Status and related factors of burnout among palliative nurses in China: a cross-sectional study

A cross-sectional design with convenience sampling method was performed in this study. From May 2021 to September 2021, we investigated a total of 340 palliative nurses from 25 hospitals or healthcare institutions in 8 cities, and 331 questionnaires were recycled; the response rate was 97.35%. Two independent researchers used an author-designed Excel spreadsheet for data entry, and this was verified by a third researcher. This study was part of our cross-sectional study about palliative nurses in China, other details about participants’ selection process please see our previous work [23].

After acquiring permissions to conduct this study from the nurse managers of each selected hospitals or healthcare institutions, two well-trained researchers assigned the questionnaires to palliative nurses by face. Subjective information was collected in the form of paper questionnaires, and objective information was obtained through the personnel management system with the help of human resource managers. Staff meetings were regularly conducted under the lead of nursing managers in each participating palliative unit to promote the data collection process and to ensure the response rate. The signed informed consent forms and the completed questionnaires were required to return together and were accessible only to researchers of this study.

The sample size was analysed using G*power 3.1 software, and the linear multiple regression algorithm was selected. There were 21 variables, including 12 sociodemographic characteristics and nine scale-associated dimensions, in this study. With 95% confidence intervals and 0.8 power, the minimum sample size in our study was 160. Considering a possible 15% wastage rate, a total of 188 participants were needed.

We used SPSS 26.0 for data analysis. Skewness and kurtosis were used for normality tests. The variables were determined to have a normal distribution when the skewness value ≤ 2 or the kurtosis value ≤ 4 [29], and we found that our data (scores for emotional exhaustion, depersonalization, reduced personal accomplishment, pessimistic personality, social support, self-efficacy and coping style) were normally distributed. The independent-sample t test (dichotomous variables) and one-way ANOVA (polytomous variables) were conducted for univariate analysis. Correlations among dimensions of subscales were analysed using Pearson correlation analysis. Finally, multiple linear regression analysis was carried out to separately test the related factors of emotional exhaustion, depersonalization and reduced personal accomplishment dimensions. We chose stepwise regression analysis to minimize the multicollinearity effect (“the alpha to enter” and “the alpha to remove” were set as 0.05 and 0.1, respectively). The multicollinearity in this regression model was checked using tolerance, and the variance inflation factor (VIF), with tolerance < 0.1 or VIF > 10, indicated multicollinearity. The result was considered statistically significant when the two-tailed p value was less than 0.05.

As shown in Table 1. The mean age and the mean years of work experience of the 319 nurses were 30.88 ± 7.22 years and 5.88 ± 5.9 years, respectively. Most of the participants were female (98.43%), and the majority were married (62.70%). The most common educational level was undergraduate or above (53.61%), and the most common professional title was junior nurse (73.89%). Only one-third of participants (31.39%) reported being satisfied with their income.

The mean scores of burnout total and emotional exhaustion, depersonalization, and reduced personal accomplishment dimensions were 23.4 ± 7.68, 8.66 ± 3.04, 7.46 ± 2.58, and 7.29 ± 2.83, respectively. The mean score of pessimistic personality was 30.89 ± 3.85. In addition, the mean scores of social support, self-efficacy and resilience were 61.26 ± 11.73, 23.36 ± 5.30, and 57.12 ± 14.60, respectively, suggesting that palliative nurses in this study generally had high levels of social support and moderate levels of self-efficacy and resilience [25‐27]. For the coping style, the mean scores for the positive dimension and negative dimension were 2.00 ± 0.59 and 1.32 ± 0.65, respectively, which means that most palliative nurses tend to adopt a positive coping style [28].

Bivariate analysis results suggested that educational level, health condition, family support, income satisfaction and end-of-life care training were statistically significantly associated with all three dimensions (p < 0.05) (Table 2).

The results of Pearson correlation analyses are presented in Table 3. Among the three subscales, the correlations were statistically significant (r = 0.699, r = 0.742, r = 0.762, all p < 0.01). Among the three dimensions, there were five common influencing factors. Specifically, pessimistic personality and negative coping style were positively correlated with emotional exhaustion, depersonalization, and reduced personal accomplishment (r > 0, p < 0.01). In addition, social support, resilience and positive coping style were negatively correlated with three dimensions (r < 0, p < 0.01), while no statistically significant correlations were found among the three dimensions and self-efficacy, age and years of work experience.

