Occupational burnout and depression among emergency medical service providers in Switzerland and Liechtenstein: a cross-sectional nationwide web-based study

Abstract

Background

Emergency medical services (EMS) providers face elevated burnout risk due to job-related stressors, organizational challenges, and physical and emotional demands. Stigma and limited support further endanger EMS providers’ wellbeing. Burnout not only affects providers but can also compromise care recipients, and system sustainability. Research on EMS providers’ wellbeing remains limited and subject to moderate generalisability. This study examines the prevalence of burnout, depression, suicidal thoughts, and turnover intention among Swiss EMS providers, offering a national multicentre perspective across all 26 cantons plus Liechtenstein.

Methods

A web-based survey was distributed to all active EMS providers in Switzerland and Liechtenstein between May and June 2024. The questionnaire included validated patient-reported outcome measures of burnout (Maslach Burnout Inventory (MBI) and Copenhagen Burnout Inventory (CBI)), and depression (Patient Health Questionnaire-9). The primary outcome was the prevalence of burnout, and secondary outcomes were prevalences of depression, suicidal thoughts (measured using two dedicated additional questions) and turnover intention (evaluated with a specific frequency-based question). Multivariable logistic regressions were run separately for MBI and CBI outcomes to identify burnout determinants.

Results

Of 3669 eligible providers, 1485 completed the questionnaire (40.5%). Burnout prevalence was 43.4% (MBI) and 36.5% (CBI), with regional variations. Night shift frequency was consistently associated with increased burnout risk across both MBI and CBI in a dose-dependent manner. Age, gender, professional title categorized as “Other”, canton of work and childlessness, were associated with burnout in a scale-specific manner. Turnover intention was high, with 67.7% of respondents reporting having considered leaving their job at least once. A total of 13.4% of participants reported moderate to severe depression, and 10.2% indicated they had contemplated suicide, of which 46.7% within the past year.

Conclusions

Burnout is a significant concern among Swiss EMS providers. Night shift frequency emerged as the only predictor consistently associated with burnout, showing a dose-dependent effect, and highlighting it as a priority target for intervention. Scale-specific associations reflect the multidimensional nature of burnout. These findings underscore the urgent need for targeted, evidence-based interventions to protect EMS providers’ wellbeing and mitigate cascading effects on prehospital care quality, safety, and the sustainability of the healthcare system.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-026-26184-z.

Introduction

Burnout is recognized by the World Health Organization (WHO) as an occupational phenomenon resulting from chronic workplace stress that has not been effectively managed [1]. Although validated assessment tools exist, its definition and diagnostic criteria remain debated [2, 3]. A 2021 consensus defined occupational burnout as “physical and emotional exhaustion due to prolonged exposure to work-related problems” [4]. The Maslach Burnout Inventory (MBI) and the Copenhagen Burnout Inventory (CBI) are the most widely patient-reported outcome measures (PROMs) used for its assessment. The MBI identifies three domains: emotional exhaustion, depersonalization (cynicism and detachment), and reduced personal accomplishment [5]; the CBI measures burnout across personal, work-related, and client-related dimensions [6].

Over the past decade, work-related stress has increased significantly. In Switzerland, it affects 23% of the population, up from 18% ten years previously, with more than half of those affected reporting emotional exhaustion [7]. Association with certain risk factors, such as staffing shortages, night shifts, working hours, and exposure to highly stressful situations, suggests certain professions may be more at risk than others, notably those working in emergency care [8–11]. Worldwide, there is a variability in burnout prevalence in emergency medical service (EMS) providers, with a recent systematic review reporting rates ranging from 16% to 56% [12]. Differences in assessment tools, study populations, and burnout definitions may partly explain these differences, however organisational factors seem to play a recurrent role, with one study reporting burnout in over 50% of EMS providers within one single ambulance trust in northern England [13]. Driven by factors such as low job satisfaction, high workload, individual traits, certification level/function, as well as poor leadership support, burnout is linked to increased sickness absence and turnover, threatening the perennity of prehospital care [14–16]. In addition, it also poses long-term health risks for personnel, such as increased risk of cardiovascular disease, shorter life expectancy, and increased use of psychotropic medications [17]. Finally, it also negatively affects patient care, increasing safety incidents and decreasing quality of care and satisfaction [18–21]. As such, burnout undermines the wellbeing of providers and care recipients, and puts the system at risk of unsustainability.

