Unethical Leadership in Critical Care Teams Can Increase Burnout Among Healthcare Professionals

Abstract

Background

Working in healthcare imposes a significant psychological burden on professionals, affecting their attitudes and job satisfaction. The pivotal role in the effective functioning of the team is that of the leader, who must exercise effective management. In addition to managerial activities, the leader’s role encompasses fostering a positive team atmosphere and ensuring the provision of adequate support. The main objective of this study was to measurably assess the impact of critical care team leaders’ negative behaviors on the incidence of professional burnout among medical professionals.

Material/Methods

Online cross-sectional survey was sent via social media groups and channels to Polish medical professionals who are members of critical care teams (but not team leaders). We obtained 146 replies. The survey consisted of a consent form, demographic questions, a self-designed questionnaire concerning non-technical aspects of leadership in their critical care teams, and the Oldenburg Burnout Inventory (OLBI) questionnaire.

Results

Principal Component Analysis (PCA) divided the self-designed questionnaire into 2 subscales: “unethical leadership” and “flurried leadership”. The Cronbach’s alpha of the questionnaire was 0.814. OLBI total score and exhaustion were positively correlated with results of our self-designed questionnaire: total and ‘unethical leadership’ subscale.

Conclusions

Unethical leadership in critical care teams can increase burnout among health professionals. Higher levels of exhaustion in the OLBI questionnaire were observed among participants who report lower level of leaders’ ethics. There is a need for further studies concerning this subject. Educational efforts are needed to improve the level of ethics among healthcare team leaders.

Introduction

Working in healthcare imposes a significant psychological burden on professionals, affecting their attitudes and job satisfaction. Medical professionals face constant evaluation and must take responsibility for their decisions, including difficult life-saving measures [1]. Furthermore, the environment of a medical facility, such as a hospital emergency department or intensive care unit, is characterized by an abundance of stimuli that can result in cognitive overload among personnel. Such an overload occurs when people receive too much information or face tasks exceeding their mental capacity. The functioning of healthcare teams has been shown to reduce the impact of these adverse factors, thereby reducing cognitive overload and increasing the efficiency and self-reliance of individual team members [2]. Nevertheless, the pivotal role in the effective functioning of the team is that of the leader, who must exercise effective management. In addition to managerial activities, the leader’s role encompasses fostering a positive team atmosphere and ensuring the provision of adequate support. Healthcare teams face many stressful situations and only a well-designed management structure can overcome team members’ emotional difficulties. Healthcare team structure is hierarchical, which ensures proper responsibility assignment but brings a risk of adverse leadership patterns. Unfortunately, the conduct of team leaders frequently falls short of ethical standards and compromises the positive atmosphere that is conducive to the well-being of team members.

Team members’ comfort should be taken under consideration and addressed. Otherwise, in severe cases, poor leadership ethics can result in professional burnout among team members. The condition, known as “professional burnout”, is characterized by a state of exhaustion in the psychological and pathological domains, stemming from occupational demands, a sense of de-personalization, and a perceived decline in personal competence [3]. It can be briefly described as a state of reduced engagement, lack of motivation, and higher level of exhaustion connected with work.

Fortunately, recent studies have underscored the profound impact of toxic leadership on healthcare professionals. Such a leadership pattern takes place, when the way orders are given is harmful for others, causing distress and increased anxiety. For instance, a 2023 study by Ofei et al [4] examined the effects of nurse managers’ toxic behaviors on nurses’ job satisfaction and intention to resign. The findings revealed that nurses subjected to toxic leadership exhibited higher turnover intentions, with job satisfaction mediating this relationship. This indicates that toxic leadership diminishes job satisfaction, which in turn increases the likelihood of nurses leaving their positions.

Similarly, a cross-sectional survey conducted by Hu et al [5] at an academic medical center explored the correlation between physician burnout and leadership behaviors. The study found that certain leadership behaviors, particularly passive management and laissez-faire approaches, were associated with higher levels of burnout among physicians [6]. Conversely, transformational leadership behaviors, such as individualized consideration and idealized influence, are correlated with lower burnout levels. These results suggest that adopting positive leadership behaviors can mitigate burnout, while negative behaviors can exacerbate it.

