.

Risk perceptions reflect people's intuitive evaluations of various hazards that they are or might be exposed to, including multiple undesirable consequences that people associate with a specific cause. The elements that increase or mitigate fear and risk perceptions during pandemic included voluntariness, knowledge, visibility and trust (Cori, Bianchi, Cadum, & Anthonj, 2020) . To evaluate different hazards, Slovic postulated two-factor representations to outlines the feature of risk perception, namely 'dread risk' and 'unknown risk'. The former one was defined at its high end by perceived lack of control, dread and catastrophic potential, while the latter one was defined at its high end by hazards judged to unobservable, unknown, new and delayed manifestation of harm (Slovic, 1987) . It was also postulated that a combination of technical perception of hazard and outrage could provide psychometric insights of risk perception and public response (Fischhoff, 2015; Sandman, 1988) . Shi et al confirmed that risk perception was a mediator between risk information and coping method and mental health of people after outbreak of an infectious disease.

The cognitive map of risk perception on Slovic's two dimensions showed only one of six SAR-related hazards was in maximum risk quadrant (Shi et al., 2003) . Past experience has implicated the critical role of risk perception in unprecedented times of COVID-19 pandemic (Cori et al., 2020) .

• Risk perception of pandemic in HCWs from designated hospitals was positively correlated with PTSD symptom score after controlling other related variables.

• Locations of COVID-19-related hazards were significant different in cognitive map of risk perception between groups with high and low levels of PTSD symptoms in HCWs from designated hospitals.

• Good sleep quality and sufficient time for leisure activities were protective factors for PTSD symptoms in HCWs working at frontline hospitals.

Previous evidences supported the relation between risk perception and PTSD. Jalloh et al conducted a cross-sectional survey to a national sample in areas of Ebola epidemic and found that perceived threat of Ebola disease was independently associated with PTSD (Jalloh et al., 2018) . Another study has reported that perceptions of risks in SARS were significantly positively associated with PTSD symptom scores in HCWs after three years of outbreak (Wu et al., 2009 ).

Studies of risk perception under COVID-19 pandemic has explored the relationship between pandemic-related information (e.g., risk communication and COVID-19 related knowledge) and preventive behaviours (Abrams & Greenhawt, 2020; Taghrir, Borazjani, & Shiraly, 2020) . However, there is still a lack of examination about the relation between PTSD and risk perception under the COVID-19 pandemic within half year of outbreak by using psychometric cognitive model of risk perception. In addition, the distribution of COVID-19-related hazards on two-dimension cognitive map of risk perception and its difference determined by PTSD severity were rarely explored.

Sleep and leisure activity were indispensable daily behaviours for regulating psychological stress. A systematic review and meta-analysis has confirmed sleep quality was impaired in individuals with higher PTSD symptoms (Lewis et al., 2020) . The studies conducted in COVID-19 pandemic also have reported sleep quality was negatively correlated with PTSD symptoms in general HCWs and residents from hardest-hit area Yin et al., 2020) . The significance of this correlation needs further examination in HCWs from designated hospital. The studies of leisure activity and mental health mainly focused on the domain of physical activity (White et al., 2017) . However, the content of leisure activity could also include hobbies, relaxation and social intercourse with others (Hakansson, Axmon, & Eek, 2016) . The 2014 Time Use Survey (2014TUS) conducted by Statistics Korea (KOSTAT) classified leisure activity into outdoor activities, indoor activities, internet use and other activities (Cha, 2018) . There are individual differences on the leisure activities and the time for such activities could a reliable measure for the involvement for the leisure activities. Previous studies found more time for leisure activity was benefit for both physical and psychological health status (Cha, 2018; Hakansson, Axmon, & Eek, 2016; Holstila, Manty, Rahkonen, Lahelma, & Lahti, 2017; Mausbach et al., 2017) . Under the impact of pandemic, the time for leisure activity was expected to be changed by increasing working load and frequent working shift in HCWs from designated hospitals. Whether this change is connected to PTSD was rarely explored.

1.5 | The aim of the current study 

The online questionnaire was designed by one of our authors (GSY) in advance, which could be accessed by a link generated by the Wenjuanxing. The link of online questionnaire was sent to the managers and administrative staffs in departments of those designated hospitals.