Beyond burnout, EMS providers experience high rates of post-traumatic stress disorder (PTSD), poor mental health, and psychological distress [22]. Unlike burnout, depression has a standardized clinical definition, now encompassing work-related factors, in the Diagnostic and Statistical Manual of Mental Disorders (DSM-V) [23, 24]. Its strong association with burnout makes it a valuable indicator of wellbeing and resilience among emergency care workers [24]. A recent systematic review suggests that burnout is associated with depression, anxiety and suicidality, underlying the importance to assess these outcomes alongside burnout [25]. In addition, depression increases the risk of shift work disorder [26], and the association between psychological vulnerability and suicidality in EMS providers demands further investigation [27]. With facilitated access to lethal means, the risk of suicide is indeed a significant concern [28]. To highlight the significance of these issues, the WHO has underlined the economic sense to invest in mental health [29]. Despite its impacts and consequences, research on EMS providers’ wellbeing is limited. In Switzerland, recent studies have provided only partial insights into wellbeing and resilience in healthcare, with EMS providers either not or underrepresented [11, 30, 31]. Such underrepresentation is problematic. EMS providers operate under working conditions that differ from those of other emergency care workers. Their role involves, among other factors, unpredictable mission profiles, physical demands, recurrent exposure to traumatic events and high-acuity events in uncontrolled environments, autonomous clinical decision-making under time pressure, prolonged shifts with irregular rest opportunities, and distinctive organizational structures [10, 13, 16, 32]. These features impose unique stressors and recovery constraints that may influence the onset, progression, and consequences of occupational mental health outcomes in ways that differ from other settings. Investigating EMS providers as a distinct professional group is therefore essential to capture these specific dynamics, to better understand their wellbeing and resilience, and to guide the development of tailored preventive strategies.

In a bid to better understand Swiss EMS providers’ wellbeing, and employing the same design as previous and ongoing national studies on emergency physicians and emergency nurses [11, 33], this study examines the prevalence of burnout, depression, suicidal thoughts, and turnover intention using the validated MBI, CBI, and PHQ-9 PROMs.

Methods

Study design and setting

This was a nationwide, cross-sectional, open web-based study among EMS providers in Switzerland and Liechtenstein. At the time of the study, the Swiss prehospital system employed 3653 EMS providers in 101 EMS organizations [34]. Liechtenstein had a single EMS organisation employing 16 providers who were trained and certified at Swiss paramedic colleges. The Swiss Interassociation of Rescue (IVR-IAS), which also certifies Liechtenstein’s EMS, provided the official EMS organizations registry [35].

In Switzerland, EMS teams typically consist of basic life support (BLS) drivers, emergency medical technicians (EMTs) with a federal diploma (requiring 1800 h of training), and paramedics with an advanced federal diploma of higher education (an extensive 5400-hour training) [36]. Paramedics may also obtain post-graduate specializations in anaesthesiology, emergency care, intensive care, or pre-clinical care. Although each of the 26 cantons differ in legal and regulatory frameworks, it is mandatory for urgent missions that each ambulance include at least one paramedic. BLS drivers, often paired with an EMT, are typically involved in referral (secondary) missions. Notably, in the canton of Ticino, BLS drivers are not present on ambulances; however, volunteer rescuers who are BLS-trained may serve as additional crew members. In Liechtenstein, ambulances are staffed exclusively by paramedics.

Study population

Eligible study participants included all active EMS providers in Switzerland or Liechtenstein (n = 3669) regardless of professional education or function. There were no exclusion criteria.