According to background described above, we assume that certain unfavorable behaviors of leaders in critical care medicine can increase burnout among healthcare workers. The main objective of this study was to measurably assess the impact of negative behaviors of a team leader on the incidence of professional burnout among medical professionals.

Material and Methods

We distributed an online survey through social media to Polish medical professionals in critical care teams, excluding team leaders. They were excluded, since the aim was to determine how team members assess ethical aspects of leadership and how it correlates with burnout. Those who usually lead could assess other team leaders in a different way than members do. We obtained 146 replies (32 physicians, 38 paramedics, and 76 nurses; 97 women and 49 men; age 38.6±11.2). Online recruitment was a convenient way to collect replies, but it brings numerous limitations (described subsequently). Otherwise, sampling would include only a few hospitals, which could result in bias. The survey consisted of a consent form, demographic questions (eg, age, work experience in years, sex, profession), a self-designed questionnaire concerning non-technical aspects of leadership in their critical care teams (Table 1), and the Polish version of the OLBI questionnaire [7].

Table 1.

Self-designed questionnaire.

Item number	Please indicate to what extent the following statements describe your experience of working with a critically ill patient	1= I fully disagree	2= I don’t agree	3= Neutral	4= I agree	5= I fully agree
IT1	I often hear: Hurry up! Faster!	1	2	3	4	5
IT2	I often hear vulgar words (eg, ‘fuck’)	1	2	3	4	5
IT3	The leader waves his/her hands when things don’t go his/her way	1	2	3	4	5
IT4	The leader throws things when he/she feels like something isn’t working	1	2	3	4	5
IT5	The leader issues many orders that cannot be carried out simultaneously	1	2	3	4	5
IT6	The leader shouts: Where is…?! Someone bring…!	1	2	3	4	5
IT7	The leader speaks in a calm voice	1	2	3	4	5
IT8	When giving orders, the leader indicates a specific person	1	2	3	4	5
IT9	The leaders do not reveal their negative emotions	1	2	3	4	5
IT10	I am being blamed by the leader for something not working properly	1	2	3	4	5

In further analysis, items 7, 8, and 9 were inverted (R=reverse); therefore, high total results in the questionnaire reveals less ethical and more chaotic leadership.

Categories of burnout levels in the OLBI questionnaire were set according to percentiles: ‘low’ (percentiles <50), ‘moderate’ (percentiles 50–85) and ‘high’ (percentiles >85).

Statistical analysis was performed with Jamovi 2.5.3 software. Elements of validation of the self-designed questionnaire were done with Cronbach’s alpha, Pearson’s correlation test, and principal component analysis with Varimax rotation.

Shapiro-Wilk tests were performed to check if the whole data meets normal distribution. Result revealed a lack of normality for age (W=0.922, P<0.001) and results of the self-designed questionnaire: total score (W=0.981, P=0.040), ‘unethical leadership’ subscale (W=0.956, P<0.001) and ‘flurried leadership’ (W=0.967, P=0.002). Only OLBI results met the criteria for normal distribution: total score (W=0.987, P=0.185), exhaustion subscale (W=0.982, P=0.057), and disengagement (W=0.984, P=0.099).

Further analysis used the non-parametric Mann-Whitney U test and Spearman’s rho correlation.

The Bioethical Commission of Gdańsk Medical Chamber (Poland) determined that a survey study does not require commission permission. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2024).

Results

Validation

The Cronbach’s alpha value of our self-designed questionnaire was 0.814. Correlations between each item and total result are presented in Table 2. Correlations between items are presented in Figure 1. Principal component analysis with Varimax rotation was performed. Two significant components were revealed, with cumulative variance 61.4% (Table 3). Bartlett’s test showed χ²=559 (P value <0.001).

Table 2.

Validation parameters of each item.