They subsequently sent this link to the online groups of their departments so HCWs could finish the questionnaire by clicking into it by mobile phone or computer. After entering into the online survey, the questionnaire could only be submitted when all items were answered.

The data were automatically stored in online system of Wenjuanxing, which could be downloaded after the survey was finished. 

According to the psychometric model of postulated by Slovic, dread and familiarity were adopted as the measurement of risk perception (Slovic, 1987) . These two dimension each was rated on a five-point Likert scale from 1 (not dreadful at all or very familiar) to 5 (extremely dreadful or not familiar at all) on a single hazard. We included following five items as hazards on the reference to previous studies: H1.

Virus: SAR-COV-2; H2. Disease: COVID-19; H3. COVID-19 patient or virus carrier; H4. Treatment and prevention of COVID-19; H5. Frontline HCWs who had contact with COVID-19 patients. Thus there were total 10 items in section (2 dimension × 5 hazards). The reason for inclusion of H4 and H5 is that people who were exposed, infected and fall ill can be considered the potential hazard (Malecki, Keating, & Safdar, 2020) . The item was rated retrospectively based on risk perception of the peak period (basically from January to March 2020) of pandemic. Factor analysis has confirmed that above items fit twofactor model better than one factor model, which is in line with the Slovic's two-dimension theory. Further, after exlusion of two items of H5 for having relatively low factor loading, the remaining eight items had an ideal model fit (adjusted two factor model) with good reliablity (Composite reliability were 0.927 and 0.953 for dread and familiarity, respectively) and validity (Average variance extracted were 0.763 and 0.835 dread and familiarity, respectively, with sigificant discriminant validity). Relevant analyses were presented in the appendices (See supplementary file).

The sleep quality was assessed by one item extracted out of Pittsburg Sleep Quality Index (PSQI) as 'How good is your sleep quality?' (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989) . The item is rated on a four point Likert scale ordered as 'very good', 'good', 'bad'

and 'very bad'. Single-item sleep measures have been used in previous studies , with evidence that single-item component of subjective sleep quality correlated highly with global sores of the full sleep questionnaire (e.g., a 11-points single item with Medical Outcomes Study (MOS) Sleep scores and a 6-points single time with Mary's Hospital Sleep Questionnaire) (Cappelleri et al., 2009; Pien, Sammel, Freeman, Lin, & DeBlasis, 2008 ).

Time change of leisure activity was measured with a study-specific single item stated as 'How was your time for leisure activity since the F I G U R E 1 Sampling process start of epidemic'. The item was rated on a five point Likert scale ordered as 'no time', 'decreased', 'as usual', 'increased' and 'a lot'.

This item was set on the reference of limited studies measuring the time for leisure activities (Hakansson et al., 2016) . Specific criteria for time for leisure activity (such as >1 hr) was not used for consideration of individual difference on the demand of time for leisure activity. This item was mainly focusing on the time change caused by COVID-19 pandemic.

Outliers were detected by box plot to find extreme outlying value. 

3.1 | The prevalence, demographic characteristics and difference of PCL-5 score in different subgroups A total of 317 participants recruited in this study, among which 38% (n = 121) were from emergency department, 29% (n = 91) were from respiratory department and 33% (n = 105) were from other departments (including hepatology unit (n = 54, 17%), digestive department (n = 25, 8%), department of cardiovascular medicine (n = 17, 5%), medical imaging department (n = 13, 3%)). About 70% (n = 222) of the sample were women and aged 35 or younger; 92% (n = 293) of the participants belong to the Han nationality. The proportion of doctor, nurse and other medical professions were 44% (n = 140), 45% (n = 144) and 11% (n = 33), respectively. Thirty-six percent of participants (n = 201) had an education level of university and 37% (n = 116) had a higher degree. Using PCL-5 cut-off score of 33 as criterion, 34 people met the criterion of a probable PTSD diagnosis, with evaluated PTSD prevalence rate of 10.7%.