Enrolment and study period

Recruitment was conducted through three parallel strategies:

• Professional association level: All 1591 members of the Swiss Paramedic Association received an invitation to participate by email; the association also shared the survey through its dedicated smartphone application.

• EMS organization level: Heads of all EMS organizations in Switzerland and Liechtenstein were contacted by email (which included information and a link to the survey) and were asked to forward it to their employees.

• Investigator level: The survey was promoted on social media platforms, including Meta (formerly Facebook), Swissrescue.ch (a website for prehospital healthcare providers), and LinkedIn.

Additionally, the survey was distributed to EMS providers who were not affiliated with the professional association, using a snowball recruitment.

The survey was open from May 21 to June 21, 2024. Two reminder emails were sent at weeks two and three. An information sheet was attached to provide additional details. No incentives were offered. Participation was voluntary and anonymous.

Data collection

The online survey was hosted on the MindGarden© platform and had been previously used in a study surveying emergency physicians [11]. It comprised four sections covering demographics, validated mental health measures, work-related concerns, and open-ended questions on wellbeing. No adaptive features or items randomization were applied, and completeness checks ensured full responses. The survey was multilingual and open-access, with built-in measures to prevent duplicate entries. A 24-hour mental health support contact was provided at the end. Full details of the data collection process are presented in Additional file 1. The results are reported in accordance with the Checklist for Reporting Results of Internet E-Surveys (Additional file 2) [37].

Variable description and outcomes

The following variables were collected and included in the multivariable models:

• Socio-demographic variables:

• Age: reported in years.

• Gender: categorized as male, female or non-binary.

• Language: primary working language of the participant (French, German, Italian, or Other).

• Relationship status: categorized as single, in a stable relationship, married, divorced, or widowed/other.

• Partner working as a healthcare professional: Yes, No, or Not applicable.

• Children: having child(ren) (Yes/No).

• Employment and work-related variables:

• Employment in %: participant’s work percentage relative to full-time employment.

• Working hours per week: average number of hours worked per week.

• Night shifts per month: average number of night shifts worked per month.

• Experience: years of professional EMS experience.

• Title: categorized as BLS driver/volunteer, EMT, Paramedic, or Other.

• Geographical variables:

• Canton (collapsed by population size): categories based on total population of the canton: Over One Million, 500,000 to 1,000,000, 250,000 to 500,000, Less than 250,000.

• Canton (collapsed by responses): canton of work, with cantons grouped into an “Other” category when individual sample sizes were < 50 to ensure stable estimates.

The primary outcome was burnout prevalence, measured using the MBI and CBI [6, 38, 39]. For the MBI, participants were considered to meet burnout criteria if they scored above the threshold in any domain: depersonalization ≥ 10, emotional exhaustion ≥ 27, and personal accomplishment < 34. For the CBI, participants were considered to meet burnout criteria if they scored ≥ 50 in any of the three domains: personal (P-CBI), work-related (WR-CBI) and client-related (CR-CBI) burnout. Using both instruments allows for cross-validation of results, enhances the robustness of our findings, and captures burnout experiences that may be specific to different professional roles within our study population.

Secondary outcomes included the prevalence and severity of depression (measured using the PHQ-9 [40–43]), suicidal thoughts, and turnover intention. The PHQ-9 cut-offs were: 0–4 (no symptoms), 5–9 (mild), 10–14 (moderate), 15–19 (moderately severe) and ≥ 20 (severe). Suicidal thoughts and turnover intention were assessed using specific additional questions:

• “Have you ever considered suicide during your career in prehospital care?”

• “If yes, have you considered suicide during the past 12 months?”

• “How often do you think about leaving your profession?”

This latter question included the following response options: once a shift, once a week, more than once a week, once a month, more than once a month, once a year, less than once a year, once, never.

Sample size

No sample size calculation was performed. Although the survey was disseminated as broadly as possible through national professional and organizational channels, no comprehensive roster of EMS providers exists in Switzerland and Liechtenstein. As participation depended on voluntary response across multiple dissemination pathways, the study relied on a non-probability convenience sampling strategy.

