Consequences of a toxic workplace environment on psychological safety in anaesthesia and intensive care

Abstract

BACKGROUND

Toxic workplace environments in healthcare, characterised by discrimination, bullying, poor communication, and lack of support, have a negative effect on physicians’ well-being and patient safety. Anaesthesiologists and intensivists are at high risk of burnout due to the intensity of their work. However, the prevalence and consequences of a toxic work environment in these specialties across Europe remain unknown.

OBJECTIVES

The primary objective of this study was to evaluate psychological safety among European anaesthesiologists and intensivists and its relationship to toxic workplace environments. The secondary objectives were to explore associated sociodemographic and professional factors, perceived support and job satisfaction, the prevalence and consequences of toxic workplace experiences, and the coping strategies used in response to them.

DESIGN

Cross-sectional, multinational survey study.

SETTING

Online anonymous survey distributed via the European Society of Anaesthesiology and Intensive Care mailing list and social media platforms between April and June 2025.

PARTICIPANTS

Some 859 respondents: 491 women (57.2%), 364 men (42.4%) and 2 gender-fluid (0.2%). They were predominantly White European (87.2%). Most participants worked in academic hospitals, mainly in anaesthesia alone (80.6%).

INTERVENTIONS

None.

MAIN OUTCOME MEASURES

Psychological safety (Edmondson's 7-item scale), support/satisfaction (4-item scale), prevalence and consequences of toxic workplace experiences, and coping strategies.

RESULTS

The Psychological Safety Scale showed high reliability (α = 0.85); the mean score was 4.12 ± 1.21. Burnout was reported by >50% of respondents and associated with significantly lower psychological safety and support/satisfaction scores (P < 0.001). Poor communication (59.6%) and poor leadership (45.9%) were the leading causes of perceived toxicity. Thirty-five percent reported health problems, 17% considered changing specialty, and 14% effectively changed workplace. Coping strategies were mainly adaptive, though 7.9% reported self-medication.

CONCLUSIONS

Toxic workplace environments are prevalent in European anaesthesiology and intensive care, strongly associated with burnout and adverse health outcomes. Organisational reforms targeting leadership, communication, and psychological safety are urgently needed.

KEY POINTS

• Many anaesthesiologists and intensivists in Europe experience a toxic work environment, where low psychological safety and poor support contribute to burnout.

• The most reported issues are poor communication, weak leadership, heavy workload and bullying.

• To foster healthier teamwork within anaesthesiology and intensive care, teams need systemic changes that incorporate trust, improve communication and hold leaders accountable.

Introduction

A healthy work environment has a significant impact on the performance of healthcare professionals, particularly those working in high-stakes fields such as anaesthesia and intensive care. A toxic workplace environment can undermine physician well-being and include discriminatory practices, such as verbal aggression, a lack of respect among colleagues, poor communication, excessive workload, and inadequate teamwork.¹ Toxic workplace environments have been described as environments that include an often violent and cruel treatment of employees that can hinder their safety and health.² Recent reports indicate that toxic work environments, marked by discrimination, harassment, bullying, and poor teamwork, can drive professionals to leave their current positions in search of healthier work settings.^3,4 Moreover, they cause physicians to change their specialty or quit medicine altogether, which is particularly relevant given the current era's global shortage of physicians.^5,6 Such violent workplace environments are universally recognised as detrimental, posing serious health risks and leading to a range of physical health problems including cardiovascular diseases,⁷ musculo-skeletal diseases⁸ and respiratory problems.⁹ The psychological impact is equally severe, with mental health issues and burnout being particularly prevalent.^9–12 It can even lead to the development of unhealthy coping mechanisms such as alcohol or drug use.^13,14

Recent work has highlighted how gender-based mistreatment contributes to toxic workplace climates in anaesthesiology. A secondary analysis of nearly 6000 European anaesthesiologists found that being a woman, younger age, and perceptions of gender as a disadvantage in leadership or research were independent predictors of mistreatment, underscoring the structural inequities that may fuel burnout and dissatisfaction in our specialty.¹⁵

Burnout is a syndrome arising from chronic workplace stress.¹⁶ Anaesthesiologists and intensive care physicians are particularly vulnerable due to the inherently high-stakes, high-pressure nature of their work.^17–19 Key contributing factors include poor communication within the Operating Room (OR) team²⁰ as well as strained interactions with colleagues and patients.¹³ In Europe, burnout affects up to 24% of anaesthesiologists and intensivists.^13,21–25 A 2023 Swiss survey reported that more than half of anaesthesiologists were at high risk of burnout.⁶ Moreover, the study identified a strong correlation between burnout and a perceived lack of support from colleagues. Being within the first 2 years of residency or having >5 years of training also increased the risk of burnout.

