Prevalence and factors associated with professional burnout in Polish oncologists—results of a nationwide survey

Abstract

Background

High rates of burnout are observed among health care professionals worldwide, which could have negative consequences on personal and organizational levels. We aimed to evaluate the burnout prevalence and factors associated with burnout among oncologists in Poland.

Materials and methods

An online survey was conducted using the validated Maslach Burnout Inventory—Human Services Survey (MBI-HSS) and additional work/lifestyle questions. Descriptive statistics, parametric and nonparametric tests, and multivariate logistic regression were used to identify factors associated with burnout.

Results

A total of 228 physicians participated in the survey, including 168 medical oncologists, 43 radiation oncologists, and 17 from other specialties. Data collected from 211 medical and radiation oncologists were included in the final analyses. Most participants were female (71.6%) and ≤40 years of age (70.1%). A self-reported feeling of burnout was present in 65.9% of participants. Based on the MBI-HSS, 74.9% showed evidence of burnout with burnout subdomains as follows: depersonalization 37.0%; emotional exhaustion 64.5%; low accomplishment 43.1%. There were no differences in burnout rates based on specialization (oncology/haematology—75.6%, radiotherapy—72.1%), career stage, gender, or age groups. Lack of work–life balance was the only significant factor associated with the risk of burnout in the logistic regression (relative risk 2.6, 95% confidence interval 1.3-5.4). Only 20.9% of physicians had access to psychological support in their workplace; however, 70.1% desired such support. Three main factors impacting burnout in cancer care workers were: bureaucracy and administrative duties overload, admissions of many patients, and poor work culture.

Conclusions

Burnout is common among medical and radiation oncologists in Poland. There is a high demand for psychological support and organizational changes in the workplace to reduce risk and mitigate the adverse effects of burnout among health care professionals.

Highlights

• •Professional burnout is prevalent in Polish oncologists affecting three-quarters of them.
• •Work overload, bureaucracy, and poor work organization are the main factors leading to burnout.
• •A majority of oncologists in Poland do not have access to psychological support.

Introduction

Burnout syndrome is a complex occupational phenomenon resulting from chronic workplace stress that has not been effectively managed. It is characterized by emotional and physical exhaustion (e.g. chronic fatigue, behavioural distress), depersonalization (DEP) (e.g. depression, demoralization), and low professional satisfaction (e.g. low productivity).^1,2 Many studies have confirmed the high prevalence of burnout syndrome among health care workers worldwide.3, 4, 5

The major risk factors for burnout syndrome consist of individual, organizational, and professional factors. Individual risk factors include inherent personality traits and sociodemographic predispositions such as female gender, age ≤55 years, short professional experience, and marital status (single, unmarried physician).^6,7 Personality traits are considered an independent risk factor for occupational burnout and include alexithymia, type A behaviour, mental resilience, conscientiousness, neuroticism, and compulsiveness.⁸ Organizational and professional factors include the increasing time of patient care, high professional requirements, increased administrative duties, extensive medical records, unclear job expectations, rapidly changing medical knowledge, and lack of social support.^9,10 Oncology is a particular field of medicine that especially predisposes to burnout syndrome because of life-and-death decisions, frequent delivery of distressing news, dealing with terminally ill patients, disease relapses, and patient death. Oncologists often deal with heavy workloads. Burnout among oncologists is often associated with decreased patient treatment quality and loss of empathy and compassion and may contribute to medical malpractice. Burnout syndrome can lead to depression and even more severe consequences, including drug addiction, suicide, and medical errors.^1,11

Results of burnout studies in oncology show that the incidence of burnout varies between 20% and 71%.¹² The highest incidence rates of burnout syndrome were reported in the European Society of Medical Oncology (ESMO) study, where burnout syndrome occurred in 71% of young oncologists. The incidence of burnout varies between regions, study populations, and burnout measures. It significantly increased during the coronavirus disease 2019 (COVID-19) pandemic.^13,14 Similar results were found in an international study among over 600 oncology physicians from 19 European countries, including Poland, where a high risk of occupational burnout was observed in 72% of physicians.¹⁵ In a study conducted by the American Society of Clinical Oncology (ASCO), 45% of oncologists exhibited symptoms of burnout in the form of emotional exhaustion (EE) and DEP.¹ Due to the high prevalence of burnout among oncologists in Central and Eastern Europe,¹⁵ and the lack of reliable Polish data, we conducted the first comprehensive study to assess burnout and its causes among Polish medical professionals working in oncology.