Statistical analysis

The complete dataset was first reviewed by T.S. and L.S. to ensure there were no inconsistencies. Specific outlier entries were excluded or recoded as missing as adjudicated by the study team. All authors had access to the finalized dataset. Statistical analysis followed the approach used by Heymann et al. [11]. Socio-demographic variables and secondary outcomes were described using median [quartiles] for continuous variables, based on their distribution, and n (%) for categorical variables.

Burnout scores for the MBI and CBI were calculated according to the respective questionnaire manuals. Logistic regression assessed associations between burnout (meeting ≥ 1 MBI or CBI criteria) and independent variables. Variables with p <0.1 in univariate analysis were entered into a backward stepwise multivariable logistic regression model (Wald removal criterion of 0.1). Likelihood ratio tests were used to assess the appropriateness of including variables in the model. Odds ratios were reported with 95% confidence interval (CI). Nagelkerke R² was used to determine the variance in burnout explained by the final model; this measure, though limited, is modified to have a maximum of 1, which facilitates interpretation compared to other methods. The same procedure was conducted for both the MBI and the CBI. Cantons with fewer than 50 responses were grouped for analysis into a single “Other” category to ensure sufficient sample size of each category. This cut-off was chosen to support a regional analysis approach while ensuring sufficient participants in each category for stable estimates in the final model. Non-binary participants were coded as missing due to insufficient variability for regression; excluding them did not affect the odds ratios or significance for female versus male comparisons.

A separate post hoc subgroup analysis was performed for one canton found to present a higher burnout prevalence, comparing it to all others. This post hoc analysis examined socio-demographic variables as well as each domain of the burnout scales to further refine the findings and identify potential factors that could explain the observed association within this canton. The purpose of this secondary analysis was to generate hypotheses and provide additional insight into potential regional differences in burnout prevalence, which could inform future studies or targeted interventions.

Depending on the distribution of the data, Chi-square test (or Fisher’s Exact Test when expected counts were < 5) was used for categorical variables; independent samples median test or Mann-Whitney U test were performed for non-normally distributed continuous variables.

Data analysis was performed in SPSS version 29 (IBM Corp., IBM SPSS Statistics for Windows, Version 29.0. Armonk, NY: IBM Corp), except for the subgroup analysis of categorical variables, which was conducted using Stata V15.1 (StataCorp LLC, College Station, TX, USA).

Results

The survey link was accessed 1736 times; 1488 surveys were completed, resulting in a dropout rate of 14.5% (251/1736). When stratified by language, the completion rates were 83.7% (118/141) for Italian, 85.2% (500/587) for French, 86.3% (867/1005) for German, and 100% (3/3) for English.

Three respondents were excluded from the analysis: one due to inactive status (retired), two because education level or professional title could not be identified. The final dataset included 1485 participants. Considering that 3653 EMS providers were currently employed in Switzerland and 16 in Liechtenstein, the participation rate stands at 40.5% (1485/3669). Socio-demographic variables are displayed in Table 1.

Table 1.