In this study, we included variables with statistically significant correlations in the bivariate analysis and Pearson correlation analysis into multiple linear regression analyses for the three dimensions. In addition, age was included as a common confounding variable [30]. The multiple linear regression results of the relationships among the three dimensions and selected variables in palliative nurses are listed in Table 4.

The results indicated that resilience, coping style, and pessimistic personality were common related factors of three dimensions (p < 0.05), with resilience (β = -0.325, β = -0.244, β = -0.179, all p < 0.01) and positive coping (β = -0.105, β = -0.113, β = -0.219, all p < 0.01) presented as negative factors, while pessimistic personality (β = 0.230, β = 0.166, β = 0.179, all p < 0.01) and negative coping (β = 0.370, β = 0.338, β = 0.380, all p < 0.01) were presented as positive factors. Regarding the emotional exhaustion model (R ² = 0.399, adjusted R ² = 0.386, F = 29.539, p < 0.01), income satisfaction (β = -0.143, p = 0.001) and end-of-life care training (no compared yes, β = 0.088, p = 0.048) were also related factors, and a total of 38.6% of the variance could be explained. In the depersonalization model (R ² = 0.291, adjusted R ² = 0.278, F = 21.367, p < 0.01), social support also reflected a related factor, and 27.8% of the total variance was explained. In the reduced personal accomplishment model (R ² = 0.360, adjusted R ² = 0.345, F = 24.961, p < 0.01), income satisfaction and social support were also revealed as related factors, and a total of 34.5% of the variance was explained. There was no multicollinearity found in the three models.

Findings from correlations between three subscales revealed that palliative nurses who experience emotional exhaustion also experience depersonalization and reduced personal accomplishments, which was consistent with the findings of a previous study [31]. In addition, compared with studies using the same burnout scale, the scores in this study seemed higher than those of Chinese general nurses [32]: 21.65 ± 6.09, p < 0.05). In addition to the influence of sample size and the variation across individuals, there may be other reasons for these differences. First, the risk of burnout in medical staff has been demonstrated to be strongly correlated with the frequency of contact with death and suffering [33]. Nurses working in palliative units experience the frequent salvage, hospice care and death of patients, as well as constant grief, helplessness and even despair [34], which places them at a high risk of burnout. In addition, the specific cultural context may make it difficult for palliative nurses to obtain favourable social support. The relatively slow development of Chinese palliative care can also lead to the lopsided coverage and imperfect discipline development in this field [14]. As a result, palliative nurses in China seem to be a more vulnerable group of job burnout than general nurses. Thus, it is urgent to identify burnout among palliative nurses and provide effective strategies to cope with it.

In this study, coping strategy was strongly associated with emotional exhaustion, depersonalization, and reduced personal accomplishment, which was consistent with the findings of previous studies [35, 36]. Coping is the cognitive and behavioural reactions exhibited to master, tolerate, and reduce stress [37]. Palliative nurses who adopt negative coping strategies such as cognitive avoidance can experience negative outcomes, including psychosomatic symptoms [38] and immune and endocrine system damage [39]. However, positive coping, such as problem solving, positive attitudes, cognitive flexibility, and optimism, can help them be less emotionally exhausted, depersonalized and complaint prone [35, 40], thus reducing their vulnerability to burnout. Therefore, coping strategy training targeted to end-of-life care should be conducted widely with nurses working in palliative units.

Resilience is another strong protective factor that has been found in many other studies as well [20, 21]. Resilience is defined as the ability to cope with adverse circumstances and minimize distress [41]. Nurses with good resilience may regard problems as a normal part of life and adopt ways of thinking that lessen the impact of clinical dilemmas [40]. Moreover, the development of resilience can also facilitate the adoption of positive coping [42], which can reduce burnout synergistically. Thus, effective interventions need to be explored and used to cultivate internal resilience and help palliative nurses survive in high-stress settings.

In addition, personality traits such as neuroticism, negative affectivity, and cooperativeness were demonstrated to have a negative influence on burnout by influencing the way people experience and cope with stressors [43]. Considering this, palliative nurses with a pessimistic personality may be less likely to be happy at work because they tend to interpret their work environment as more negative. They more easily suffer further step distress and social disengagement and are caught in a downwards spiral of burnout [44]. However, as one’s personality is difficult to change or modify, more intention can be focused on other factors when considering practical intervention, and it might be necessary to take personality as an admission qualification of palliative nurses.