Materials and methods

Study design and questionnaire

The study was conducted between 15 February and 15 April 2022, using an online questionnaire. Information about the study was disseminated through the Polish Society of Clinical Oncology (PTOK) website, the PTOK Young Oncologist Facebook group, the PTOK newsletter, and targeted emails to department chairs, oncologists, and other physicians working with cancer patients. Participation was voluntary, and no incentives were offered to encourage participation. The questionnaire consisted of a standardized burnout assessment using the Maslach Burnout Inventory—Human Services Survey (MBI-HSS) and 64 questions covering demographic data, lifestyle choices, professional activities, and factors potentially associated with the risk of burnout (Supplementary Material, available at https://doi.org/10.1016/j.esmoop.2023.102230). The selection of questions was based on the data from previous studies and agreed upon by the Young Oncologists Section of the PTOK. Due to the nature of the study (survey study), ethical approval by the local ethics committee was not required.

Respondents self-assessed their subjective feeling of burnout through a yes/no question in the questionnaire. Objective rates of burnout were measured using the validated MBI-HSS questionnaire, which consists of 22 items, each scored using 7-level frequency ratings from ‘never’ (0 points) to ‘every day’ (6 points). The MBI-HSS measures three dimensions of burnout: EE, DEP, and low personal accomplishment (PA). The score ≥27 points on the EE subscale defined a high level of EE, ≥10 on the DEP subscale a high level of DEP, and <32 on the PA subscale a low level of personal and professional accomplishment. Each of these results was considered a surrogate of a high level of burnout in that subscale. The overall burnout rate (definition 1) was defined as the percentage of study participants with a high level in the EE or DEP subscale or a low level in the DEP subscale. Due to a lack of methodological consistency, additional, more strict definitions of burnout were applied. For exploratory analyses, burnout was defined as a high level in both the EE and DEP scales (definition 2), a high level in either the EE or DEP scale (definition 3), or a high level in the EE and DEP scales and a low level of PA (definition 4).^2,16

The study was dedicated to all physicians involved in the care of cancer patients, mainly clinical/medical oncologists, radiation oncologists, haematologists, surgical oncologists, and gynaecological oncologists. Due to a low number of responses from certain groups, only responses from clinical/medical oncologists, radiation oncologists, and haematologists were analysed. Due to the similar scope of work, clinical/medical oncologists and haematologists were grouped together. A threshold of 40 years of age was used to distinguish the young oncologist population based on the ESMO definition of a young oncologist.

Statistical analyses

Descriptive statistics were used to summarize the data. For comparison of continuous variables, nonparametric Mann–Whitney or Kruskal–Wallis tests were used. Categorical data were compared with chi-square or Fisher’s exact test. The association between selected demographic, lifestyle, and work-associated factors and burnout (binary variable, yes versus no) was analysed using a multivariate logistic regression. Results were considered statistically significant when two-tailed P values were <0.05. Statistical analyses were carried out using PS Imago PRO 7.0 (Predictive Solutions, New York, NY). Figures were drawn in the Flourish.studio (Canva UK Operations Ltd., London, UK) or MS Excel (Microsoft, Redmond, WA).

Results

Study population

A total of 228 physicians filled out the survey, including 168 medical oncologists or haematologists, 43 radiation oncologists, and 17 from other specialties. Due to the low response rates in the other specialties, the analyses were conducted using data obtained exclusively from medical oncologists, haematologists, and radiation oncologists.

Overall, 211 responders were included (Table 1). Women comprised the majority of the study population (71.5%, n = 151). Most physicians were board-certified specialists in their discipline (69.2%, n = 146) with a median experience in oncology of 10 years (range 1-45 years). Participants represented equally cancer centres and multidisciplinary hospitals (48.8%, n = 103 and 51.2%, n = 108, respectively) and worked mainly in large cities with >250 000 inhabitants (67.8%, n = 143).

Table 1.

Characteristics of the study population

Factor		% (n) n = 211
Gender	Male	28.0 (59)
	Female	71.5 (151)
	I do not want to specify	0.5 (1)
Age	Median (range)	37.0 (26-70) years
	≤40 years old	70.1 (148)
Marital status	Single (single, widower, divorcee)	19.9 (42)
	In a relationship	80.1 (169)
Having children		64.9 (137)
Number of children	Median (range)	2 (1-5)
Oncology specialty	Clinical oncology and haematology	79.6 (168)
	Radiation therapy	20.4 (43)
Stage of training	Board-certified specialist	69.2 (146)
	Doctor in training	30.8 (65)
Years working in oncology (including training)	Median (range)	10 (1-45)
Type of medical facility of workplace and practice	Cancer centre	48.8 (103)
	Multidisciplinary hospital	51.2 (108)
Size of the city where working (number of inhabitants)	<50 000	4.3 (9)
	50 000-100 000	7.6 (16)
	100 000-250 000	20.4 (43)
	>250 000	67.8 (143)

