Factors Associated With Circadian Rhythm, Job Burnout, And Perceived Stress Among Nurses In Chinese Tertiary Hospitals: A cross-Sectional Study

Background: Shift work is a common work pattern for nurses worldwide. Circadian rhythm dysregulation affects the quality of nurses’ work and their physical and psychological health. Eveningness and occupational stress might be associated with burnout syndrome. Methods: A Cross-sectional study design. Between July and September 2020, 23 tertiary hospitals were randomly selected from among 113 tertiary hospitals in Hunan Province for participation in this study. At least 25% of the nurses working in each hospital were targeted. Nurses’ circadian rhythm, job burnout, and perceived stress were evaluated. This paper complies with the STROBE reporting guideline for cross-sectional studies. Results: A total of 28.1% and 17.6% of nurses reported eveningness type and morningness type, respectively. The scores for emotional exhaustion, depersonalization, and perceived stress of eveningness nurses were higher than those of their morningness counterparts. Eveningness nurses also reported a lower sense of personal accomplishment (P<.05). Predictors of MEQ score included age 36-40 years, widowed or divorced, BMI ≥ 25 kg/m 2 , history of chronic disease, working for 11-20 years or 21 years and above, emotional exhaustion, personal accomplishment, perceived stress. Conclusions: A large proportion of nurses experience circadian rhythm dysregulation. It is necessary to understand the predictors and inuencing factors of circadian rhythm dysregulation in nurses to relieve their job burnout and reduce perceived stress.

Emotional exhaustion, depersonalization, and a reduced sense of personal accomplishment are the three aspects of tridimensional syndrome, which can be de ned as burnout (Maslach & Jackson, 1981). Nurses are exposed to occupational burnout in the clinical environment due to shortages of medical resources and nursing manpower and increased numbers of patients (Orgambídez-Ramos et al., 2017). Burnout is more prevalent among nurses than among workers in other professions (Gómez-Urquiza, De la Fuente-Solana, et al., 2017). Guo et al., (2019) reported that 74.1% of Chinese nurses have experienced burnout. Burnout can affect professionals in many occupations associated with helping others, with nurses being the most at risk due to potential impacts on safety and quality of clinical care (Gómez-Urquiza, Vargas, et al., 2017). On the other hand, burnout is also associated with resilience and turnover intention in nurses (Guo et al., 2019;Harker et al., 2016). Burnout syndrome in the nursing profession is also considered a risk factor for turnover intention and job dissatisfaction (Jiang et al., 2017). (Hosseinabadi et al., 2019) suggested that it is necessary to reduce burnout syndrome in shift nurses and identify risk factors that in uence individual differences in nurses' circadian rhythms.
"Perceived stress" is de ned as a person's response to an environment that is perceived as threatening to his or her abilities and health (Richard & Susan, 1984). Perceived stress refers not only to measurable stress itself but also to how a person feels about stress. Thus, the perception of a stressful condition is more essential than the objective measurement of the stressor and can impact a person's performance (Abdollahi et  A previous study also showed that differences in circadian rhythm and occupational stress can cause burnout syndrome (Hosseinabadi et al., 2019). Although burnout and stress have been assessed separately in Chinese nurses to some extent, few studies have investigated the relationship between circadian rhythm, burnout, and perceived occupational stress in Chinese tertiary hospital nurses. The results of this study can assist nursing managers in better managing circadian rhythm dysregulation among Chinese nurses. This study could also help improve both the physical and mental health of nurses, reduce turnover intention, enhance nursing retention, and provide a basis for effective interventions. The purposes of this study were to (1) evaluate chronotype, job burnout, and perceived occupational stress among Chinese tertiary hospital nurses and (2) understand the predictors of circadian rhythm in this group of nurses.

Methods
This study is compliant with Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline for cross-sectional studies (Supplementary File 1).

Design
A cross-sectional study of multi-strati ed, grounded, random sampling method was used to select the nal sample of nurses.