The comparison between groups of demographic information, sleep quality and time change of leisure activity was shown in Table 1 

The results of the hierarchical regression analysis are presented in Table 2 . Fear in risk perception used the mean value of fear in five hazards as continuous variable. Knowledge in risk perception used the mean value of knowledge in five hazards as continuous variable. Table 1 

Locations of COVID-19-related hazards on dread-familiarity twodimension cognition map were shown in Figure 2 Locations of COVID-19-related hazards on dread-familiarity twodimension cognition map in different subgroups divided by PCL-5 cutoff score (33) were shown in Figure 3 . The difference of the rated scores of dread and familiarity were presented in Table 3 . On the group of PCL-5 ≥ 33 (n = 34), overall risk perceptions towards pandemic were 3.17 (mean, 95% CI: 2.89-3.45) for dread, 2.72 (mean, 95% CI: 2.53-2.91) for familiarity. The mean values of dread for SAR-COV-2 virus (H1), disease of COVID-19 (H2), COVID-19 patient or T A B L E 1 Demographic information, sleep quality, time change of leisure activity and group differences of PCL-5 scores in participants. T A B L E 2 Regression analyses with PCL-5 score as the dependent variable (n = 317).

Variable PCL-5 score

Step 1 There were significant differences on the rated score of risk perception for five hazards between groups divided by PCL-5 cut-off 

The prevalence of PTSD in current sample was relatively high compared with recent-published studies concerning PTSD in HCWs (Chew et al., 2020; Yin et al., 2020) and was close to the prevalence rate from another study conducted in stratified sample of HCWs from

Beijing 3 years after outbreak of SARS (Wu et al., 2009) . Different from above studies, the current study mainly focused on the population of HCWs in designated hospitals. Over two thirds of our respondents were frontline department staff who continued suffering long hours of work and confronted with high risk of infection. And those who worked in non-frontline department in a designated hospital also, when consultation of COVID-19 was needed, were required to have face-to-face contact with patients, exposed to deleterious environment. Relatively high working pressure and risk potentially raised the possibility of enhanced PTSD prevalence rate in those population group. Furthermore, the onset of psychiatric symptom could be delayed by the cumulative effect of time. For instance, PTSD symptoms of HCWs would become severer after long time of chronic stress (overloading of work, lack of sleep, worry and fear) and experiencing multiple traumatic event (such as witnessing patient suffering or dying) (Kilpatrick et al., 2013; Schmeltzer et al., 2015) . Therefore, the prevalence of PTSD in HCWs might tend to be higher at half year after outbreak than other time points close to the onset of epidemic.

However, this finding should be taken cautiously for the current investigation did not directly study the higher prevalence of PTSD by comparing controlled and matched groups of HCWs from designated and non-designated hospitals with appropriate sampling method.

More qualified studies are needed to clarify this issue.

Consistent with our hypothesis, results from regression analysis and SEM supported the significant correlation between the risk perception of COVID-19 pandemic and PTSD. This finding is similar to the points from previous studies concerning risk perception of infectious disease and PTSD (Jalloh et al., 2018; Lau et al., 2005; Wu et al., 2009) . Our results also demonstrated that those HCWs who had significant PTSD symptoms perceived higher pandemic risk. Nevertheless, the causality between risk perception and PTSD remains unclear. Based on psychopathology characteristics of PTSD, it could be inferred that HCWs with significant PTSD symptoms were prone to formulate negative cognition (e.g., 'Now the world is even more dangerous') towards COVID-19 pandemic and elevated their risk perceptions (Brown et al., 2018) . On the other hand, high level of risk perception could worsen HCWs' vulnerability towards traumatic experience or chronic stress. The interaction might also exist between those two pathways.

Further understanding of this subtle relation demands follow-up study in large sample at multiple time points.

To add more bricks on previous evidence, we further described two-dimension (dread-familiarity) cognition map based on Slovic's paradigm to give more intuitive presentations. In line with our hypothesis, a noteworthy separation of five hazards was observed between two group divided by PTSD severity (four of five hazards from severer PTSD group were separately in another quadrant), which further explained the covered details in the correlation between risk perception and PTSD. In both groups, H5 (frontline HCWs who had contact with COVID-19 patients) distinctly deviates from the other four hazards, confirming the results of factor analysis of structure of COVID-19 perception. This result implicated that frontline HCWs, even they were fully exposed to COVID-19-related risk, could not appropriately (Shi et al., 2003) , which implicated general population had higher risk perception of source of the widespread infectious disease than HCWs.

Similarly, our results also supported the that sleep quality and time change of leisure activity were significantly correlated with PTSD,

respectively. Sleep quality was consistently found positively correlated with PTSD symptom. One main reason could be that sleep disturbance was one of the self-reported items in PTSD scales (Foa et al., 2016a; Foa et al., 2016b; Weathers et al., 2013) . Ered, Cooper, and Ellman (2018) suggested that depression and PTSD mediated the relationship between psychotic-like experiences and sleep quality.

