A Cross-Sectional Study on Key Factors of Burnout Among Tunisian Oncology Nurses

2.1 Study population

This descriptive cross-sectional study was conducted over one month, from April 15 to May 15, 2018, at the Salah Azaiez Institute (ISA) in four departments: cancer surgery, otorhinolaryngology, medical oncology, and radiotherapy. Eligible nurses had at least one year of oncology experience and sufficient French proficiency to complete the questionnaire. The exclusion criteria were nurses in training, those with administrative roles or without direct patient contact, those with a first-degree relative with cancer, and those suffering from a severe, disabling somatic condition (including cancer). Nurses with incomplete questionnaire responses or thise who withdrew their consent were also excluded.

2.2 Evaluation

Data were collected using an anonymous form that gathered socio-demographic and occupational information Table I (Extended data).⁸ Burnout was assessed using the Maslach Burnout Inventory (MBI), a tool developed by Maslach and Jackson in 1981.⁹ For this study, a copyright license was obtained to authorize the use of this proprietary instrument. The MBI evaluates three key dimensions of burnout: Emotional Exhaustion (EE), Depersonalization (DP), and Loss of Personal Achievement (PA). Burnout diagnosis was based on MBI scores exceeding Maslach’s defined thresholds in any of the three components.¹⁰ We categorized burnout into three levels: no, full, and intermediate.⁸

2.3 Ethical considerations

This study was conducted in accordance with the ethical standards outlined in the Declaration of Helsinki. Ethical approval was obtained from the Institutional Review Board (IRB) of the Salah Azaiez Institute, with the approval reference number ISA/2024/01. The committee approved the study on 30 January 2024, ensuring strict adherence to ethical guidelines and the protection of participants’ rights and confidentiality throughout the research process

2.4 Statistical analysis

Data analysis was performed using SPSS software (version 23.0, IBM Corp). Continuous variables were tested for normal distribution using the Shapiro-Wilks test. Owing to significant deviations from the normal distribution, medians, interquartile ranges, and extreme values were calculated for quantitative variables. Group differences for continuous variables were analyzed using Mann-Whitney or Kruskal-Wallis tests, and percentages were compared using Fisher’s exact test. The Spearman correlation coefficient was used to examine associations between quantitative variables, considering p < 0.05 as indicative of correlation. Multiple linear regression identified independent predictors of EE, DP, and PA, with the significance level set at 0.05. Please refer to the relevant underlying data.⁸

Assessment of burnout among nurses included in our study

Our study identified a substantial burnout rate (89.9%) among nurses. In the literature, the comparison between different burnout levels in oncology nurses in published series remains difficult. Indeed, so far there is no clear and precise consensus on two essential points: the precise definition of burnout as a syndrome, and the interpretation of the obtained scores in each of the three dimensions. Moreover, the majority of studies on burnout in oncology comprised all caregivers (doctors, nurses and assistant nurses); the published results were heterogeneous.

In the studies that defined burnout as the elevation of one of the three dimensions of the MBI, our results were quite comparable to the literature data.

A French study conducted in 2008 among caregivers in oncology showed that 39% of the studied population was affected by burnout with predominance among doctors (38%), followed by assistant nurses (35.2%) and nurses (25%) 18.¹¹

A study conducted in 2010 in Italy compared the burnout levels among Oncology nurses (59 nurses working in the oncology units at the hospital and 33 at hospices, and found a higher level of burnout among nurses working at the Hospital.¹²

For other authors, burnout was defined by a high level of EE.

In a study published in 2005, American authors studied burnout among 305 nurses in Oncology, and found a high level of EE in 39% of cases.¹³

A multicenter Chinese study published in 2013 including 708 nurses working in Oncology showed that 36.8% of the study population had high levels of burnout.¹⁴

In other works, the burnout was retained in the presence whether high levels of EE and DP, or a low level of PA.

