Working condition and health status of 6,317 front line public health workers during the COVID-19 epidemic across 5 provinces in China: a cross-sectional study

Background: Public health workers at Chinese Center for Disease Control and Prevention (China CDC) and primary health care institutes (PHI) constitute one of the main workforces for implementing prevention and control measures to contain the COVID-19 epidemic, but their efforts and health status have not been well documented. We aimed to investigate the working conditions and health status of frontline public health workers in China during the epidemic. Methods: Between February 18 to March 1, 2020, we conducted a cross-sectional survey among 2,313 CDC workers and 4,004 PHI workers in ve provinces across China experiencing different scales of COVID-19 epidemic. We interviewed all participants about their work conditions, roles, burdens, perceptions, mental health, and self-rated health using a self-constructed questionnaire and standardized measurements (i.e., Patient Health Questionnaire and General Anxiety Disorder scale). To examine the independent associations between working conditions and health outcomes, we used multivariate regression models controlling for potential confounders. Results: The prevalence of depression, anxiety, and poor self-rated health was 21.3%, 19.0%, and 9.8%, respectively, among public health workers (27.1%, 20.6% and 15.0% among CDC workers; and 17.5%, 17.9% and 6.8% among PHI workers). The majority (71.6%) participated in both eld and non-eld work and gave immense efforts. Nearly 20.0% have worked all night for more than 3 days and 45.3% had worked during the whole period of Chinese New Year. Three risk factors and two protective factors were found independently associated with all three health outcomes in our nal multivariate models, including working all night for >3 days (ORm=1.67~1.75, p<0.001), concerns about being infected at work (ORm=1.46~1.89, p<0.001), perceived troubles at work (ORm=1.10~1.28, p<0.001), starting COVID-19 prevention work after January 23 (ORm=0.78~0.82, p=0.002~0.008), and capability to persist for more than 1 month at the current work intensity (ORm=0.44~0.55, Conclusions: Chinese public health workers gave immense efforts and personal sacrices to curb the COVID-19 epidemic and were exposed to risk of mental health problems. Efforts should be taken to improve the working condition and health status of public health workers to maintain the morale and effectiveness of public health workers in the ght against the epidemic.


Background
Since December 29, 2019, the coronavirus disease 2019 (COVID-19) [1] has spread rapidly across China and now has become a global epidemic. As of March 7, 2020, a total of 80,695 con rmed cases and 3,097 deaths have been reported in China [2]. Despite some initial missteps, China implemented a series of draconian prevention and control measures which have been favorably assessed by the WHO-China joint mission [3], and may have reduced the intensity of the epidemic in China [4]. Since January 27, 2020, the daily new cases in China have been steadily declining [5], whereas community outbreaks of  have quickly emerged in many other countries, leading to stretched health systems in containing the epidemic [6].
Due to a lack of effective treatments and vaccines, preventive measures including controlling the source of infection, blocking transmission, and protecting the susceptible population are the most effective ways to contain the spread of diseases [5,7]. In an outbreak of the scale of COVID-19, China has made an unprecedented effort to mobilize its public health workers in those containment efforts. Though public health systems vary across countries [8][9][10], during the outbreak of COVID-19, different countries may share needs in COVID-19 prevention and control, and public health workers may face similar challenges. China's experiences can inform other countries of the types of public health workers who can be mobilized, the work roles and functions that they can effectively perform, the challenges that they encountered, and, in particular, the heavy toll of the work on their physical and mental health. With those lessons learned, other countries may better prepare their human resources in the ght against this epidemic.
In previous outbreaks and major public health emergencies, there were studies and reports on medical workers who joined the front line to treat patients to document their distress, di culties, and health problems [11][12][13][14]. It was found that during public health emergencies, such as SARS, medical workers had a high prevalence of depression, anxiety and perceived poor physical health [12,[15][16][17]. Increased workload, fear of infection due to patient care, inadequate training and equipment, and lack of support were risk factors for poor mental and physical health [18][19][20][21][22]. Similar ndings were reported in studies of medical workers who participated in the treatment of MERS patients [23][24][25]. However, we have identi ed no studies that focused on public health workers during outbreak control and prevention, despite their indispensable role in containing the epidemic.
China's current public health system was established after the 2003 SARS [26]. It is composed of specialized institutions (e.g., disease prevention and control, maternal and child health, mental health) for technical guidance, and primary health care institutes (PHIs) (i.e., community health centers/stations in urban areas and township health centers and village clinics in rural areas) for public health service delivery [26]. Structurally the Centers for Disease Control and Prevention (CDC) in China have units at national, provincial, municipal, and county/district levels, while PHIs operate at the town/subdistrict (i.e., community)level. During the COVID-19 epidemic, CDCs and PHIs at all levels undertook various tasks for prevention and control, such as the development of technical instructions, epidemiological investigation of patients and close contacts, surveillance of high-risk populations, specimen collection and examination, data collection and reporting, health education and promotion, training and so forth [27].
It is important to document the working and health status of public health workers in China during the outbreak and to harness China's experiences to improve the world's efforts in ghting this epidemic. In this study, we conducted a large survey among front line public health workers across 5 provinces of China, intended to address the following key questions concerning the effective mobilization and use of public health workers: (1) the roles and job functions (e.g., work contents) of public health workers involved in the epidemic containment efforts, (2) the working conditions and challenges associated with their work, (3) their perceptions (e.g., perceived support and perceived troubles) related to COVID-19 and work, and (4) factors that were associated with their mental and physical health (i.e., depression, anxiety, and self-rated health).