Treating symptoms of depression and PTSD might improve several domains of psychotic illness (Ered et al., 2018) . Majority of the recruited HCWs reported that they had less or none time for leisure activity. Those who had more time for leisure activity, which might reflect less hours of work, lower level of COVID-19 risk exposure and more time for self-adjustment, had less self-reported PTSD symptoms.

Results from sleep quality and time for leisure activity indicated that sufficient rest and rational working shift were protective factors for psychological impact of pandemic.

The current study also found that related factors of PTSD symptom score included ethnic group, and working department (respiratory vs. other departments). HCWs of minorities had relatively high PTSD severity. This finding needs further confirmation for number of minorities was quite small. However, there have been studies reporting the reporting the relation between the identity of minorities and PTSD (Koo, Hebenstreit, Madden, & Maguen, 2016; Spoont, Sayer, Kehle-Forbes, Meis, & Nelson, 2017; Wei, Wang, Heppner, & Du, 2012) . In terms of the difference between the working department, HCWs in respiratory department directly handled the affairs of treatment and diagnosis for COVID-19 patients, with increased risk and workload.

An acute overwhelming stress might induce rapid changes in psychological status among those frontline HCWs and effect could last for a long time (Su et al., 2007) .

Although current study demonstrated some meaningful outcomes, several limitations need to be considered in this study. First, we intended to focus on the groups of HCWs with relative high risk of COVID-19 epidemic-those who were in designated hospitals. However, instead of using probability sampling method, convenience sampling is adopted to extract samples, making it difficult to generalize our results to HCWs in designated hospitals. To balance the sample at best of our ability, we both investigated frontline and non-frontline departments. Second, in order to control the length of online questionnaire, only one factor (dread and familiarity) was used in each dimension of the Slovic's risk perception cognitive model (dread risk and unknown risk). Whether two factor in our analysis could fully represent the Slovic's model still awaits further confirmation. More factorial questions were required to comprehensively evaluate the risk perception in the future. Third, there was only one single item to investigated sleep quality and time change of leisure activity which limited the reliability and validity of those investigations. Fourth, in the difference analysis, the difference between the number of participants in each subgroup is large, which could result in potential bias.

Last, the causal relation and psychological mechanism and mediators between risk perception and PTSD were not explored in this article.

Based on the significant associations between risk perceptions and PTSD in the current study, it implicated the management or intervention of risk cognition, such as health education or propaganda for strengthening the knowledge of the risk and effective prevention of COVID-19 pandemic, could help to regulate stress for HCWs who work in frontline, which further contributes to reducing psychological distress, including PTSD symptoms. Since sleep quality and time change of leisure activity both had significant relation to PTSD severity, importance should be attached to having HCWs realize the necessity to arrange enough time for daily adjustment to restore health status against adverse effect triggered by negative events.

In HCWs from designated hospitals, risk perception independently 

Risk communication during COVID-19

The posttraumatic stress disorder checklist for DSM-5 (PCL-5): Development and initial psychometric evaluation

A critical review of negative affect and the application of CBT for PTSD

The Pittsburgh sleep quality index: A new instrument for psychiatric practice and research

Psychometric properties of a singleitem scale to assess sleep quality among individuals with fibromyalgia

Real time data of distribution of COVID-19 epidemic

Daily report of situation of COVID-19 in Guandong Province

Correlation between leisure activity time and life satisfaction: Based on KOSTAT time use survey data

Mental health care for medical staff in China during the COVID-19 outbreak

COVID-19 control in China during mass population movements at new year

A multinational, multicentre study on the psychological outcomes and associated physical symptoms amongst healthcare workers during COVID-19 outbreak

Risk perception and COVID-19

WHO Director-General's opening remarks at the Mission briefing on COVID-19-2

Sleep quality, psychological symptoms, and psychotic-like experiences

The realities of risk-cost-benefit analysis

Psychometric properties of the posttraumatic diagnostic scale for DSM-5 (PDS-5)

Psychometric properties of the posttraumatic stress disorder symptom scale interview for DSM-5 (PSSI-5)

Insufficient time for leisure and perceived health and stress in working parents with small children