Participants
The inclusion criteria were as follows: (1) nurses on duty; (2) registered nurses with at least 1 year of working experience; and (3) those who provided informed consent. Participants were excluded if they were (1) in an internship, undergoing advanced training and study or on vacation during the study period; or (2) had experienced a major stressful event or suffered from a serious physical or mental illness. All subjects voluntarily participated in this study.

Date Collection
Between July and September 2020, 23 tertiary hospitals were randomly selected for this study from among 113 tertiary hospitals in Hunan Province. Twenty-ve percent of the nurses working in each hospital were targeted for selection. A self-administered questionnaire was developed by researches. It was sent to the participants online or o ine.

Demographic information
We collected the following demographic data from each participant: sex, age, BMI, education, marital status, number of children, comorbid chronic diseases, professional title, working year, job satisfaction, and sources of stress.

Maslach Burnout Inventory-Human Services Survey (MBI-HSS)
The MBI-HSS has been widely applied in studies of burnout among members of healthcare groups. The MBI-HSS consists of 3 dimensions: (1) Emotional Exhaustion (EE), to evaluate the emotional reaction to work stress; (2) Depersonalization (DP), to assess work stress-induced attitudes and feelings toward patients; and (3) Personal Accomplishment (PA), to appraise the effect of work stress on the perception of one's own work (Zhang et al., 2006). EE and DP are positive scores; that is, the higher the score, the more serious the job burnout is. EE and DP scores range from 0-42 and 0-18, respectively. The PA score ranges from 0 to 42 and is reverse scored; the lower the score, the more severe the burnout is. The Cronbach's α coe cient was .828, and the Spearman-Brown coe cient was .823 (Zhang et al., 2006).

Perceived Stress Scale (PSS)
The PSS was used to assess the level of stress perceived by the participants. It was designed by (Cohen et al., 1983) and translated into Chinese in 2003 (Yang & Huang, 2003). The PSS consists of 2 dimensions: a sense of lack of control and tension. The PSS also includes 14 items scored on a 5-point Likert scale, ranging from 0 (never) to 4 (always). The total score ranges from 0 to 56, with higher scores indicating higher perceived stress.
The PSS has good reliability, with a Cronbach coe cient of .780 (Yang & Huang, 2003).

Pittsburgh Sleep Quality Index (PSQI)
The PSQI was developed to measure the quality and patterns of sleep (Buysse et al., 1989). It consists of 19 self-

Ethics and Consent to Participate
This study was approved by the IRB of The Third Xiangya Hospital, Central South University and supported by the heads of the nursing departments of the various hospitals. Moreover, the participating nurses were asked to give their written consent before completing the required data collection forms. The consent forms informed all participants of the purpose, processes, bene ts, and potential risks of this study. The participants had the right to decide whether to participate and could withdraw at any point in time. The privacy of the participants was rigorously protected, and no identifying information was collected.

Statistical analysis
Data from different hospitals were analyzed using SPSS 24.0 (IBM Corp., Armonk, NY, USA). Continuous variables are described by the mean and standard deviation, whereas frequency and percentage are used for categorical variables. Pearson's correlation analysis was used to analyze the degree of association between 2 variables. The correlations between demographics, pressure sources, perceived stress, burnout, sleep quality, and circadian rhythm were assessed using the chi-square test and one-way ANOVA. Multiple linear regression analysis was employed to explore the predictions of circadian rhythm, accounting for all variables, and categorical data were transformed into continuous variables by creating dummy variables. Alpha < .05 indicated statistical signi cance.