Socio-demographic and work-related variables

Variable	N = 1485 (unless otherwise specified)
Age (n=1484)	38.0 [33.0-46.0]
Gender
Male	887 (59.7)
Female	596 (40.1)
Non-Binary	2 (0.1)
Language
French	490 (33.0)
German	877 (59.1)
Italian	118 (7.9)
Employment in %	100.0 [80.0–100.0]
Working hours per week (n=1439)	42.0 [36.0–45.0]
Night shifts per month (n=1478)	6.0 [4.0–7.0]
Experience	12.0 [6.0–20.0]
Title
BLS driver/volunteer	22 (1.5)
EMT	103 (6.9)
Paramedic*	1283 (86.4)
Other	77 (5.2)
Description of «Other» title (n=77)
Instructor	1 (1.3)
Emergency dispatcher	6 (7.8)
Emergency dispatcher and specialization in emergency care	1 (1.3)
HEMS technical crew	1 (1.3)
Managerial function	7 (9.1)
Nurse	1 (1.3)
Physician	3 (3.9)
Specialized nurse (anaesthesiology, ICU or emergency care)	6 (7.8)
Trainee EMT	1 (1.3)
Trainee paramedic	44 (57.1)
Not specified	6 (7.8)
Canton
Aargau	65 (4.4)
Appenzell Inner Rhine**	2 (0.1)
Appenzell Outer Rhine**	11 (0.7)
Basel-City	60 (4.0)
Basel-Landschaft**	27 (1.8)
Bern	211 (14.2)
Fribourg	58 (3.9)
Geneva	96 (6.5)
Glarus**	1 (0.1)
Grisons	85 (5.7)
Jura**	15 (1.0)
Liechtenstein **†	8 (0.5)
Lucerne**	14 (0.9)
Neuchâtel	62 (4.2)
Nidwalden**	3 (0.2)
Obwalden**	2 (0.1)
Schaffhausen**	21 (1.4)
Schwyz**	31 (2.1)
Solothurn**	41 (2.8)
St. Gallen**	39 (2.6)
Thurgau**	15 (1.0)
Ticino	103 (6.9)
Uri**	16 (1.1)
Valais	92 (6.2)
Vaud	171 (11.5)
Zug**	12 (0.8)
Zurich	224 (15.1)
Canton collapsed by population size
Over One Million	435 (29.3)
500,000 to one million	275 (18.5)
250,000 to 500,000	446 (30.0)
Less than 250,000	329 (22.2)
Relationship status
Single	274 (18.5)
In a stable relationship	473 (31.9)
Married	621 (41.8)
Divorced	82 (5.5)
Widowed	24 (1.6)
Other	11 (0.7)
Partner working as a healthcare professional
Yes	571 (38.5)
No	749 (50.4)
Not applicable	165 (11.1)
Children
Yes	755 (50.8)
No	730 (49.2)
Number of children (n=755)	2.0 [1.0–2.0]
Age of the youngest (n=754)	8.0 [3.0–14.0]

Data expressed as median [quartiles] or n (%)

*Paramedics with a post-graduate specialization (anaesthesiology, ICU, emergency care or preclinical specialist) were categorized under “Paramedic”

HEMS Helicopter emergency medical services, ICU Intensive care unit, BLS Basic life support, EMT Emergency medical technician

**Cantons with fewer than 50 participants were grouped into the category “Other cantons” for the regression analysis

†Liechtenstein is a country, not a canton

Maslach burnout inventory

Of the 1485 participants, 644 (43.4%) met at least one burnout criterion according to the MBI. Specifically, 425/1485 (28.6%) met the depersonalization threshold (≥ 10), with a median score of 6.0 [quartiles 3.0–10.0], 137/1485 (9.2%) the emotional exhaustion threshold (≥ 27), with a median score of 12.0 [quartiles 7.0–18.0], and 345/1485 (23.2%) the threshold for low personal accomplishment (< 34), with a median score of 38.0 [quartiles 34.0–42.0]. Among participants meeting at least one burnout criterion (43.4%, 644/1485), the proportion meeting each specific criterion is shown in Fig. 1.

Fig. 1.

Distribution of Maslach Burnout Inventory burnout domains among participants meeting at least one burnout criterion (43.4%, 644/1485)

Each additional year of age was associated with a 2.1% decrease in the OR of meeting any MBI burnout threshold. Female participants had a 33.9% lower OR compared to males. Each additional night shift was associated with a 3.8% increase in the OR (Table 2). The model correctly classified 56.7% of cases and explained 3.0% of the variance in burnout (Nagelkerke R² = 0.03).

Table 2.