To our knowledge, no study has yet explored the prevalence of a toxic workplace environment in anaesthesiology and intensive care in Europe, nor its impact on health and workplace satisfaction. The primary objective of this study was to evaluate psychological safety among European anaesthesiologists and intensivists and its relationship with toxic workplace environments. The secondary objectives were to explore associated sociodemographic and professional factors, perceived support and job satisfaction, the prevalence and consequences of toxic workplace experiences, and the coping strategies used in response to them. We hypothesised that anaesthesiologists and intensivists across Europe experience varying levels of workplace toxicity and psychological safety, and that lower psychological safety and perceived support would be associated with higher reported rates of burnout and adverse health outcomes.

Methods

This study employed a cross-sectional, primarily descriptive survey design, with exploratory analytical analyses to examine associations between psychological safety, workplace factors and self-reported outcomes. The study was conducted in accordance with the Declaration of Helsinki and the Tri-Council Policy. An ethical committee waiver was obtained (Kantonale Ethikkommission Zurich; Switzerland; Req-2024-01129). Reporting follows the Checklist for Reporting of Survey Studies (CROSS) guidelines,²⁶ depicted in the Supplemental Digital Content 1.

An anonymous, voluntary online survey was conducted between April and June 2025 to align with ESAIC's mailing and communication calendar and to maximise response rates before the summer holiday period. This timing also aligned with ethical approval, ensuring dissemination at the earliest feasible opportunity. Recruitment was carried out via the European Society of Anaesthesiology and Intensive Care (ESAIC) mailing list and social media platforms, and the ESAIC–National Anaesthesiologists Societies Committee (NASC). NASC representatives were encouraged to further disseminate the survey within their professional networks. Participation was voluntary and anonymous. Completion of the survey implied consent. No personally identifiable data was collected. Participants accessed the survey by scanning a QR code or by clicking on a link provided in the ESAIC E-Mail or social media post. After reading the introductory information, they proceeded directly to the survey, and completion of the first section indicated their consent to participate.

The survey was administered using SurveyMonkey (Momentive Inc., San Mateo, California, USA), a platform compliant with the Health Insurance Portability and Accountability Act (HIPAA)²⁷ and the General Data Protection Regulation (GDPR).²⁸ Data were securely stored on SurveyMonkey's encrypted servers. Access to the survey database was password-protected and restricted to the study authors. Upon closure of the survey, the dataset was exported to statistical and data management software for analysis (Sigmaplot, R, Excel).

Survey design

Following a literature review, the authors developed a survey that included demographic questions, validated scales, and new questions. The final survey (presented in full in the Supplemental Digital Content 2) consisted of three sections. The first section included sociodemographic data (including gender, country of medical practice and training, years of experience, and history of psychiatric or psychological conditions). The second section included questions on job satisfaction, along with the validated seven-item Psychological Safety Scale developed by Edmondson et al.²⁹ The Edmondson 7-item Psychological Safety Scale measures the extent to which individuals feel comfortable taking interpersonal risks within their team, such as admitting mistakes or asking for help. Each item is rated on a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree), with higher scores indicating greater perceived psychological safety.²⁹ Because no single validated definition of a toxic workplace environment exists, we evaluated its core dimensions, psychological safety and perceived support, using the above instrument.²⁹ The last section included questions on the toxic work environment itself (prevalence, perceived causes, consequences, and coping strategies related to toxic workplace experiences). Maladaptive coping was assessed through a single item addressing self-medication; no further details on alcohol, recreational drug use, or other addictive behaviours were collected, to preserve anonymity and minimise the burden of response.

Before deployment, the survey was piloted among ten anaesthesiologists and intensivists from five European countries (Switzerland, Portugal, Romania, Greece, the United Kingdom and Norway). Based on their feedback, minor adjustments were made to improve clarity and consistency of wording and item sequencing.

Statistical analysis

Descriptive statistics were used to summarise sociodemographic characteristics. Categorical variables were reported as counts and percentages, and continuous or ordinal variables were summarised as mean ± (SD) or median [IQR], as appropriate.

