Burnout is a chronic response to emotional and interpersonal stressors on the job and is defined by the three dimensions of personal accomplishment, emotional exhaustion and depersonalisation (Maslach & Leiter, 2008). It has been established that burnout is highly prevalent among nurses (Cañadas-De la Fuente et al., 2015). A systematic review by Gómez-Urquiza et al. (2016) showed that nurses present high levels of emotional exhaustion and reduced personal accomplishment. In particular, oncology nurses experience moral distress and burnout because of the intensity and proximity to pain, suffering, and death. In addition, these nurses experiencing burnout often have higher levels of emotional exhaustion (Cheng et al., 2015; Kutluturkan et al., 2016).
Consequences of burnout, such as chronic fatigue and impaired health, have serious implications on both patients’ and nurses’ safety. For example, high emotional exhaustion and depersonalisation are significantly associated with low patient safety grades (Halbesleben et al., 2008) and anxiety/insomnia, social dysfunction and severe depressive symptoms (Khamisa et al., 2015). Burnout could also lead to a shortage of nurses specialising in cancer care since it increases the intention to leave the profession (Lin, 2012).
A previous study has demonstrated promising evidence that burnout could be ameliorated by improving the psychosocial work environment, such as rewards received, reduced work-family conflict, and increased leadership quality (Rahman et al., 2017). However, due to the multi-dimensional factors in the psychosocial work environment, which consists of 28 factors (Kiss et al., 2013). This complex relationship warrants further research to improve understanding and design better interventions. In Brunei, no study has examined burnout among oncology nurses, and this research is timely to shed light on these nurses’ levels of burnout. Brunei adopts universal healthcare services where the citizens enjoy free health services. With a large influx of patients at the only cancer centre, it becomes necessary to measure the burnout of nurses, particularly oncology nurses, where evidence is non-existent. Therefore, this study aimed to investigate the relationship between burnout and psychosocial factors in oncology nurses. The secondary aim was to estimate the prevalence of burnout among oncology nurses in the main cancer care centre in Brunei.
Study Design and Participants
This was a cross-sectional study using self-administered questionnaires on all oncology nurses working in the main cancer care centre.
The questionnaire consisted of three sections: First part is the demographic data which collects participants’ age, gender, and nationality, duration of work experience, highest qualification, designation, and working hours. The second part gathers data on burnout using the Maslach Burnout Inventory (MBI), and the third part collects data on psychosocial factors using The Copenhagen Psychological Questionnaire (COPSOQ II).
The English version of the Maslach Burnout Inventory (MBI) questionnaire was developed by Maslach and Jackson (1981). The MBI questionnaire was obtained from the Mind Garden website, and permission to use it was granted upon purchasing the License to Produce. It addressed three subscales, namely 1) emotional exhaustion, which measures feelings of being emotionally overextended and exhausted due to work, 2) depersonalisation which measures the unfeeling and impersonal response towards one’s service and care treatment; and 3) personal accomplishment, which measures one’s successful achievement and competency at work. MBI is a 22 item questionnaire established to measure burnout of people whose job description involved human services (Maslach et al., 2001).
The English version of the Copenhagen Psychological Questionnaire (COPSOQ II) was developed by the National Research Centre for the Working Environment, Denmark. COPSOQ II was obtained from National Research Centre for the Working Environment website and can be used freely for research purposes. The medium version of COPSOQ II was used for this research. It has 87 items measuring 28 psychosocial factors, including quantitative demands, work pace, emotional demands, influence, possibilities of development, the meaning of work, commitment to the workplace, predictability, rewards (recognition), role clarity, role conflicts, quality of leadership, social support from supervisor, social support from colleagues, the social community at work, job satisfaction, work-family conflict, trust regarding management, mutual trust between employees, justice and respect, self-rated health, burnout, stress and sleeping troubles. In addition, offensive behaviours are also to be assessed, including sexual harassment, threats of violence, physical violence and bullying (Kiss et al., 2013).
Data were collected from February to March 2018. All oncology nurses at the only cancer centre in Brunei were recruited for this study. Seventy-one sets of self-administered questionnaires were distributed by the researchers.
A validation procedure was conducted to re-establish validity and reliability estimates of MBI. Inter-scale correlation (correlation matrix), corrected item-total correlation (CITC), average variance extracted, and Cronbach’s alpha were computed to establish discriminant validity, convergent validity, composite reliability and internal consistency reliability, respectively. In addition, Cronbach’s alpha coefficient and CITC were also established for COPSOQ II.
Descriptive statistics were calculated for the characteristics of participants. Missing data were replaced with centred mean due to small (less than 5%) missing values. The scores of MBI was reported using proportion based on the categorisation of high, moderate or low. The scores of COPSOQ II were reported using mean and standard deviation for numerical scales and frequency and percentage for categorical scales. Multivariate regression (Structural Equation Modelling) using maximum likelihood procedures (bootstrapping with 1000 subsample and 95% confidence interval) was performed to determine the relationship between psychosocial factors and burnout variables. Statistical analysis was computed with IBM|SPSS v21 and IBM|AMOS v.25. P-value less than 0.05 is considered evidence of statistical effect (two-tailed).