Study design
This cross-sectional study was conducted from February 18 to March 1, 2020. Data was collected from ve provinces (Hubei, Guangdong, Sichuan, Jiangsu and Gansu) which were purposely selected to cover different levels of epidemic severity de ned by numbers of reported cases (by March 5, 2020, 67466, 1351,   539, 631, and 102 COVID-19 cases were reported in the selected ve provinces, respectively), and to cover different regions of China (center, southern, western, eastern, and northern). In each province, 3-5 cities were selected; within selected cities, 3-5 districts/counties and 5-10 subdistricts/towns were further selected using similar methods to represent both different levels of the outbreak and to represent different regions. At province, city and district/county levels, CDCs workers were investigated, and at subdistrict/town level, PHI workers were investigated. We targeted at least 5000 public health workers, with a ratio of CDC/PHI workers of around 1/2.

Participants recruitment
Eligible participants were: 1) aged 18 years old or above; 2) working at CDC or PHI of the selected places during the study period; 3) participated in COVID-19 control and prevention-related work. Site investigators (e.g., CDC workers) of each province distributed the survey link through their Wechat/QQ working groups. Wechat and QQ are popular communication/social networking mobile-phone applications used in China that are ubiquitous in workplace settings. All participants were informed of the background, aims, anonymous nature and length (about 8-12 minutes to complete) of the survey. They were also informed that completing the questionnaire signi ed their informed consent. No compensation was provided to the participants. The study was approved by the ethics committee of the School of Public Health, Sun Yat-sen University (Reference no.: 2020-012).
Information about age, sex, job title, and whether they have children under 6 years old (i.e., the school age) were collected. Participants' areas of routine work before COVID-19 outbreak were collected in Guangdong Province. We only collected this information in Guangdong province, and it was deleted in other provinces after feedback from the pilot trial in Guangdong that the questionnaire was too long.