The Edmondson Psychological Safety Scale (seven items rated on a 7-point Likert scale) was used to assess psychological safety.²⁹ An average score was computed across the seven items for each respondent. The internal consistency was evaluated using Cronbach's alpha. A threshold of 0.70 or higher was considered indicative of acceptable reliability.³⁰ Scores for items 2, 4, 6 and 7 were inverted since those were positive questions. A secondary 4-item scale assessing perceived support (job satisfaction, communication, peer support and leadership support) was similarly assessed for internal consistency.

The Mann–Whitney rank sum test was used for binary comparisons, and the Kruskal–Wallis test was used for comparisons involving more than two groups, with Dunn's post hoc comparison using the Benjamini–Hochberg adjustment. Associations between categorical variables were assessed using χ² tests. Effect sizes were quantified using Cramer's V, with values of 0.1, 0.3 and 0.5 interpreted as small, medium and large associations, respectively.³¹

Not all respondents completed every section of the questionnaire. The number of valid responses for each section and item is indicated in the corresponding table. Missing data were not imputed; analyses were performed using available responses only, consistent with the CROSS reporting guidelines.²⁶ For multiitem scales, mean scores were computed using available (nonmissing) responses for each participant. All statistical tests were two-sided, and a P-value < 0.05 was considered statistically significant. We performed all analyses using SigmaPlot 14 (Systat Software, San Jose, CA, USA) and R version 4.1.2 (R Foundation for Statistical Computing, Vienna, Austria) with RStudio (Posit PBC, Boston, MA, USA) with the following packages: tidyverse, rstatix, ggpubr, psych, vcd, readxl, ggplot2, janitor, dplyr, forcats.

Results

Sociodemographic data

A total of 859 people responded to the survey. Characteristics of participants, including gender, ethnicity, region of practice, professional experience and psychiatric history, are presented in Table 1. Of approximately 20 000 invited anaesthesiologists, 900 responded, corresponding to a response rate of about 4.5%.

Table 1.

Sociodemographic data, shown as counts and percentages (n = 859)

Gender (missing, n = 2)
Woman	491 (57.2)
Man	364 (42.4)
Genderfluid	2 (0.2)
Ethnicity (missing, n = 0)
Arab/Arab European	16 (1.9)
Asian/Asian-Pacific/Asian-European	14 (1.6)
Black/African/Caribbean/Black European	11 (1.3)
I prefer not to answer	10 (1.2)
Mixed/Multiple ethnic groups	29 (3.4)
South-East Asian/ South-East Asian European	13 (1.5)
South Asian/ South Asian-European	17 (2.0)
White European	749 (87.2)
Country of worka (missing, n = 4)
Eastern Europe	100 (11.6)
Northern Europe	93 (10.8)
Southern Europe	257 (29.9)
Western Europe	315 (36.7)
Outside Europe	90 (10.5)
Work in the same country as born in (missing, n = 2)
No – both medical school and work in another country	78 (9.1)
No – medical school in another country, work where I was born	37 (4.3)
No – medical school where I was born, work in another country	136 (15.8)
Yes – medical school and work in the country I was born in	606 (70.5)
Level of experience (missing, n = 2)
Board-certified/practicing anaesthesiologist (>5 years)	428 (49.8)
Board-certified/practicing anaesthesiologist (0–5 years)	124 (14.4)
Fellow or postgraduate/post completion trainee	30 (3.5)
Junior trainee/resident (0–2 years)	72 (8.4)
Senior trainee/resident (3 or more years)	197 (22.9)
Retired anaesthesiologist	6 (0.7)
Primary professional activity (missing, n = 0)
Anaesthesia	692 (80.6)
Anaesthesia and intensive care	30 (3.5)
Intensive care only	113 (13.2)
Emergency department	7 (0.8)
Leadership & management	3 (0.3)
Pain medicine & palliative care	10 (1.1)
Other b	4 (0.4)
Healthcare institution (missing, n = 2)
Regional/secondary hospital	172 (20.0)
Academic/tertiary hospital	612 (71.2)
Private practice	30 (3.5)
Mix (shared time between private and public hospitals)	43 (5.0)
Diagnosed or experienced symptoms of burnout/depression/anxiety or had dark thoughts (missing, n = 1)	454 (52.9)