In a Swedish multicenter study conducted in 2015 including 7412 nurses, the authors compared nurses according to their assignment services: a first group of nurses working in services and treating cancer patients (80% of treated patients, n = 1440) and a second group with a lower number of cancer patients (10-70% of treated patients, n = 5972).¹⁵ The authors identified a burnout rate of 22.8% in the first group versus 25% in the second one.

Other authors retained a burnout syndrome in the presence of an alteration of the three dimensions of the MBI scale (high levels of EE and DP, and a low level of PA).

After applying this definition, an Iranian study published in 2017 on 67 nurses working in oncology found a burnout rate of 38.8%.¹⁶

The results of a meta-analysis published in 2018, including 17 studies of which 82.3% were cross-sectional and using the MBI scale to assess the levels of burnout among oncology nurses¹⁷ found high scores of EE and DP in 30% and 15% of the cases, respectively, and low PA scores in 35% of cases.

The concept of burnout in Arab Muslim countries has not been well explored. A Tunisian study published in 2014, that assessed 60 nurses caring for end-of-life patients, revealed a burnout prevalence of 70%. Notably, 81.7% of participants exhibited high levels of burnout. Among them, 80% experienced a high level of emotional exhaustion, 70% reported a high level of depersonalization, and 17% demonstrated a low sense of personal accomplishment.¹⁸

Other studies conducted in some countries such as Algeria and Saudi Arabia showed that nurses, in these countries, suffered from Burnout.^19,20 A Jordanian study conducted among 181 psychiatric nurses found high levels of EE and DP in 32.7% and 27.7% of cases, respectively, and low levels of AP in 16.8% of cases).²¹ In a systematic review assessing burnout among healthcare professionals in Arabic countries, the authors recorded high EE (81.0%), high DP (80.0%), and low PA (85.8%).²²

Association between burnout and the different variables of the study

In our study, we found an association between high EE levels and female sex.

In the literature, the role of sex in the occurrence of burnout remains unclear. According to a meta-analysis published in 2010 that analyzed 183 studies, women were more emotionally exhausted than men, while men had higher levels of depersonalization than women.²³

The results of a Norwegian study published in 2011 also suggested that female nurses exerienced more burnout than male nurses (especially higher levels of EE).²⁴ An American study of all caregivers in oncology showed that men had a greater sense of PA than women.²⁵

A greater prevalence of burnout among women was explained by greater involvement in the emotional relationship with their patients as well as more difficulties in reconciling their professional and family lives.²⁶ Therefore, Burnout tend to be more common among women because of more important life roles (housework and childcare) and work–family conflict.

We found that younger age was associated with lower AP levels. According to Maslach and Goldberg,²⁷ the older one gets, the more adaptive mechanisms against stress are improved. For Ullrich and FitzGerald,²⁸ young oncology health professionals would have more difficulties communicating with their patients, would be more stressed at work, and would express more criticism at work.

Some authors have explained this fact by three points: (1) experience and capacity of adaptation (coping strategies) improves with age; (2) distancing from many factors of stress occurs with age, and (3) more dissatisfied nurses would leave at a young age.²⁹

In our study, we did not find statistically significant associations between burnout and other socio-demographic factors (marital status, number of children, habitat type and socio-economic level).

These results match the majority of studies published on the burnout in the literature.^30,31

However, some authors have suggested that having a partner and children is a protective factor against burnout.³¹ Maslach and Jackson³² found that individuals without children are more likely to experience burnout. Another study showed that having children is correlated with the occurrence of EE.³³

In our study, a high DP level was significantly associated with the presence of a medical history. In a publication dealing with the assessment of burnout among 67 nurses in oncology, Quattrin et al. found a statistically significant association between personal medical history and the severity of burnout.³³ The study showed that EE was correlated with the presence of a medical surgical personal history and the experience of having a family member affected by cancer.

This study showed a statistically significant association between the level of high EE and the presence of a personal psychiatric history, taking psychotropic drugs in the last year and asking for psychological help because of work exhaustion.