COVID-19 control and prevention work-related variables.
We collected information from all participants about their work in terms of work contents, readiness for the work, and time of starting the work. Detailed variables included are listed below.
1) work contents. The preset list of work contents included 14 eldwork questions covering for example face-to-face epidemiological investigation of the patients/close contacts, epidemiological investigation by phone or video call, medical observation of the close contacts, specimen collection and shipment, health education, and community-based investigation; and 11 non-eld work topics covering for example preparation of technical guidelines, data analysis and report writing, laboratory test, comprehensive coordination and publicity, and technical training. Participants were asked to select work that they have been involved in from the preset list and ll in other contents that were not included on the list if any.
3) knowledge of COVID-19 prevention and control, from 1 'adequate' to 5 'very inadequate'. 4) Date the participant started COVID-19 prevention and control work. In data analysis, we use chose the cut-off date of January 23 because it is the date of closure of Wuhan city and also it is the day just before the Chinese New Year. 5) severity of the epidemic in their provinces, from 1 'very low' to 5 'very high' according to con rmed cases in each province.
Efforts and sacri ces.
Participants were asked about their efforts and sacri ces during the outbreak: 1) number of days that they have worked all night, 2) whether they have worked during the whole period of Chinese New Year, and 3) family sacri ces such as not going home or sending children to parents' home, to avoid infecting family members.
Perceptions related to COVID-19 and work were also collected. One item was used to assess their concerns about being infected at work, from 1 'none' to 5 'very worried'. One item was used to assess how long they can persist with their current work intensity, coded as 1-2 weeks, 3-4 weeks, 1-3 months, or >3 months.
Perceived support and perceived troubles at work were measured by self-constructed items which were developed after discussions with CDC and PHI workers and among the research team. The perceived support scale consisted of three items to measure perceived support from colleagues, family, and society rated on a Likert-type scale from 1 'none' to 5 'very much'. The three items showed acceptable internal consistency in this study (Cronbach's alpha = 0.760). Perceived troubles at work consisted of ve items, which were rated on a 5-point Likert scale, from 1 'none' to 5 'very much'. For example, participants were asked how often they have been treated unfairly at work. The Cronbach's alpha for perceived troubles was 0.842 in this study.
Health outcomes included overall health status, depression, and anxiety. Overall health was measured by self-rated health status from 1 'very poor' to 5 'very good', which has been widely used globally and in China [28,29]. The 9-item patient health questionnaire (PHQ-9) was used to assess the presence of depressive symptoms. The Chinese version of the PHQ-9 has been validated for the general population, showing good internal reliability [30]. Participants were asked to rate how often they have experienced the depressive symptoms in the past two weeks, on a 4-point Likert scale, from 0 'none' to 3 'nearly every day'.
The total score ranges from 0 to 27, with a higher score re ecting greater severity. A score of ten or more was classi ed as a major depressive disorder. In this study, the Cronbach's alpha value was 0.922. The 7-item General Anxiety Disorder scale (GAD) was used to measure anxiety [31]. Each item was rated on a 4-point Likert scale ranging from 0 'never' to 3 'often (almost every day)'. The cutoff point of ten or above is used to de ne a probable case of moderate anxiety disorder. In this study, the Cronbach's alpha value was 0.937.

Statistical analysis
Descriptive analysis was conducted to characterize the study. Chi-square tests, t-tests, and rank sum tests were used to investigate differences among CDC workers and PHI workers. To explore potential factors of the three health outcomes (i.e., self-rated health, depression, and anxiety), three sets of logistic regression models were performed in parallel. First, bivariate logistic regression analyses were used to examine the associations between all variables of interest and the three outcomes. Then adjusted logistic regression models were performed to identify the associations between COVID-19 related variables (COVID-19 control and prevention work related variables, efforts and sacri ces during the outbreak, perceptions) and the three outcomes, after controlling for potential confounders (sex, age, having children under 6 years and job title). In the nal models, multivariate forward stepwise logistic regression models were tted, using all COVID-19 related variables that were found to be signi cant in the univariate analysis as candidates for selection, with sociodemographic variables entered in the model. Unadjusted odds ratios (ORu) from univariate logistic regression models, adjusted odds ratios (AOR) from multiple logistic regression models, and their respective 95% con dence intervals (CIs) were reported. IBM SPSS Statistics 25 was used for data analysis. Signi cance referred to p value <.05.

Results
Of the 7090 completed questionnaires, 528 (7.4%) did not pass the consistency checks and 245 (3.4%) did not report any COVID-19-related work. We performed the complete-data analysis based on the effective sample size of 6,317 (89.1%).

Socio-demographic characteristics of the participants
Of the 6,317 participants, 64.6% were female; the mean age was 38.7 (SD=9.43); 66.9% were 30-49 years old; 77.0% had an intermediate or Junior job title; and 27.9% had children under 6 years old (Table 1). More participants among CDC workers were male, younger and with senior job title than PHI workers. Factors associated with depression, anxiety, and poor self-rated health Signi cant socio-demographic variables associated with depression included being female (OR=1.29, Table 2), with children under 6 years old (OR=1.41, Table  2), and being aged above 30 (OR=0.63, Table 2). Only one signi cant socio-demographic variable (i.e., with children under 6 years old, OR=1.39, Table 2) was found to be associated with anxiety. Women (OR=1.13, Table 2), participants with children under 6 years old (OR=1.12,  Table 2) were more likely to have poor self-rated health. Participants from Hubei had a higher level of anxiety but not depression and poor self-rated health than those from other provinces. Adjusted for socio-demographic variables, 10 factors out of 14 factors of interest were signi cantly associated with all three health outcomes (i.e., depression, anxiety, and poor self-rated health), including three protective factors and seven risk factors ( Table 2).
In our nal multivariate models (Table 3), ve factors were found to be independently associated with depression, anxiety and poor self-   Table 2 were used as candidates of forward stepwise models, after adjusting for sociodemographic variables (sex, age, having child <6 and job title) ariables which were non-significant for all outcome variables were not included in the Table 3, including trainings on COVID-19 and send children to parents' home to avoid infecting family members. ariables with p<.05 were in bold.