^aNorthen Europe (Denmark, Estonia, Finland, Iceland, Ireland, Latvia, Lithuania, Norway, Sweden, United Kingdom), Southern Europe (Albania, Andorra, Bosnia and Herzegovina, Croatia, Gibraltar, Greece, Italy, Malta, Montenegro, North Macedonia, Portugal, San Marino, Serbia, Slovenia, Spain), Western Europe (Austria, Belgium, France, Germany, Liechtenstein, Luxembourg, Monaco, Netherlands, Switzerland), Eastern Europe (Belarus, Bulgaria, Czechia, Hungary, Poland, Republic of Moldova, Romania, Russian Federation, Slovakia, Ukraine).

^bOther included Internal Medicine, Physiotherapy, Prehospital and Research. Data are given as n (%).

Psychological safety

The Edmondson 7-item Psychological Safety scale demonstrated high internal consistency (Cronbach's α = 0.85) after reverse-scoring the positively worded items (items 2, 4, 6 and 7). The average psychological safety score across participants was 4.12 ± 1.21, with scores ranging from 1.00 to 7.00. In the overall cohort, 23.3% agreed and 8.2% strongly agreed with the statement, “If you make a mistake on this team, it is often held against you.”

Responses to the Edmondson Psychological Safety Scale varied widely (Fig. 1). For the negatively worded item “It is difficult to ask other members of this team for help”, most respondents disagreed, indicating that they generally felt comfortable seeking assistance from colleagues. For “If you make a mistake on this team, it is often held against you” and “Members of this team sometimes reject others for being different”, agreement was less frequent but still notable.

Fig. 1.

Distribution of psychological safety items in the overall cohort.

Stacked horizontal bar charts shows the proportion of respondents selecting each option on a 7-point Likert scale for seven items assessing psychological safety. Items 2, 4, 6 and 7 are positively worded, whereas items 1, 3 and 5 are negatively worded. Colours indicate the degree of agreement, from “Strongly disagree” (red) to “Strongly agree” (dark blue). Percentages are shown relative to the total number of respondents for each item.

For positively worded items, the highest endorsement was for “Members of this team are able to bring up problems and tough issues” and “Working with this team, my unique skills and talents are valued and utilised”. However, a significant minority disagreed with these statements. Responses to “It is safe to take a risk on this team” and “No one on this team would deliberately act in a way that undermines my efforts” were more evenly distributed across agreement and disagreement.

Regional differences and the influence of gender, experience, professional activity, and hospital type are presented in the Supplementary Figures S1 to S5, featured in the Supplementary Digital Content File 3.

Group comparisons

Table 2 presents the average psychological safety scores across sociodemographic and professional factors. Women reported symptoms of burnout more frequently than men, but average psychological safety scores did not differ significantly by gender. The two participants identifying as gender-fluid reported both burnout symptoms and the lowest psychological safety scores.

Table 2.

Average psychological safety score (PS score) influenced by sociodemographic data (n = 752)