Item number	Mean	SD	Skewness	Kurtosis	Item – total correlation (r Pearson value)	P value	Measure of Sample Adequacy (KMO)
IT1	2.37	1.21	0.402	−0.968	0.551	<0.001	0.901
IT2	2.50	1.15	0.209	−1.020	0.338	<0.001	0.842
IT3	2.24	1.07	0.529	−0.560	0.652	<0.001	0.856
IT4	1.75	1.03	1.250	0.626	0.662	<0.001	0.878
IT5	2.45	1.22	0.320	−1.070	0.592	<0.001	0.835
IT6	2.63	1.23	0.217	−1.050	0.606	<0.001	0.858
IT7 (R)	2.67	1.13	0.502	−0.506	0.261	<0.001	0.674
IT8 (R)	2.65	1.13	0.521	−0.622	0.328	<0.001	0.719
IT9 (R)	3.04	1.20	−0.032	−0.861	0.438	<0.001	0.739
IT10	1.94	1.03	0.812	−0.231	0.548	<0.001	0.894

Figure 1.

Correlations between items of self-designed questionnaire.

Table 3.

Loadings of components in PCA with Varimax rotation. Loadings <0.3 are not presented.

Item number	Component 1	Component 2	Specificity
IT1	0.721		0.477
IT2	0.595		0.614
IT3	0.823		0.320
IT4	0.774		0.371
IT5	0.785		0.381
IT6	0.759		0.412
IT7 (R)		0.852	0.274
IT8 (R)		0.873	0.234
IT9 (R)		0.840	0.261
IT10	0.681		0.514

PCA – principal components analysis.

Since the PCA with Varimax rotation divides items into 2 groups, items IT7 (R), IT8 (R) and IT9 (R) formed a subscale named ‘flurried leadership’ (Cronbach’s alpha=0.827) and the others (items 1–6 and 10) formed a subscale named ‘unethical leadership’ (Cronbach’s alpha=0.858).

The Cronbach’s alpha of the OLBI questionnaire was 0.782. Categories of burnout levels were set according to percentiles: ‘low’ (percentiles <50), ‘moderate’ (percentiles 50–85) and ‘high’ (percentiles > 85). Cut-off levels for total score were: ‘low’ (≤38), ‘moderate’ (39–44) and ‘high’ (≥45). Cut-off levels for exhaustion subscale were: ‘low’ (≤19), ‘moderate’ (20–23) and ‘high’ (≥24). Cut-off levels for disengagement subscale were: ‘low’ (≤18), ‘moderate’ (19–21) and ‘high’ (≥22).

General Results and Statistical Tests

The study group consisted of 32 physicians (mostly younger ones, who were residents), 38 paramedics, and 76 nurses; 97 participants were females while 49 were males. Females had statistically higher results in ‘unethical leadership’ subscale in comparison with males (16.4±5.62 versus 14.8±6.21; P=0.047; r=0.202) with no significant difference in OLBI test results between sexes. There were no differences between professions in the OLBI and our self-designed questionnaire results. Work experience (years) and age were not correlated with the OLBI and self-designed questionnaire results.

OLBI total results and exhaustion were positively correlated with the results of the self-designed questionnaire total and ‘unethical leadership’ subscale. These findings are presented in Table 4.

Table 4.

Correlations between OLBI and self-designed questionnaire results (Spearman’s rho).

OLBI	Self-designed questionnaire
	Total	Unethical leadership subscale	Flurried leadership subscale
OLBI total	rho 0.240* P value 0.003	rho 0.247* P value 0.003	rho 0.098 P value 0.240
OLBI exhaustion	rho 0.231* P value 0.005	rho 0.245* P value 0.003	rho 0.087 P value 0.295
OLBI disangagement	rho 0.197* P value 0.017	rho 0.189* P value 0.023	rho 0.092 P value 0.270

^*Statistical significance (P values <0.05).

Participants with moderate or high OLBI total scores and exhaustion scores had statistically higher scores in the self-designed questionnaire (Total and Unethical leadership subscale). These finding are presented in Figure 2 and Table 5.

Figure 2.

Comparisons of participants’ self-designed questionnaire results between ‘low’ versus ‘moderate/high’ subgroups in OLBI questionnaire (Table 5). Statistically significant differences are marked with a star.

