Factors associated with burnout among Chinese nurses during COVID-19 epidemic: a cross-sectional study

In December 2019, Novel coronavirus pneumonia began erupting in Wuhan, Hubei Province, as the quake's epicenter. As the epidemic has developed, it has spread around the world. Since the outbreak of the COVID-19, China's national and local government leaders have worked together to take a series of effective prevention and control measures. On 23 January 2020, Wuhan City COVID-19 Prevention and Control Center issued a circular requesting the city to suspend operations and cut off transmission lines [1]. Hospitals play an irreplaceable role in prevention and treatment as effective treatment institutions. As of 28 February, more than 90 national and provincial medical teams, about 13,000 senior healthcare workers had gathered in Wuhan to work with local medical staff to treat patients [2].

Nurses are the most significant part of the healthcare professionals, and they have played an extraordinary role in combating COVID-19. During the pandemic, especially the initial stage, nurses struggled with insufficient resources, intensive work, and uncertainty. In their daily work, they witness the pain and suffering of the patients [3], wear cumbersome personal protective equipment (PPE), and perform tedious and detailed care for confirmed patients in isolated areas. In addition, they are at a higher risk of contracting the virus than others. Because of the possibility of transmitting the virus to families and traffic restrictions, most medical personnel could not return home and had to stay in hospital dormitories and hotels. Therefore, facing such high levels of psychological pressure and work-related stress, they experience mental health problems such as depression and anxiety [4] and experience fatigue, burnout, and even consider quitting their jobs [5, 6].

Burnout is a state of mind caused by chronic emotional or psychological stress [7]. Maslach described this condition as emotional exhaustion (the state of being emotionally drained), depersonalization (the loss of compassion and concern), and a diminished sense of personal accomplishment at work [8, 9]. On 28 May 2019, the World Health Organization (WHO) declared in the 11th revision of the International Classification of Diseases (ICD-11) that burnout is a "professional phenomenon", a chronic workspace stress syndrome that has not yet been successfully managed [10]. Previous studies on epidemics have shown that frontline nurses have different levels of burnout. In the Hu et al [11]. study, about half of frontline nurses reported moderate and high work burnout, and the emotional exhaustion rate was 60.5%, depersonalization was 42.3%, and personal accomplishment was 60.6%. In the Zhang et al. [12] study, 78.5 and 92.5% of nurses presented mild levels of emotional exhaustion and depersonalization, respectively, while participants experienced a severe lack of personal accomplishment. It is related to nurses’ job satisfaction and patients’ outcome [13].

Previous studies showed that burnout factors mainly contain sociodemographic and work-related factors. Sociodemographic factors mentioned in previous studies include sex, age, marital status [14, 15], education [16, 17], having children, living with family [18], personality [19], etc. Work-related factors include employment status [14], work experience [15], work department [20‐22], type of shift [14, 15], salary level [23, 24], social support [17, 25, 26], and lack of material resources [25], job stress [16, 25, 27], coping strategies [15, 27], job security and satisfaction [15], education and training [27]. Studies have shown that burnout is related to a range of self-reported physical and psychological stressors, such as anxiety and depression [11], fatigue [28].

However, considering the peculiarity of COVID-19, the level and factors of burnout among nurses at the epicenter are not fully understood, especially concerning different facets of the work environment and concomitant psychological responses, such as anxiety. Therefore, the purpose of this study is to investigate the status of burnout among nurses and its influencing factors during the COVID-19 epidemic to provide intervention decisions for managers.

The study was conducted from 19 to 25 February 2020, and a self-reported questionnaire was distributed anonymously online to 1011 nurses. Participants have worked in Wuhan tertiary hospitals for at least one week since the epidemic. They were informed of the purpose, and their participation was interpreted as informed consent, and all data were guaranteed confidentiality. Questionnaires with missing values or similar answers for most items were excluded. The final number of valid questionnaires was 885, and the effective response rate was 87.5%.

Burnout was measured with the Maslach Burnout Inventory-General Survey (MBI-GS), developed by Maslach et al [8]. and revised by Li et al. [29]. And its validity has been verified [30]. The 15-item questionnaire contains five questions on emotional exhaustion, four on cynicism, and six on personal accomplishment. Each item is a self-reported question made up of a 7-point scale (0–6 points from "never" to "every day"). The higher score of emotional exhaustion and cynicism is, the higher level of burnout will be. Emotional exhaustion level can be interpreted as mild (0–16), moderate (17–24), or high (≥ 25); cynicism dimension is classified as mild (0–7), moderate (8–11), or high (≥ 12). The score in personal accomplishment is the opposite and is classified as mild (≥ 16), moderate (12–15), or high (0–11) based on item scores [31]. In this study, the Cronbach's a of the MBI-GS was 0.854. The Cronbach's a for three dimensions were 0.957, 0.929, and 0.936, respectively.