	Median [IQR]	P value
Gender		0.126
Woman (n = 429)	4.14 [3.29 to 5.00]
Man (n = 321)	4.00 [3.29 to 5.00]
Gender fluid (n = 12)	2.29 [2.00 to 2.57]
Ethnicity		0.009
Arab/Arab European (n = 14)	3.50 [3.00 to 3.96]
Asian/Asian-Pacific/Asian-European (n = 10)	4.00 [3.50 to 4.21]
Black/African/Caribbean/Black European (n = 11)	3.57 [3.29 to 4.00]
I prefer not to answer (n = 10)	3.00 [2.75 to 3.61]
Mixed/Multiple ethnic groups (n = 27)	4.43 [3.79 to 5.36]
South-East Asian/ South-East Asian European (n = 12)	3.86 [3.57 to 4.79]
South Asian/ South Asian-European	3.57 [2.86 to 3.86]
White European (n = 657)	4.14 [3.29 to 5.00]
Country of work a		< 0.001
East Europe (n = 82)	3.29 [2.71 to 4.39]
North Europe (n = 87)	4.86 [4.00 to 5.71]
Outside Europe (n = 76)	3.86 [3.29 to 4.75]
South Europe (n = 224)	3.71 [3.14 to 4.57]
West Europe (n = 281)	4.57 [3.57 to 5.29]
Work in the same country as born in		0.641
No – both medical school and work in another country (n = 68)	4.00 [3.57 to 5.00]
No – medical school in another country, work where I was born (n = 34)	4.07 [3.43 to 4.57]
No – medical school where I was born, work in another country (n = 121)	4.00 [3.29 to 4.71]
Yes – medical school and work in the country I was born (n = 529)	4.14 [3.29 to 5.00]
Level of experience		0.229
Board-certified/practicing anaesthesiologist (>5 years) (n = 384)	4.14 [3.43 to 5.14]
Board-certified/practicing anaesthesiologist (0–5 years) (n = 107)	3.86 [2.86 to 5.00]
Fellow or postgraduate/post completion trainee (n = 21)	3.71 [2.50 to 5.00]
Junior trainee/resident (0–2 years) (n = 62)	4.14 [3.29 to 5.00]
Senior trainee/resident (3 or more years) (n = 174)	4.00 [3.14 to 4.86]
Retired anaesthesiologist (n = 4)	4.00 [4.00 to 4.25]
Primary professional activity		0.015
Anaesthesia (n = 610)	4.14 [3.32 to 5.00]
Anaesthesia and intensive care (n = 28)	4.29 [3.39 to 4.75]
Emergency department (n = 5)	4.29 [4.14 to 5.14]
Intensive care (n = 99)	3.71 [3.00 to 4.57]
Pain medicine (n = 7)	3.00 [2.79 to 4.00]
Other b (n = 5)	5.86 [5.43 to 6.10]
Healthcare institution		0.822
Academic/tertiary hospital (n = 544)	4.14 [3.29 to 5.00]
Regional/secondary hospital (n = 148)	4.00 [3.29 to 4.86]
Private practice (n = 27)	3.86 [3.54 to 4.86]
Mix (shared time between private and public hospitals) (n = 34)	4.00 [3.43 to 5.07]
Diagnosed or experienced symptoms of burnout/depression/anxiety or had dark thoughts (n = 397)	3.71 [3.00 to 4.71]	< 0.001

Data are given as median [IQR]. For statistical analysis Kruskal–Wallis test was use with Dunn's post hoc analysis with Benjamini–Hochberg correction when appropriate. The Mann–Whitney Rank test was used for those variables where only two groups were present. Differences in the number of observations (count = n) between Table 1 and Table 2 are due to missing values. P values are reported as overall P values; pairwise analysis is reported in the supplementary data file.

^aNorthern Europe (Denmark, Estonia, Finland, Iceland, Ireland, Latvia, Lithuania, Norway, Sweden, United Kingdom), Southern Europe (Albania, Andorra, Bosnia and Herzegovina, Croatia, Gibraltar, Greece, Italy, Malta, Montenegro, North Macedonia, Portugal, San Marino, Serbia, Slovenia, Spain), Western Europe (Austria, Belgium, France, Germany, Liechtenstein, Luxembourg, Monaco, Netherlands, Switzerland), Eastern Europe (Belarus, Bulgaria, Czechia, Hungary, Poland, Republic of Moldova, Romania, Russian Federation, Slovakia, Ukraine).

^b(Leadership n = 1, management n = 1, palliative care n = 1, physiotherapy n = 1, prehospital n = 0, research n = 1, internal medicine n = 0).

Participants with a history of burnout, depression, anxiety, or dark thoughts had significantly lower psychological safety scores compared with those without such experiences (P < 0.001). Regional variation was marked, with the lowest psychological safety scores in Eastern and Southern Europe and the highest in Northern and Western Europe (P < 0.001). Psychological safety also differed significantly by ethnicity (P = 0.009), although posthoc testing only confirmed a difference between those who preferred not to answer and participants identifying with multiple ethnic groups.

Psychological safety scores did not vary significantly by level of experience, place of birth, education, or hospital type. However, scores differed across primary professional activities, with intensive care and pain medicine showing lower values than anaesthesia (P = 0.015).

Four-item support/satisfaction scale

The 4-item Support/Satisfaction scale demonstrated acceptable reliability (Cronbach's α = 0.768). Median scores were similar between men and women and did not differ significantly by level of experience or hospital type. History of burnout, however, was strongly associated with lower scores (P < 0.001), reinforcing the link between workplace strain and reduced perceived support.