The study was approved by the Institutional ethics review board of the Universiti Brunei Darussalam and the Ministry of Health. The participants were given a week to read the participant information sheet. Those who agreed have signed the written informed consent prior to completing the survey and returned it to the researchers. Data were taken anonymously to protect the participants’ identities.
A total of 63 questionnaires (89% response rate) were used as valid data points for analysis. Table 1 shows the sociodemographic data of participants. Participants’ age ranges from 20 to more than 50 years old. The majority of the participants are female (77.8%), married (55.6%), and local (57.1%). The highest respondents are from the speciality nursing department (33.3%), and most are staff nurses (96.8%). The work experience ranges from 0 to more than ten years. Only 29% of the respondents have oncology nursing qualifications.
|20 – 29||19||(30.2)|
|30 – 39||37||(58.7)|
|40 – 49||6||(9.5)|
|More than 50||1||(1.6)|
|Day care unit||10||(15.9)|
|0 – 5||21||(33.3)|
|6 – 10||15||(23.8)|
|More than 10||17||(42.9)|
|Highest education level|
|Oncology nursing qualification|
|No (General nursing)||45||(71.4)|
n = frequency
Table 2 demonstrates the validity and reliability estimates for MBI. In terms of validity, the instrument was modified corresponding to the changes suggested by participants to improve face and content validity. CITC for all numerical scales range from 0.5 to 0.8, indicating satisfactory to good convergent validity except for Item 14 (CITC=0.148) of emotional exhaustion subscale and Item 18 (CITC=0.119) of personal accomplishment subscale, which showed low correlation with overall items. The correlation matrix showed that personal accomplishment was negatively associated with emotional exhaustion and depersonalisation, which was intended by the scale. In terms of reliability, all the subscales had acceptable to good estimates. The average variance extracted estimates were above satisfactory (above 0.5). Cronbach’s alpha coefficient was 0.70 and above, indicating good internal consistency reliability. The cumulative variance explained by these factors for burnout was 70.5%.
1 = Personal accomplishment (8 Items)
2 = Emotional Exhaustion (9 Items)
3 = Depersonalization (5 Items)
AVE = Average Variance Extracted (cut-off 0.5)
Alpha = Cronbach’s alpha (cut-off 0.6)
Table 3 shows the internal consistency reliability coefficients for COPSOQ II. CITC for all numerical scales ranges from 0.4 to 0.6, indicating acceptable convergent validity, except for items in emotional demand, influence at work, skill discretion, role conflict and trust (horizontal), which coincided with low Cronbach’s alpha indicating poor internal consistency reliability amongst those subscales. On the other hand, other subscales demonstrated good to excellent internal Cronbach’s alpha between 0.7 and 0.9.
|1.||Quantitative demand (4 Items)||0.683|
|2.||Work pace (3 Items)||0.560|
|3.||Emotional demand (4 Items)||0.349|
|4.||Influence at work (3 Items)||0.378|
|5.||Skill discretion (4 Items)||0.559|
|6.||Meaning of work (3 Items)||0.634|
|7.||Commitment to the workplace (4 Items)||0.715|
|8.||Predictability (2 Items)||0.820|
|9.||Rewards (3 Items)||0.866|
|10.||Role Clarity (3 Items)||0.756|
|11.||Role Conflict (3 Items)||0.398|
|12.||Quality of Leadership (4 Items)||0.942|
|13.||Social support (colleague) (3 Items)||0.616|
|14.||Social support (superiors) (3 Items)||0.910|
|15.||Social community at work (3 Items)||0.678|
|16.||Job satisfaction (4 Items)||0.846|
|17.||Work-family conflict (4 Items)||0.731|
|18.||Trust (horizontal) (3 Items)||0.567|
|19.||Trust (vertical) (4 Items)||0.697|
|20.||Justice and respect (4 Items)||0.829|
|21.||Sleeping troubles (4 Items)||0.869|
|22.||Burnout (4 Items)||0.919|
|23.||Stress (4 Items)||0.856|
Table 4 demonstrated that, in terms of prevalence, we estimated using a 95% confidence interval that personal accomplishment was low to moderate (20% to 43%), emotional exhaustion was high (55% to 79%), and depersonalisation was low (3% to 19%) among the population of oncology nurses in Brunei.
n = frequency
Table 5 demonstrated that the participants reported a high level of stress, quantitative demands and sleeping troubles. On the other hand, 44.5% of the participants reported excellent to a very good level of health compared to fair (44.4%) and poor health (11.1%).