The State-Trait Anxiety Inventory (STAI) was developed by Spielberger et al [32]. And it was revised by Li and Qian [33]. This measure consists of two subscales: State Anxiety Inventory (SAI) and Trait Anxiety Inventory (TAI). The TAI reflects the relatively stable aspect of “anxiety proneness”, while the SAI assesses the current state of anxiety or feeling now [34]. It consists of 40 items, each scored on a 4-point scale (1–4 points), with 20 items for each subscale. The total score is 20–80. The higher the score, the greater the anxiety. It has better reliability and validity [34, 35]. In our study, the Cronbach’s a of SAI was 0.925.

Data analysis was performed using IBM SPSS version 25.0 for windows [36]. Descriptive statistics were used to summarize sample characteristics and outcome variables. Nurses’social-demographic, work-related, and categorical outcome variables were shown as number (N) and percentage (%). Continuous outcome variables were shown as mean and standard deviation (SD). Multivariate Linear Regression analysis was performed to detect independent factors related to anxiety and burnout. The correlation between burnout and anxiety was examined by Pearson correlation. A two-tailed P < 0.05 was identified as statistically significant.

The potential range score, mean values, standard deviations (SD), and the lowest and highest scores of the subscales are listed in Table 2. The prevalence of low, moderate, and high emotional exhaustion, cynicism, and personal accomplishment is shown in Table 3.

The linear correlation between burnout scales and the anxiety scores was determined in Table 3. The emotional exhaustion scale presented statistically significant correlations with SAI and TAI scores, and statistically significant relationship was observed between cynicism and SAI and TAI items. Similarly, significant correlations were found between personal accomplishment and SAI and TAI scores. There was a significant positive correlation between emotional exhaustion and cynicism but no significant correlation with personal accomplishment. Cynicism was negatively related to personal accomplishment (Table 4).

Multiple linear regression analysis was compiled for each dimension of the burnout scale. For emotional exhaustion, the variables personnel agency (β = -2.01, P < 0.01), five years or less working experience (β = -1.75, P = 0.001), living in hospital dormitory (β = -1.86, P = 0.001), three or more night shifts weekly (β = 2.43, P < 0.001), the best (β = -3.24, P < 0.001) and better (β = -1.25, P < 0.05) level of knowledge of COVID-19, and having confirmed or suspected medical staff around (β = 2.05, P < 0.001) were statistically significant predictors. For cynicism, the variables intermediate title (β = 1.21, P < 0.01), personnel agency (β = -1.72, P < 0.01), working in isolation ward (β = 1.04, P < 0.05), living in hospital dormitory (β = -1.31, P = 0.001), 3 or more night shifts weekly (β = 2.10, P < 0.001), the best level of knowledge of COVID-19 (β = -1.83, P < 0.001), and having confirmed or suspected medical staff around (β = 1.71, P < 0.001) were predictors. Finally, for personal accomplishment, no siblings (β = -1.73, P < 0.05), Wuhan medical team (β = 1.37, P < 0.05), living in hotel (β = -1.33, P < 0.05), nine or more hours of daily work (β = 3.37, P = 0.001), the best (β = 5.67, P < 0.001) and better (β = 2.62, P = 0.001) level of knowledge of COVID-19, and having confirmed or suspected medical staff around (β = -1.68, P = 0.01) were predictors (Table 5).

This study aimed to investigate the levels of anxiety and burnout of nurses working in Wuhan during the COVID-19 epidemic and analyze factors affecting burnout. The findings showed that frontline nurses had severe anxiety and mild to moderate burnout. The influencing factors of burnout were analyzed based on social demography and work-related characteristics, and finally, 12 variables were proposed. It will provide a reference for nursing managers and policymakers, especially considering that the epidemic will be a long process. Local or national longitudinal studies will be further conducted to understand nurses' occupational psychological dynamics. And it is immediate to perform individual or group psychological interventions. Make it possible to establish a stable and harmonious nursing team for challenges in the future.