Demographic information
Out of 3,050 nurses from the 23 hospitals invited, 2,780 nurses completed and returned the questionnaire. As a result, the overall response rate was 91.15%. The participants were predominately female (97.7%). More than 80% of participants had a bachelor's degree, and the BMI range was 18.5-24.9 for 77.7% of the participants.
Approximately 70% of nurses were married and had one or more children. Approximately one-fth of participants had a chronic disease. Regarding job-related factors, only 7.8% had obtained a senior professional title. With regard to working experience, most of the respondents had worked for 5-10 years (33.6%) or 10-20 years (33.8%).
A total of 46.2% of the participants were satis ed with their job (See Table 1).
Chronotype, burnout, perceived stress, and sleep quality of the nurses As shown in Table 1, the eveningness type and morningness type of circadian rhythm were reported by 28.1% and 17.6% of participants, respectively. Regarding sources of stress, 82.4% of the participants expressed that late-night shift work was a source of stress. In addition, income, workload, and worktime were indicated as sources of stress by 56.2%, 53.9% and 51.8% of participants, respectively. For the emotional exhaustion and depersonalization dimensions, morningness participants scored 17.44±9.039 and 4.99±4.123, respectively. In addition, the personal accomplishment score was 21.79±8.166. The average PSS score was at a normal level (27.12±8.166), although the perceived stress levels were higher for eveningness (28.56±8.15) than morningness (25.43±8.45), which had an ordinary level. For the seven dimensions of sleep quality, participants belonging to the eveningness type presented higher scores on the PSQI and its subscales (P<.05).

Bivariate analysis of circadian rhythm
Except for sex and BMI and other individual characteristics, all job-related factors were signi cantly associated with circadian rhythm. Nurses who were younger, unmarried, without children, and did not participate in sports, had a lower level of educational background and professional title, had fewer years of service, were dissatis ed with their job, and reported their sources of stress to be from late-night shift work and their workload, work time, income, work di culty, and marital relationship were signi cantly more likely to be eveningness types. Table 2 summarizes the correlations between different variables.

Multiple linear stepwise regression analysis
For nurses who were 36-40 years old, had worked for 11-20 years or 21 years or more, and who did not report parent-child relationships as sources of stress, PA scores were positive predictors of the MEQ score. In contrast, for nurses who were widowed or divorced, without children, did not participate in sports, and had a BMI ≥25 along with a chronic disease, EE scores, perceived stress, hypnotic drug use, sleep latency, and sleep duration were negative predictors of the MEQ score. A total of 19.2% of the variance in the circadian rhythm was explained by the above variables (F=44.955, P<.001, R 2 =0.196, Adjusted R 2 =0.192) ( Table 3).

Discussion
In this study, the mean MEQ score was at the lower end of the median range (54.41±8.92). This may be related to the professional characteristic that shift work is a common pattern for nurses (Moreno-Casbas et al., 2014). The participants with an eveningness type of circadian rhythm accounted for 28.1% of the study group, which represented a higher percentage than the morningness type (17.6%). Approximately 82.4% of the 2,780 nurses we surveyed were required to work the late-night shift. Shift work, on the other hand, tends to disrupt nurses' circadian rhythms (Niu et al., 2015). Eveningness was related to a higher score on the PSQI and its subscales, Perceived stress was inversely related to circadian rhythm among nurses who participated in this study. This is in line with the mechanism revealed by former studies. The stressor of late-night shift work was also signi cantly related to chronotype. Law & Clow (2020) reported that disruption of the circadian rhythm of cortisol secretion was caused by stress. Roeser et al., (2012) indicated that "night owls" have higher levels of perceived stress than "early birds" among university students. In this study, we also surveyed nurses about their sources of stress.
Workload, work time, work di culty, income, and marriage relationships were identi ed as stressors that were signi cantly associated with circadian rhythm. Robat et al., (2021) also found that some of these stressors affect hospital nurses. The stressors workload, work time, and work di culty can cause nurses to work overtime to complete their work, which leads to delays in daily life activities.
Regarding job-related factors, professional title, working years, and job satisfaction were all signi cantly correlated with circadian rhythm. Our study showed that working 11-20 years as well as 21 years or more were predictors of MEQ score. In China, the working years of nurses is positively correlated with professional title; that is, the longer nurses have worked, the more senior their title is. Therefore, experience can help reduce