Moreover, in this study, the self-rated health condition also revealed strong relations. This function could be explained by the interaction between burnout and health outcomes. According to current review [45], burnout can directly affect health conditions through neuroendocrine mechanisms to deplete the resources necessary for coping, leading to negative health outcomes such as fatigue, depression, and somatization. As for health problems caused by burnout, nurses may not be able to achieve the desired work performance, thus forming a vicious cycle [46]. The impact of burnout on health outcomes found in this study highlighted the importance of identifying health conditions early and adopting preventive interventions in palliative working environments.

Palliative nurses who had accepted end-of-life care training received lower emotional exhaustion scores in this study, which highlights the importance of job training in palliative care. Palliative care work requires strong professional knowledge and skills. However, nearly 20% of the participants in this study reported they have received no end-of-life care training, which might result in incompetent and ineffective care and a lack of emotion, communication, and stress management skills [35]. In addition, under the influence of Confucian culture, people in China rarely talk about death, which might indirectly result in the shortage of palliative medical staff and the slow development of palliative care in China. Therefore, death education should be fully covered for palliative care staff in order to adjust their attitude towards hospice care and death, and it is urgent to extend this out to the public to improve their cognition and understanding of and support for palliative work and thus further promote the development of Chinese palliative care.

In this study, social support and income satisfaction also showed statistically significant correlations with burnout. For nurses, the relationship between healthcare-related occupational stress and burnout can be mediated by perceived social support [47]. In addition, for palliative nurses, support from the work environment, such as perceived organizational support and peer support, seemed especially effective for decreasing their burnout [22]. Thus, it is noted that nurse managers in palliative units should provide support resources for palliative nurses and create a harmonious working atmosphere to help them deal with stress and difficulties.

Past review has confirmed that salary is an important factor for nurses’ job satisfaction [48]. In the US, nurses are paid an average of 98,190 USD/year [49]. However, in China, nurses are paid only 9,500–15,100 USD/year [18], revealing a notable gap in nurse salaries between China and developed countries. In our study, palliative nurses were generally paid 5,600–9,500 USD/year, which was even lower than the Chinese average. The disparity between pay out and income surely contributes to their risk of burnout and intention to leave the field [18]. Thus, a carefully considered and reasonable salary system needs to be established to gradually improve palliative nurses’ job satisfaction.

Some other factors did not manifest significance in this study, including age, educational level, and self-efficacy. It was noted that nurses with older age are vulnerable to burnout due to the reduction of physical and psychological work abilities [50]. However, the majority of nurses in this study were at young age (30.88 ± 7.22), and age did not seem to be the main reason for their burnout. In addition, this study coded educational level as a dichotomous variable (junior college/ undergraduate and above), failing to clarify a more detailed relationship between educational level and burnout, and further studies with larger sample sizes should be conducted to identify such relationships in palliative nurses. Some publications have explained that nurses with higher self-efficacy experienced lower level of burnout [51, 52], which was not detected in this study. This might due to the reason that Chinese palliative nurses take on a relatively overloaded caring tasks, and it is hard for them to overcome work stressors through themselves’ personal abilities [53].

Our results suggest that training programs directed at the Chinese cultural context, including death education, coping strategies training, and resilience cultivation, are badly needed to help palliative nurses adjust their cognition and attitudes regarding palliative care, as well as to improve the quality of nursing care and decrease burnout among such nurses; this point is important for nursing managers in formulating an effective training system. Second, palliative nurses with pessimistic personalities and poor health conditions in this study were more likely than other nurses to experience burnout. As such factors were relatively stable and are hard to change [19], it should be noted that the corresponding entry qualifications should be developed before potential palliative nurses are recruited. Third, palliative care policy is relatively limited in China, and the current policy stipulates palliative standards, including structure, human resources, and service environment [14]. However, to address the shortage of palliative nurses through policy solutions and developing localized palliative care in China, policy makers and nurse managers should take particular care to optimize the current policy in light of relevant cultural characteristics and build favourable teamwork, a harmonious atmosphere, and a reasonable salary system in the Chinese context [19, 20].

First, due to the nature of the cross-sectional design, we cannot identify causal relationships among burnout and its influencing factors. Future longitudinal studies are therefore warranted to robustly analyse the correlative relationships among these factors. Second, although we included nearly all palliative nurses from Sichuan Province, the representativeness of the results might be affected due to the lack of samples from other areas in China. As one year has passed since the data was collected, the timeliness of the results may be affected. Third, although information on the possible influencing factors were collected to the best of our knowledge, the results explained less than half of the total variance; thus, more exploratory studies should be carried out in the future.