Table 5.

Results of U Mann-Whitney test comparing low versus moderate/high OLBI test results.

OLBI		Self-designed questionnaire
		Total		Unethical leadership subscale		Flurried leadership subscale
OLBI total	Low (n=65)	22.80±6.54	U=2029 P=0.017* r=0.229	14.68±5.48	U=2053 P=0.022* r=0.220	8.12±2.80	U=2371 P=0.301 r=0.099
	Moderate/high (n=81)	25.46±7.18		16.85±5.99		8.60±3.16
OLBI exhaustion	Low (n=54)	22.37±6.38	U=1822 P=0.007* r=0.267	14.06±5.09	U=1775 P=0.004* r=0.285	8.31±3.01	U=2364 P=0.626 r=0.048
	Moderate/high (n=92)	25.39±7.15		16.96±6.03		8.43±3.02
OLBI disangagement	Low (n=70)	23.11±6.39	U=2157 P=0.048* r=0.189	15.13±5.51	U=2283 P=0.139 r=0.142	7.99±2.67	U=2293 P=0.149 r=0.138
	Moderate/high (n=76)	25.34±7.41		16.58±6.10		8.76±3.26

^*Statistical significance (P values <0.05).

Discussion

Our study shows that a leader’s behavior significantly impacts the mental well-being of medical team members. Burnout affects staff at work and this state of exhaustion often gives rise to a deterioration in interpersonal relationships. Occupational burnout has been associated with impaired job performance, poor mental and physical health, and deterioration of relationships with family and friends [8]. These symptoms contribute to an increased risk of developing mental illnesses such as addiction to psychoactive substances, anhedonia, anxiety, or depression [9], which can directly impact quality of life.

We have shown that the increasing tendency to burnout in healthcare team members is closely correlated with negative leader behavior (significant weak correlations with rho values about 0.2). In this aspect, the results of our study are in line with the findings of many studies that have unequivocally shown correlations between a leader’s behavioral style and increased levels of workplace stress and burnout [6,10,11].

We divided the leader’s behavioral patterns into 2 subscales: the unethical leader and the flurried leader. These 2 subscales showed different levels of job burnout. The subscale describing the unethical leader were correlated with higher levels of job burnout (a significant but weak correlation) compared to the subscale describing the flurried leader (no statistically significant correlation). According to previous research, an authoritarian or aggressive leader’s behavior contributes to increased emotional exhaustion and decreased job satisfaction [12]. In contrast, chaotic, unstable, and emotional leader behaviors, described as the flurried leader, did not show a strong correlation with the level of job burnout. Flurried leadership is rather a sign of a leader’s disorganization and chaotic thinking, not aggressive behavior. This chaos often needs to be balanced by team members’ structured and calm performance. It can act as a motivator for non-leadership professionals, which is why it does not increase burnout. Chaos is not always damaging, and motivation can arise from a willingness to maintain a positive self-concept [13]. Some team members try to restore proper team functioning, which may act as an intrinsic motivator for them.

This result suggests that disorganization and impulsive decision-making affect dynamics in the workplace, but do not necessarily contribute to chronic stress or emotional exhaustion. A similar observation was made by Skogstad and Yukl, who found that team leader unpredictability leads to a higher incidence of temporary frustration for team members, but not lasting psychological strain [14,15].

The results of our self-designed questionnaire and individual subscales were checked using Cronbach’s alpha, reaching a high value of 0.8, which demonstrates the reliability of the results.

Experienced negative behaviors of the team leader did not show a statistically significant difference between subgroups. The doctors, nurses, and paramedics had similar levels of burnout. This finding is a noteworthy outcome of our research. This suggests that, despite the great advances in healthcare, there are still major shortcomings in terms of work culture and attitudes towards colleagues. One potential solution that merits exploration is the development of specialized courses for team leaders, accompanied by periodic recertification. These courses could focus on cultivating the skills necessary for effective group management and mitigating the adverse effects of toxic leaders on their employees [16]. The training programs for these courses should prioritize the development of emotional regulation, communication skills, and ethical decision-making. A study by West et al (2020) demonstrated the efficacy of early interventions to enhance leadership skills, as evidenced by a reduction in burnout and improved team performance within medical facilities [17].