The vast majority of participating physicians were in a relationship (80.1%, n = 169) and had children (64.9%, n = 137) (Table 1). Regarding lifestyle factors, nearly one-quarter did not carry out any physical activity (23.2%, n = 49), over half slept ≤6 h a day (53.1%, n = 112), and 55.0% (n = 116) reported lack of effective sleep that gives a feeling of relaxation (Supplementary Table S1, available at https://doi.org/10.1016/j.esmoop.2023.102230). Antidepressant or anti-anxiety medications are used by 16.6% (n = 35) of responders, 12.3% (n = 26) smoke cigarettes or used other nicotine-delivery systems, and 24.7% (n = 53) drank alcohol two or more times a week.

Half of the responders worked between 8 and 10 h a day (50.7%, n = 107), while 22.8% (n = 48) worked over 10 h (Supplementary Table S1, available at https://doi.org/10.1016/j.esmoop.2023.102230). Nearly half of the participants had additional research and teaching activities besides clinical work. A very low proportion of physicians reported possibility of having a break time during the workday or to eat regularly at work, 32.2% (n = 68) and 28.4% (n = 60), respectively. The vast majority complained of not having a work–life balance (61.6%, n = 130) and not using all credited vacation days per year (56.4%, n = 119). A detailed description of lifestyle- and work-associated factors in the study population is provided in Supplementary Table S1, available at https://doi.org/10.1016/j.esmoop.2023.102230.

Self-reported feeling of burnout

A self-reported feeling of burnout was declared by 65.9% (n = 139) of participants, with no significant differences between medical oncologists/haematologists and radiation oncologists (P = 0.079), 67.9% and 58.1%, respectively (Figure 1A). Main signs of burnout reported in the study included feeling of excessive workload (73.1%, n = 154), lack of energy, constant fatigue, and exhaustion (72.0%, n = 152), and inability to recharge (52.1%, n = 110) (Figure 1B). Only 4.3% (n = 9) of physicians did not report any of the signs included in the questionnaire. 62.6% of responders exhibited at least one of the adverse somatic signs possibly related to their work. The most frequently reported symptoms were tension headaches (31.3%, n = 66), pain, muscle tension, and stiffness (29.4%, n = 62), and increased heart rate (28.0%, n = 59) (Figure 1C).

Figure 1.

Prevalence of self-reported burnout among Polish oncologists. Self-reported feeling of burnout in the overall study population, medical oncologists/haematologists, and radiation oncologists (A); prevalence (in %) of self-reported symptoms (B) and signs (C) of burnout reported by questionnaire responders.

Level of burnout according to MBI-HSS

The median scores for EE, DEP, and feeling of a low level of personal and professional achievement on the MBI-HSS scale were 30.97, 11.21, and 32.99, respectively (Figure 2A). A high level of EE or DEP was present in 64.5% (n = 136) and 51.2% (n = 108), respectively, while a low level of personal and professional achievement in 43.1% (n = 91) of responders (Figure 2B). In the whole study population, 77.7% (n = 164) of physicians were classified as burned out based on the MBI-HSS score (high level of EE, high level of DEP, or low level of personal and professional achievements). Considering more strict criteria for burnout definition, e.g. a high level in both EE and DEP, or a high level in EE and DEP and a low level of PA, 40.8% (n = 86) and 24.2% (n = 51) of responders, respectively, would be classified as burned out (Supplementary Table S2, available at https://doi.org/10.1016/j.esmoop.2023.102230). Only 22.3% (n = 47) can be classified as engaged (having low levels of both EE and DEP).

Figure 2.

Objective burnout among Polish oncologists and selected subgroups. Scores for each domain of MBI-HSS questionnaire presented as median and quartiles (A) or classified into low, moderate, or high levels (B). Prevalence of professional burnout, high level of EE, high level of DEP, and low level of personal accomplishments based on the specialty (C), age (D), and gender (E). Multivariate logistic regression of factors potentially associated with a high risk of professional burnout (F). CI, confidence interval; DEP, depersonalization; EE, emotional exhaustion; MBI-HSS, Maslach Burnout Inventory—Human Services Survey; PA, personal achievements; RR, relative risk.