Scores also varied significantly by region, with the highest values reported in Northern Europe and the lowest in Eastern Europe (P < 0.001). Differences across professions were significant, with intensive care and pain medicine showing lower scores than anaesthesia (P = 0.047).

Cramer's V values indicated weak associations between demographic/professional variables. The largest effect size was for profession vs. gender (V = 0.108), followed by country vs. gender (V = 0.110). All other associations were ≤ 0.09. (Data not shown)

Toxic work environment

Discomfort at work due to professional interactions was reported by 15.4% of respondents as experienced daily and by 34.1% at least weekly. Monthly and a few-times-per-year frequencies were reported by 18.4% and 16.1%, respectively, while 3.3% reported never experiencing such discomfort.

The most frequently perceived causes were poor communication (59.6%), poor leadership, and resistance to change (45.9%). Excessive workload or poor work-life balance (38.6%), lack of support and inadequate resources (34.3%), and discrimination/ gossip and office policies (27.9%) were frequently reported. Bullying or unethical behaviour was reported by 30.7% of participants, and a lack of staff by 27.0%. Consequently, 27.5% reported feeling that patient or self-safety was endangered. Health problems were reported by 35%, and 79% of those reporting mental health issues, 44.8% physical health problems, 3.9% absenteeism, and 3.6% substance abuse. As a result, 17% of the responders thought about changing specialties, and 14% changed workplaces.

Cramer's V analysis revealed several statistically significant associations between toxic work environment variables and demographic and professional characteristics (Figure S2, Supplementary Table S1, Supplemental Digital Content). The strongest associations were observed between history of burnout and mental health issues (V = 0.388, P < 0.001) as well as between history of burnout and self-reported health problems due to the work environment (V = 0.406, P < 0.001), both indicating medium effect sizes. Country of work was also associated with reporting health problems (V = 0.236, P < 0.001) and high-frequency discomfort (V = 0.208, P < 0.001), with small-to-moderate effects. History of burnout, country differences, and differences by profession are noted in the supplementary material (Supplementary Digital File 3)

Coping strategies

Overall, 83.4% of respondents reported using at least one adaptive coping strategy. The most common were seeking support from family/friends (57.4%) and from colleagues/supervisors (42.6%), engaging in stress-relief activities (39.3%), and removing oneself from harmful situations (34.1%). Less often, the participants reported seeking professional help (14.4%) or making an official complaint (9.2%). Maladaptive coping was less common, but 7.9% reported self-medication.

Discussion

This multinational survey of toxic work environments revealed several patterns. More than half reported symptoms of burnout, depression, or anxiety. A history of burnout was strongly linked to lower psychological safety and support/satisfaction scores. Women reported having endured burnout more frequently than men, although the average psychological safety did not differ by gender. psychological safety scores varied widely by region, with the highest scores in Northern and Western Europe and the lowest scores in Eastern Europe. Similar trends were seen in support and satisfaction scores. Most participants worked in their birth country, in academic hospitals, and practiced anaesthesia alone. Despite most of the participants being board-certified and having significant professional experience, a substantial percentage still perceived difficulties in asking for help or taking risks at work.

While this work suggests that psychological safety is uneven between regions of Europe, a substantial proportion of professionals still encounter poor communication and overall low psychological safety. Our findings confirm trends observed in other studies,^15,18 suggesting that many operating rooms and intensive care teams operate in environments with limited trust and openness. In high-stakes settings, where rapid decision-making, collaboration, and error reporting are essential, this lack of psychological safety can suppress the desire to speak-up, hinder learning, and potentially compromise patient outcomes and patient safety.^32–34 The frequent reporting of difficulties in asking for help suggests that hierarchical barriers remain a significant obstacle to effective teamwork.

The link between previous experience of burnout and toxic workplace indicators, including mental and physical health issues, is consistent with the latest single-country study findings.^15,35 Notably, our data reveal a gender imbalance, with women more frequently reporting previous experience of burnout. Several studies have reported higher burnout rates among female anaesthesiologists,^6,36 with a recent Swiss study also revealing an association with the impostor phenomenon.³⁶ A phenomenon which may be reinforced – or even imposed externally – by a toxic workplace climate and exclusionary cues encountered in professional settings.^37,38