The available literature specifically emphasizes the balance between different leadership styles. The main distinction is between 2 types of leadership: transformational and transactional. Transformational leadership is characterized by motivation and an individual approach to each team member, while transactional leadership is based on a system of reward and punishment. Transformational leadership has greater potential to reduce levels of job burnout and simultaneously increase job satisfaction [18,19], and transactional leadership has been shown to fail in high-stress environments [20].

A further analysis of the surveyed participants, categorized according to gender, indicated that women were more likely than men to report experiencing “unethical leadership” (P=0.047), but the effect of this finding was very small (r=0.202). This disparity may be attributed to divergent perceptions of gender-based stimuli. A study conducted by Fnais found that women in healthcare facilities were more likely to experience rudeness and emotional stress due to hierarchical dynamics in their workplace [21]. This aspect of gender alignment of interventions should also be included in the curriculum of a team management preparation course.

The findings of the present study underscore the validity of the crucial role of leadership in shaping the psychological well-being of critical care team members. Recent studies have demonstrated that ethical leadership can reduce emotional exhaustion, depressive episodes, and promote the shared objective of providing optimal patient care [22]. This finding is consistent with the results of the Investigating the Impact of Ethical Leadership on Aspects of Burnout study, which suggests that ethical leaders create a supportive environment that protects against workplace stress and increases mental resilience among healthcare workers, thereby helping to reduce burnout rates [23].

Our survey has certain limitations. The main limitation is the relatively small number of people participating in the survey, which may limit the reliability of our results, especially for the physicians subgroup, which was smaller than the others. The other major limitation of the study is the restriction of the study group to people working in very specific conditions, which are characterized by high levels of stress, often working in difficult and demanding conditions in which patients’ lives are at risk. Consequently, future research should be expanded to include healthcare workers who do not provide direct patient care, which could result in more diverse responses. Nurses, paramedics, and doctors have very different jobs, but they are all influenced by their emotions. Unfortunately, emotional frustration and subjective perception of events at work affect the answers given in our survey and can lead to bias, which directly affects the results. The presented limitation of our work certainly affects the results, but regrettably remains a difficult factor to eliminate.

Two strategies of emotional expression at work have been described. Deep one takes place when a person tries to alter felt emotions to harmonize displayed and experienced feelings. Surface strategy brings to dissonance, since individuals adjust shown emotions, but they hide within themselves true feelings. Even if surface acting is adverse for employees’ well-being, this effect is unclear. This suggested that work-related stressors are more relevant than emotional expression strategy [24], showing that unethical leadership is a stronger stressor than the emotional environment.

We administered the questionnaires via social media channels, which carries a risk of self-selection bias, meaning that individuals with strong negative feelings toward leadership behavior or those experiencing burnout are more likely to participate in the survey. Future studies should have wider sample selection in clinical settings. Moreover, it would be beneficial to recruit participants from all around the world, since differences in culture, education, and healthcare system organization could affect the results. Other variables (eg, workload, team atmosphere, personal traits) were not included in the study. The use of all scales known to be correlated with burnout is impossible in a single study, since few participants would be willing to fill out such a long questionnaire.

We did not use Bonferroni correction of statistical tests because the study was an exploratory not definitive, and the size effects were moderate. Thus, the risk of a type II error may overcome potential benefits. Harman’s single-factor test was not performed, since it no longer regarded as reliable [25].

Conclusions

Unethical leadership in critical care teams can increase burnout among health professionals. Higher levels of exhaustion shown by the OLBI questionnaire responses was observed among participants who reported unethical team leadership. There is a need for further studies concerning this subject. Educational efforts are needed to improve the level of ethical team leadership. Non-technical skills are becoming widely introduced in education of healthcare workers, and ethical aspects of leadership need to be discussed as well. Postgraduate training would benefit from such an educational intervention, as many new healthcare team leaders have little experience.