There were no significant differences in burnout rates between clinical oncologists/haematologists and radiation oncologists (79.2% versus 72.1%, P = 0.320), physicians ≤40 and >40 years old (78.4% versus 76.2%, P = 0.727), or females and males (78.8% versus 74.6%, P = 0.695) (Figure 2C-E). There were no differences between the abovementioned groups for alternative definitions of burnout or performance in each MBI-HSS subscale (Supplementary Tables S3 and S4, available at https://doi.org/10.1016/j.esmoop.2023.102230). Factors significantly associated with a higher risk of burnout were lack of relaxing sleep, lack of possibility to have break time at work, lack of possibility to eat regularly at work, and lack of work–life balance. In a multivariate logistic regression, only lack of work–life balance was significantly associated with the risk of burnout (relative risk 2.76, 95% confidence interval 1.31-5.82) (Figure 2F).

Factors leading to increased risk of burnout

To assess factors that lead to a high risk of burnout, all study participants were asked to score potential factors as having a high, moderate, or low impact on the risk of burnout. Over 80% of responders rated bureaucracy and administrative overload as highly related, followed by poor work organization, patient overload, and lack of time for patients, each rated as high by >60% of them (Figure 3A). When participants were asked to select three most important factors, bureaucracy and administrative overload were the most frequently chosen (63.0%, 133) (Figure 3B). Less than 10% of physicians pointed out that difficulties in communication with colleagues, night/on-call duties, and social conditions at the workplace are among the top factors leading to burnout.

Figure 3.

Factors associated with professional burnout. Impact of selected factors on the risk of professional burnout in subjective assessment of study participants (A). Main factors leading to professional burnout (study participants were asked to select three most important factors for them) (B).

Professional and psychological support

Over one-third of participating physicians (34.6%, n = 73) have used psychological support at their workplace or outside. Only 18.5% (n = 39) of study participants had access to psychological support in the workplace, 60.7% (n = 128) did not have such an opportunity, and 20.9% (n = 44) were not sure about it. The majority of participants, 71.1% (n = 150), would like to have access to psychological support, mostly in the form of individual consultations in the workplace (56.4%, n = 119) or group meetings (25.1%, n = 53). Nearly 25% of oncologists would like to participate in Balint group meetings in the workplace or outside of it, 23.7% (n = 5) and 24.2% (n = 51), respectively. At the time of the study, only 5.82% (n = 11) had participated in any Balint group meetings.

Thought about changing specialty or workplace

Nearly half of the oncologists (46.4%, n = 98) were thinking or had thought about changing their specialty, and 21.3% (n = 45) would not choose oncology as their main specialty again. The main reasons for considering changing specialty by oncologists were mental workload (68.4%, n = 67), too many responsibilities (62.2%, n = 61), and bad work organization (50.0%, n = 49) (Supplementary Table S5, available at https://doi.org/10.1016/j.esmoop.2023.102230). 70.1% (n = 148) of oncologists have been thinking about changing their current employer or workplace. The main reasons included bad work organization (72.3%, n = 107), too many responsibilities at work (63.5%, n = 94), and dissatisfaction with current salary (49.3%, n = 73).

Burnout as the reason for the change of specialty was reported by 37.8% (n = 37) and 41.2% (n = 61) of participants were considering changing their employer. Thoughts about changing their employer were more frequent among oncologists with a high risk of burnout in the MBI-HSS questionnaire compared to those who are not burned out (78.0% versus 24.6%, P < 0.001). A similar, but not statistically significant, correlation was observed regarding changing specialty (51.2% versus 29.8%, P = 0.019). Additionally, the trends were maintained when a more strict definition of burnout was employed (Supplementary Table S6, available at https://doi.org/10.1016/j.esmoop.2023.102230).

Discussion

Burnout among health care providers, especially those in cancer care, has become an increasingly recognized issue in recent years. Multiple studies have analysed the risk of burnout among oncologists, cancer nurses, or other specialists involved in providing care for patients with cancer. This study presents the findings of the first nationwide investigation into burnout among clinical and radiation oncologists in Poland. Our research sheds light on the issue of burnout, with over 75% of oncologists in our study demonstrating burnout or a high risk of it. This aligns closely with previous European and American studies, where burnout rates hovered around 71%-72%.^15,17,18 In the study conducted by the ESMO Young Oncologist Committee, the burnout rate in Eastern Europe was 73% which is similar to our findings, underlining the lack of changes between 2013 and 2022. Interestingly, in our study, we have observed a much higher proportion of physicians showing a high level of EE (64.5% versus 47.3%) and a slightly lower level of high DEP (51.2% versus 59.5%) compared to the ESMO study.¹⁸ Despite no differences between oncologists below or above 40 years of age, caution is needed when generalizing study findings to a broad population of oncology care professionals since the younger population was overrepresented in the study.