Discussion
This is one of the few studies that timely documented working status and health-related conditions of public health workers during a new and emerging infectious disease epidemic. We found that the public health workers made a huge effort and personal or family sacri ce during the COVID-19 control and prevention response, with 27.1% and 20.6% having depression and anxiety. Public health workers' working conditions were associated with their health status.
Our survey found that China e ciently and effectively mobilized its public health workforce in preparation for the epidemic and the public health workers responded rapidly to the epidemic. Approximately 1 out of 5 CDC workers had already been called into the epidemic control program before January 20, 2020, even though by then most of the municipalities surveyed except for those in Hubei Province had not yet reported cases. In the following week, during the Chinese New Year, the participation rate increased substantially to 90%. Because the preventive work at the community level is coordinated by the CDCs, PHI workers' participation began slightly later than that of the CDC workers. Meanwhile, public health workers made huge efforts to curb the epidemic: about half of them worked throughout the whole period of Chinese New Year, the most important holiday in China, and one in ve had worked all night for more than 3 days by the time of this survey. Again, CDC workers had a higher rate of working during the whole holiday period and had more work-nights.
The prompt response and great efforts made by public health workers likely helped hasten the decline of the epidemic, together with other important strategies implemented by the Chinese government. Modeling studies showed that highly effective contact tracing and case isolation can control a new outbreak of COVID-19 within 3 months [32], suggesting the importance of the traditional epidemiological measures such as identi cation and isolation of infected cases, contact tracing, and health education on protective behaviors. Our survey con rmed that the Chinese public health workers were actively engaged in those traditional epidemiological roles that most likely contributed signi cantly to the quick control of the outbreak in China. As a part of the mobilization of the whole society to ght the epidemic, these results also re ect the status of many other governmental departments (such as transportation, communities, etc.) during the epidemic.
However, our survey also indicated a lack of appreciation of the communities toward the work performed by the public health workers [33]. Our data showed that support from family and colleagues is high, whereas the support from society is relatively low. Compared with physicians and nurses, the work by the public health workers is not widely understood and respected by the public [33]. During the epidemic, prevention and control work that they were engaged in, such as the isolation of close contacts (at home or designated hotels) and home inspections, was very likely to cause negative emotions and even objection in the community. Such negative emotions and con icts not only hindered preventive work but also had negative impacts on the mental health of the public health workers. Our analysis found that the perceived troubles and di culties (e.g., have been treated unfairly and concerns about routine works) at work were independent risk factors of mental health problems and poor self-assessed health after controlling for potential confounders. Therefore, to increase the understanding of the importance of public health work in the normal state and at the beginning of the epidemic is extremely important for both epidemic control and the health of public health staff.
The mental health status of public health workers during COVID-19 needs attention. Due to the lack of previous surveys on public health workers, we can only compare our results with those of medical workers, as a proxy for the public health workers. Self-reported depression (21.3%) and anxiety (19.0%) among participants in our study were substantially higher than those of hospital medical staff in daily work in China (18.3% and 14.7%), using the same measurements and cut-off points [34]. Furthermore, in our study, a larger proportion of CDC workers reported mental health problems and poor physical health than PHI workers, as they might bear more responsibilities in regional disease prevention [26], and had longer working hours. In studies targeting medical workers during SARS epidemic, 38.5% of hospital workers dealing with SARS patients developed symptomatic depression [12], while in MERS-Cov epidemic, two thirds of medical workers reported psychological problems [24]. Although ndings of these studies are incomparable as different measurements were used, front line medical doctors and nurses treating patients may have higher prevalence of mental health problems than the public health workers. In addition to the immediate personal welfare concerns of these health works, mental health problems can in turn affect their work [35], so mental health care is greatly warranted for these frontline health workers.
Longer working time and worrying about being infected were associated with mental health and self-rated health status, which were similar to the ndings of other studies [19,20]. Those involved in non-eld work were more likely than others to have probable depression and poor self-reported health. It is possible that time-consuming paperwork, data analysis, and laboratory work, especially under emergency conditions, add to people's physical and mental health burden [36]. Our data also found that participants involved in non-eld work had longer working hours. The health status of public health workers working at o ces or laboratories in outbreak control should not be neglected.