This association has been previously described in the literature. Catt and al.³⁴ found a psychiatric comorbidity associated with burnout in 5-27% of cases in 10 carcinology teams in England. For example, n Italy, Bressi et al. reported a 28.8% prevalence of psychiatric disorders among oncology nurses.³⁵ The authors also found a statistically significant association between psychiatric morbidity and the three dimensions of burnout. Finally, in a literature review burnout among caregivers in oncology was found to be associated with somatic, psychological and behavioral disorders.³⁶

In our population, lack of leisure activities (sports, music, etc.) was associated with EE. Leisure and social extra-professional activities would play an important role in easing and tension stress. Quattrin and al. showed that the different strategies adopted by oncology nurses to manage stress (physical activity, medical examinations, no conventional strategies to manage stress, healthy eating behaviors, rest, healthy lifestyle and hobbies) were associated with a reduction in EE levels.³³

In addition, the fact that not knowing or not being able to take time to relax has been mentioned in several publications as being one of the first disturbance in burnout.³⁶ In our study, tobacco and/or alcohol consumption did not correlated with the severity of the burnout.

Kash et al.²⁵ found that the consumption of cigarettes, alcohol or drugs to relax is a predictor of EE and DP. In our study, the number of years worked in oncology was negatively correlated with a high level of DP.

In the literature, there is conflicting data on the relationship between burnout and professional experience.

For some authors, the more old a caregiver is in a team, the less they are at risk of burnout. Inexperienced caregivers would be made fragile and disappointed with the reality on the ground which is opposed to the ideals of the profession at the beginning of their career, while more experienced caregivers stood and went past this stage. On the other hand, prolonged exposure to suffering and difficult patients, and the accumulation of a significant workload without compensatory recognition could lead these experienced caregivers to exhaustion in the middle or at the end of their careers.

In our study, 55.2% of the nurses were somewhat unsatisfied or unsatisfied at work. A statistically significant association was found between high EE, low PA and dissatisfaction at work. In the multivariate analysis, lack of job satisfaction was found to be an independent predictor of increased EE and decreased PA scores.

Some authors stated that poor job satisfaction would represent the roots of burnout; the causes would be chronic devaluation and insufficient recognition from the hierarchy, reflected by weak PA scores.³⁷

Shang et al.¹⁴ found that lack of satisfaction was predictive of psychiatric illness and burnout. Gulalp and al.³⁸ have shown that lack of job satisfaction is associated with the three dimensions of burnout.

Bressi et al. found that 10% of oncology nurses were not satisfied at work; this dissatisfaction was correlated with low levels of PA.³⁵

A Turkish study published in 2017 used the “Minnesota Job Satisfaction score” to determine the level of satisfaction of 129 nurses working in oncology, and found a negative and significant correlation between the burnout level measured by the MBI and the satisfaction score.³⁹

The burnout of nurses was associated in the literature with psychiatric symptomatology manifesting itself through a loss of sleep, a feeling of stress, depression and dissatisfaction with daily activities.¹¹ In addition, the burnout of caregivers lowers the quality of care, and influences the working conditions of health care providers.⁴⁰ Burnout reduces satisfaction with work, involvement in the organization and performance. This increases the desire to change jobs as well as absenteeism. Even in his position, a caregiver suffering from burnout is less effective and therefore less productive in economic terms.⁴¹

Study limitations

While his study, provide valuable insights into burnout among Tunisian oncology nurses, it has certain limitations. Firstly, the relatively small sample size necessitates caution when generalizing the results. Additionally, the study’s cross-sectional design limited our ability to draw conclusions about longitudinal changes or establish causal relationships between the variables studied and the level of nurse burnout. Future studies with larger samples across various oncology services in Tunisia are recommended to provide more comprehensive insights.

Clinical implications

The findings of this study indicate a high prevalence of burnout among oncology nurses, which has significant clinical implications. The association between burnout and factors such as job satisfaction and personal psychiatric history highlights the need for targeted interventions. These could include mental health support, improved working conditions, and strategies for enhancing job satisfaction. Addressing these factors is essential for maintaining nurse’s well-being and ensuring the delivery of quality patient care in oncology settings.