In our study, we have found that lack of work–life balance is significantly correlated with the risk of burnout. A high proportion of Polish oncologists work overtime, do not have regular break times at work, and are not able to use all the vacation days every year, which can lead to a dysfunctional work–life balance. Similarly, in Spanish young oncologists, poor perception of work–life balance and lack of leisure time or vacation time were independent prognostic factors for the risk of burnout.¹⁹ The majority of responders reported administrative workload as the most important factor leading to burnout. As previous studies have shown, >50% of gynaecological oncologists in the United States reported that excessive administrative work negatively impacts their work–life balance.²⁰

Along with other trials, other external factors such as daily workload and long working hours contribute significantly to burnout.²¹ Some previous studies have reported higher levels of burnout in females,^17,20 while others, including ours, have not.²¹ Among Spanish young oncologists, burnout level was higher in males.¹⁹ We did not observe any difference regarding age, while other studies showed higher burnout in younger oncologists.^17,22 Some studies reported a higher rate of burnout among medical oncologists and the lowest in radiation oncology,²³ while others did not document such a difference,²⁴ which is consistent with our findings. A small sample size for radiation oncologists could be a potential confounding factor. Relationship status or not having children was reported by other studies as potential risk factors for burnout; however, in our study, we were not able to document such a correlation.¹⁷

Consequences of oncology team members’ burnout can be classified into three categories: organizational, professional, and personal. Organizational consequences can lead to reduced patient safety, higher mortality rates, prolonged hospital stays, and a larger number of medical errors. From a professional perspective, burnout can result in the erosion of teamwork, suboptimal performance, poor team communication, high absenteeism among oncology team members, or desire to quit medicine or change specialty.²⁴ On a personal level, burnout can have a profound impact on the well-being of oncology professionals and lead to somatic and psychological disturbances. Moreover, burnout brings a large burden on the health care system. The estimated economic costs associated with physician burnout are substantial, averaging US$7600 per physician per year. These costs result from factors such as physician turnover and reduced working hours due to burnout.²⁵

It is important to acknowledge the limitations of our study. Personal characteristics or personality may affect the risk of burnout, i.e. neuroticism, agreeableness, and conscientiousness are personality traits that correlate with a higher risk of burnout.²⁶ In our study, we have not analysed personal traits, which should be considered as a study limitation. Additionally, potential biases and the relatively small sample size for radiation oncologists should be considered in interpreting our results. Finally, our study was carried out as an online survey, which can lead to a higher willingness to participate by people with higher burnout and lead to overrepresentation of young oncologists (≤40 years old), who accounted for 70% of the respondents. We are not able to estimate how many oncologists received the invitation to participate in the study, but considering statistics by the National Board of Physicians showing that there are around 1100 medical oncology specialists in Poland and around 300 oncologists in training, we estimate that participants of this survey constitute between 10% and 15% of oncologists in Poland.

Methodological differences in defining burnout should be considered when comparing between studies. In our research, we adopted the same definition as the ESMO study, where burnout was characterized by a high level of DEP or EE or a low level of PA. In contrast, studies by Shanafelt et al.¹⁷ and Blanchard et al.²⁴ focused solely on DEP or EE as criteria for burnout. Even with this consistent definition, our study still revealed a burnout rate exceeding 70%. Recently, a more strict definition was proposed that incorporates high levels of EE and DEP and a low level of PA. On the opposite end of the spectrum are physicians showing low levels of both EE and DEP, called ‘engaged’. With 24.5% of responders defined as burnout according to this definition, and only 22.3% classified as engaged, >50% can be classified in the grey zone (disengaged, overextended, or ineffective) that could be considered as a ‘pre-burnout’ state. This distribution closely resembles a recent study of young oncologists in Spain.¹⁹

Addressing burnout among oncologists demands a multi-pronged approach. Our study showed that at least three-quarters of oncologists in Poland may require tailored support to tackle their professional burnout. Previous study among neurooncologists in Europe and the United States showed that oncologist having institutional mechanism to support oncologists and provide psychological care had significantly lower burnout rates.²⁷ Individual interventions should focus on education, increasing awareness of burnout, providing psychological support, and offering communication training.²⁸ Additionally, organizational interventions, including resource provision, reducing administrative burdens, expanding the workforce, and infrastructure improvements, are equally crucial.¹⁹ Awareness about rising problems of professional burnout in oncology should also be increased at all levels since the problem is still underestimated, as shown by Holmes et al.,²⁹ and combining those approaches is likely to yield the most effective results, with organizational-level enhancements holding promise.

Declaration of generative AI and AI-Assisted technologies in the writing process

During the preparation of this work the author(s) used ChatGPT version 4 in order to improve readability and language. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.