First case of 2019 novel coronavirus in the United States

Changes in leisure-time physical activity and physical and mental health functioning: A follow-up study

Impact of Ebola experiences and risk perceptions on mental health in Sierra Leone

Assessing the presence of post-traumatic stress and turnover intention among nurses post-Middle East respiratory syndrome outbreak: The importance of supervisor support

The mental health of medical workers in Wuhan, China dealing with the 2019 novel coronavirus

National estimates of exposure to traumatic events and PTSD prevalence using DSM-IV and DSM-5 criteria

PTSD detection and symptom presentation: Racial/ethnic differences by gender among veterans with PTSD returning from Iraq and Afghanistan

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and coronavirus disease-2019 (COVID-19): The epidemic and the challenges

Prevalence of psychiatric disorders among Toronto hospital workers one to two years after the SARS outbreak

SARS-related perceptions in Hong Kong

Sleep disturbance in post-traumatic stress disorder (PTSD): A systematic review and meta-analysis of actigraphy studies

The psychological effect of severe acute respiratory syndrome on emergency department staff

Prevalence and predictors of PTSS during COVID-19 outbreak in China hardest-hit areas: Gender differences matter

Physical distancing, face masks, and eye protection for prevention of COVID-19

Clinical Infectious Diseases: An Official Publication of the Infectious Diseases Society of America

Engagement in pleasant leisure activities and blood pressure: A 5-year longitudinal study in Alzheimer caregivers

Immediate and sustained psychological impact of an emerging infectious disease outbreak on health care workers

Diagnosis and treatment of novel coronavirus pneumonia

Predictors of sleep quality in women in the menopausal transition

The effect of control strategies to reduce social mixing on outcomes of the COVID-19 epidemic in Wuhan, China: A modelling study

COVID-19 and Italy: What next? Lancet

Risk communication: Facing public outrage. Management Communication Quarterly

History of chronic stress modifies acute stress-evoked fear memory and acoustic startle in male rats

Post-traumatic stress disorder

The risk perceptions of SARS and Scio-psychologcal behaviors of urban people in China

Perception of risk

A prospective study of racial and ethnic variation in VA psychotherapy services for PTSD

Prevalence of psychiatric morbidity and psychological adaptation of the nurses in a structured SARS caring unit during outbreak: A prospective and periodic assessment study in Taiwan

Risks to healthcare workers with emerging diseases: Lessons from MERS-CoV, Ebola, SARS, and avian flu

COVID-19 and Iranian medical students: A survey on their related-knowledge, preventive behaviors and risk perception

Psychological impact of deploying in support of the U.S. response to Ebola: A systematic review and meta-analysis of past outbreaks. Military Medicine

Immediate psychological responses and associated factors during the initial stage of the 2019 coronavirus disease (COVID-19) epidemic among the General population in China

Assessing the underlying dimensionality of DSM-5 PTSD symptoms in Chinese adolescents surviving the 2008 Wenchuan earthquake

The PTSD checklist for DSM-5 (PCL-5). National Center for PTSD

Ethnic and mainstream social connectedness, perceived racial discrimination, and posttraumatic stress symptoms

Domain-specific physical activity and mental health: A meta-analysis

Coronavirus disease (COVID-19) Situation Report-171

The psychological impact of the SARS epidemic on hospital employees in China: Exposure, risk perception, and altruistic acceptance of risk

Posttraumatic stress symptoms of health care workers during the corona virus disease

SUPPORTING INFORMATION Additional supporting information may be found online in the Supporting Information section at the end of this article. How to cite this article

The authors would like to acknowledge the volunteers who participated in the study.

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

GSY carried out statistical analyses and wrote the protocol and manuscript draft. ZYG and ZWY individually searched for relevant studies, eligibility assessments for inclusion criteria, and quality checks for the obtained studies. ZYG also participated in specific parts of data analysis and preparation of manuscript. LAN, CYH, XJY and SJ assisted with the preparation and proof-reading of the manuscript, LX designed the study, and supervised the data analysis and the writing of the paper.LWZ and ZWM provided the advices on statistical analysis.

Data in this study will not be shared, for data protection reasons and participants were informed that the data would only be used in the form of publication and would not be shared.

https://orcid.org/0000-0002-8775-6665Xu Li https://orcid.org/0000-0002-0765-5494