Coping strategies reported by participants include adaptive approaches, such as seeking social support or engaging activities to relieve stress. Interventions such as mindfulness, group coaching, and art therapy have been advocated to help cope with stress at work and alleviate some of the strains of a toxic workplace environment.^39–42 However, they usually work solely on individual-level strategies and interventions, risking the oversight of the root causes of a toxic work environment.⁴³ Pro-active measures, such as filing formal complaints or engaging in mentorship, were reported as uncommon, which may reflect a lack of access or trust in institutional systems or concerns about potential reprisal.⁴⁴ Future work should endeavour to uncover the root causes of these environments by studying how the culture of medicine itself promotes toxic behaviour and stress. This supports the interpretation that the problem is systemic rather than individual, necessitating organisational rather than solely personal interventions⁴⁵ and underlining the importance of transparent, accessible reporting mechanisms and leadership accountability.³⁵

Finally, a notable number of respondents reported maladaptive behaviour, including self-medication. This finding is particularly concerning given the well documented susceptibility of anaesthesiologists and intensive care physicians to addiction and suicide,^13,46 Such risks underscore the urgent need for systemic efforts to address toxic work environments and to foster a culture that safeguards physician well-being.

Limitations

This study has several limitations. Its cross-sectional design allows us to identify associations but not to establish causal relationships between toxic work environments, burnout, and health outcomes. The low overall response rate (~4.5%) may have introduced selection bias, as individuals with more pronounced experiences or opinions about workplace culture may have been more likely to respond. Consequently, the prevalence of toxic workplace environments and low psychological safety observed in our sample may overestimate the true values. Nevertheless, the multinational nature of the survey and consistency of our findings with single-country reports suggest that the observed trends remain relevant and credible. The use of self-reported data may have introduced recall or social desirability bias, especially regarding sensitive topics such as mental health, burnout, or workplace behaviour. In addition, participation was voluntary, and recruitment occurred through ESAIC mailing lists and social media platforms, which may have led to selection and nonresponse bias, as individuals with stronger opinions or difficulties were more likely to respond. Due to the method of recruitment, no sample size could be assessed. Notably, our sample included an overrepresentation of women, which prevents us from drawing firm conclusions regarding gender disparities. Other subgroups, such as gender-diverse participants, ethnic minorities, and professionals outside Western and Southern Europe, were underrepresented, as the sample was not weighted to reflect the broader European anaesthesiology and intensive care workforce. This lack of accurate representation limited our ability to conduct meaningful subgroup analyses. Although appropriate statistical methods with adjustments were applied, under-sampling of smaller groups may have affected the precision of our estimates. Given the exploratory nature of this study and the number of statistical comparisons performed, the risk of type I error cannot be excluded. Therefore, the reported P-values should be interpreted with caution, and the findings should be considered hypothesis-generating rather than confirmatory.

Furthermore, cultural and language differences may have influenced how items were interpreted, as the survey was administered in English across many countries where English is not the primary language. This may also have introduced a bias in favour of academic physicians, who are more accustomed to engaging with English-language materials. Although validated scales were used, psychological safety and support/satisfaction were assessed using relatively brief instruments that may not fully capture the complexity of these constructs. Finally, the absence of longitudinal data prevents us from determining changes over time or evaluating the impact of institutional interventions, while unmeasured contextual factors such as hospital policies, national healthcare systems, and local labour conditions may also have influenced perceptions of workplace toxicity.

Future directions

Despite these limitations, this study highlights important avenues for future research. Further work should explore the structural and cultural drivers of regional variation in psychological safety and workplace climate, moving beyond descriptive analyses toward evaluating the impact of systemic interventions. Studies assessing leadership accountability, communication frameworks, and institutional safety culture will be particularly valuable. In addition, qualitative approaches could provide deeper insights into barriers to reporting, trust in institutional systems, and models of successful cultural transformation.⁴⁷ Embedding principles of psychological safety, together with training in healthy coping strategies, into postgraduate education and institutional policies represents an urgent next step.

Conclusion

This European survey underscores that toxic work environments and uneven psychological safety remain a pressing concern for anaesthesiology and intensive care across Europe. The consistent association between low psychological safety, burnout, and adverse health outcomes highlights the need for urgent institutional and cultural reform. Addressing these challenges requires moving beyond individual-focused coping strategies towards systemic interventions that build trust, support open communication, and ensure leadership accountability. Ultimately, creating safe, respectful, and supportive workplaces is not optional; it is an essential prerequisite for sustaining physician well-being and delivering high-quality patient care.