|4.||Influence at work||3.2||(0.59)|
|6.||Meaning of work||1.8||(0.56)|
|7.||Commitment to workplace||2.6||(0.84)|
|12.||Quality of Leadership||2.4||(0.82)|
|13.||Social support (colleague)||2.3||(0.78)|
|14.||Social support (superiors)||2.4||(1.06)|
|15.||Social community at work||1.9||(0.68)|
|20.||Justice and respect||3.1||(0.86)|
|25.||Sexual harassment (Yes)||5||(7.9)|
|26.||Threats of violence (Yes)||10||(15.9)|
|27.||Physical violence (Yes)||3||(4.8)|
SD = Standard deviation, n = frequency, Scoring: lowest = 0, highest = 5
Figure 1 illustrates the overall relationship between psychosocial factors and burnout variables. It was observed that there was a highly significant relationship between psychosocial factors and emotional exhaustion. Table 6 shows that emotional exhaustion had a high positive significant relationship towards the quality of leadership (β = 0.876, p <0.001), justice and respect (β = 0.755, p <0.001) and rewards (β = 0.719, p <0.001). Whilst having a moderate negative significant relationship with sleeping troubles (β = -0.583, p <0.001), stress (β = -0.573, p <0.001), and work-family conflict (β = -0.364, p = 0.006). However, the overall relationship between psychosocial factors and personal accomplishment and depersonali-sation was not significant.
|Personal Accomplishment||Emotional Exhaustion||Depersonalization|
|Influence at work||-0.047||0.717||-0.054||0.728||-0.196||0.097|
|Meaning of work||-0.420||<0.001||0.609||<0.001||-0.137||0.378|
|Commitment to workplace||-0.290||0.004||0.570||<0.001||-0.119||0.379|
|Quality of leadership||-0.026||0.775||0.876||<0.001||-0.499||<0.001|
|Social support (colleague)||-0.136||0.273||0.240||0.105||0.071||0.638|
|Social support (superiors)||-0.062||0.587||0.517||<0.001||-0.093||0.505|
|Social community at work||-0.094||0.417||0.373||0.007||-0.047||0.672|
|Justice and respect||0.002||0.986||0.755||<0.001||-0.053||0.737|
Estimates = Standardised regression coefficient, bold = significance at 0.05 level
This study investigated the prevalence of burnout and its relationship with psychosocial factors among oncology nurses using MBI and COPSOQ II. Both measuring tools demonstrated acceptable to good validation estimates except for items in emotional demand, influence at work, skill discretion, role conflict and trust (horizontal) of COPSOQ II. This is congruent to previous studies (Nübling et al., 2006; Ibtissam et al., 2012).
The main finding from this study suggested improving the psychosocial work environment could significantly lower emotional exhaustion. In this study, quality of leadership (extent by which immediate superior is considerate of staff satisfaction and good development opportunities), justice and respect (fair treatment at work) and rewards (recognition and appreciation by management) were significantly related to emotional exhaustion. Therefore, management and policymakers could consider these specific factors in future policies while also accounting for characteristics of high quantitative demand, stress and sleeping troubles among these oncology nurses. These results were consistent with a previous study demonstrating that oncology nurses experience more emotional exhaustion (Kutluturkan et al., 2016). Another study by Cheng et al. (2015) showed that oncology nurses suffered a high prevalence of emotional exhaustion (45% from 358 nurses) and low personal accomplishment (66.7% from 358 nurses). The main implication is to reduce nurses’ intention to leave and retain them to tackle a nurses’ shortage as a persistent issue for decades (Brossoit et al., 2020).
Burnout among nurses compromises nurses’ and patients’ safety (Halbesleben et al., 2008; Pogoy & Cutamora, 2021). Despite the identification of this specific relationship, burnout management should not neglect. On the contrary, it should be complemented by other interventions, which could consist of identifying and recognising the signs and symptoms of burnout such as chronic fatigue, insomnia, forgetfulness, anxiety and depression, isolation and detachment, and lack of productivity and poor performance (Gómez-Urquiza et al., 2016). The sooner the nurses identified the signs, the sooner they can get appropriate interventions. There are also various stress and burnout management approaches, such as counselling, support groups, and relaxation methods (Braunschneider, 2013; Berg et al., 2016).
There should be acknowledging and accepting the expected reality of compassion fatigue and burnout, education on how to recognise symptoms of compassion fatigue and burnout, portray professional coping skills as a team and promote social support and positive relationships, and engage healthcare team in discussions about coping and make it part of regularly scheduled meetings. The management could also develop interventions to reduce burnout among oncology nurses, such as communication and team-building training, feedback techniques, and goal-setting (Maslach & Leiter, 2008).
In terms of study limitations, the results of this study should be interpreted with caution because several COPSOQ II factors had low reliability, such as emotional demand, influence at work and role conflict. Small sample size may impair external validity and hence generalisability as there may be cross-cultural or other demographic differences. Also, a small sample could impair model fitness for the multivariate regression analysis. However, this study aimed not to model the relationship but instead examine how psychosocial factors interact in terms of burnout variables and see the practical significance of this relationship, especially in a small country where a larger sample size requires collaboration with external nations. The cross-sectional and quantitative nature of the study might result in not much depth and limited implications.
This study showed that improvement in quality of leadership, justice and respect and rewards could minimise emotional exhaustion among oncology nurses. Therefore, management and policymakers could target these specific factors in addition to using other interventions to counter the harmful effects of burnout.