Our survey suggests that to reduce stress and improve their commitment to work, child care support, which had been reported to be one of the top support needs among health care employees working in a disastrous event, is highly warranted in the future [37]. Approximate one third of the surveyed public health workers were parents of young children (28%). For health care workers, child care obligations can make it harder to commit to frontline work in the event of a catastrophic event [38]. A survey on health care workers who participated in the SARS epidemic response showed that living with children was signi cantly associated with an increased level of concern for personal or family health [39]. The high volume of work and concern about personal and family health can lead to burnout [40]. Other countries in preparation for the epidemic control must consider those support measures to ensure an effective public health workforce.
The epidemic outbreak is likely to lead to constrain the health workforce. In our study, we found 63% PHI workers involved in COVID-19 prevention were nonpublic health workers in their routine work. They were recruited into public health work from different elds (i.e., 27.6% clinicians, 26.6% nurses, and 5.1% pharmacists). Through some rapid training, they appeared to effectively take on the new role in community-based prevention during the outbreak.
Furthermore, we found the training rate reached 98% and most training was conducted through web-based online modules. Meanwhile, in epidemiological surveys, the smartphone-based online survey was widely used in the eld, suggesting that mobile technologies could improve the e ciency in outbreak prevention. China's experiences suggest that task-shifting in public health work is possible in an epidemic crisis, particularly when timely training and the use of technology are concurrently provided.
There were several limitations to this study. First, selection bias may exist as this study used a non-random sample. Although our strati ed sampling covered provinces with different levels of the epidemic from different regions of China, ndings of this study might not be generalizable to all public health workers in China. Second, reporting bias may exist due to the nature of self-reported data. As the current study used an online questionnaire, people who are less skilled or have no access to the internet and smartphones may be underrepresented. Also people experiencing work strain may be less likely to engage with their Wechat group, and thus the people most at risk might not be recruited in this study. Since online surveys were used in this study, the response rate may not be high. Third, the measurements of perceived support and perceived troubles at work were self-constructed and have not been validated, although the Cronbach's alpha for these self-constructed measurements was high for this study. Fourth, participants' areas of routine work were only investigated in Guangdong province. Though answers in other provinces may be similar under the national prevention strategy, this nding might not be generalizable to other provinces. Fifth, we cannot derive any conclusive causal relationship from a cross-sectional design.
Despite those limitations our study is a rare study focusing on CDC employees and PHI workers. It is critical to understand the experiences of those workers as they are the backbone of pandemic control efforts. We also examined not only their physical health but also their mental health, and their ability to maintain their workload, all of which are crucial components of an effective pandemic response. The survey was also conducted in a timely fashion, which reduced recall bias and provided urgently needed information. Lastly, although we did not have a random sample, we did reach a large number of workers within a short period of time. Our study suggests the government should build into their pandemic preparedness plan the components that will provide needed mental and physical support to the PHI workers. Our study shows that pandemic response work can be exhausting and associated with mental and physical distress, but also that workers from outside of public health can be effective in this role, providing evidence that rapidly training non-public health workers to step into pandemic prevention roles can be an effective and valid strategy for expanding the pool of public health workers during a crisis, improving pandemic responsiveness while protecting staff wellbeing. It would also be useful to improve the coordination and integration of public health workers and PHI workers.

Conclusions
In summary, this study is among the rst to document the working situation and health status in a relatively representative large sample of public health workers during the COVID-19 epidemic. With the rapid spread of the COVID-19 epidemic globally, all countries with different public health systems will face serious challenges. It is important to provide the necessary support to public health workers to ensure their health and working conditions, which is also the key to control the epidemic.

Declarations
Funding This study is supported by the National Science and Technology Major Project of China (grant ID 2018ZX10715004). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.