Meeting any MBI burnout criteria: regression analysis

	Univariate			Multivariable a
Variable	Odds ratio	95% CI	p-value	Odds ratio	95% CI	p-value
Age	0.98	0.97–0.99	0.001**	0.98	0.97–0.99	<0.001***
Gender
Male	Reference	Reference	Reference	Reference	Reference	Reference
Female	0.74	0.60–0.91	0.004**	0.66	0.53–0.83	<0.001***
Language
French	Reference	Reference	Reference
German	0.84	0.67–1.05	0.13
Italian	0.89	0.59–1.33	0.57
Employment in %	1.01	1.00–1.01	0.006**
Working hours per week	1.01	1.00–1.03	0.014*
Night shifts per month	1.06	1.02–1.10	0.001**	1.04	1.00–1.08	0.044*
Experience	0.99	0.98–1.01	0.28
Title
BLS Driver/volunteer	1.26	0.54–2.93	0.59
EMT	0.8	0.53–1.21	0.29
Paramedic	Reference	Reference	Reference
Other	0.64	0.40–1.05	0.08
Canton collapsed by population size
Over One Million	Reference	Reference	Reference
500,000 to one million	0.87	0.72–1.32	0.84
250,000 to 500,000	1	0.77–1.31	0.99
Less than 250,000	0.97	0.72–1.29	0.82
Canton collapsed by responses
Aargau	0.96	0.55–1.66	0.83
Basel-City	0.59	0.32–1.06	0.08
Bern	0.82	0.57–1.19	0.29
Fribourg	1.26	0.71–2.23	0.42
Geneva	0.73	0.45–1.18	0.19
Grisons	0.76	0.46–1.26	0.29
Neuchâtel	1.64	0.94–2.86	0.08
Ticino	0.98	0.62–1.55	0.93
Valais	1.02	0.63–1.64	0.95
Vaud	1.01	0.69–1.49	0.96
Zurich	1.16	0.81–1.66	0.43
Other	Reference	Reference	Reference
Relationship status
Single	1.52	1.14–2.02	0.004**
In a stable relationship	1.07	0.84–1.36	0.61
Married	Reference	Reference	Reference
Divorced	1.08	0.68–1.73	0.74
Widowed/other	1.23	0.62–2.43	0.56
Partner working as a healthcare professional
Yes	Reference	Reference	Reference
No	0.87	0.70–1.08	0.2
Not applicable	1.2	0.85–1.69	0.31
Children
Yes	Reference	Reference	Reference
No	1.21	0.99–1.49	0.07

^aVariables removed at univariate stage: language, experience, canton collapsed by population size, and partner working as a healthcare professional; variables removed during backward stepwise regression: employment, working hours per week, title, canton collapsed by responses, relationship status and children

HEMS Helicopter emergency medical services, ICU Intensive care unit, BLS Basic life support, EMT Emergency medical technician

Levels of significance: * for p <0.05; ** for p <0.01; *** for p <0.001

Copenhagen burnout inventory

Of the 1485 participants, 542 (36.5%) met at least one burnout criterion according to the CBI. Specifically, 444/1485 (29.9%) met the personal burnout cut-off with a median score of 37.5 [quartiles 25.0–50.0], 283/1485 (19.0%) the work-related burnout cut-off with a median score of 32.1 [quartiles 21.4–42.9], and 195/1485 (13.1%) the client-related cut-off with median score 20.8 [quartiles 8.3–37.5]. Fig. 2 illustrates the distribution of burnout across domains among participants meeting at least one CBI criterion.

Fig. 2.

Distribution of Copenhagen Burnout Inventory burnout domains among participants meeting at least one burnout criterion (36.5%, 542/1485)

Each additional night shift per month was associated with a 4.7% increase in the OR of meeting any CBI burnout threshold. Working in the canton of Neuchâtel was associated with a 2.3-fold increase in the OR. Conversely, holding a professional title categorized as “Other” and not having any child were associated with 48.6% and 22.8% reductions in the OR, respectively (Table 3). The model correctly classified 64.5% of cases and explained 3.6% of the variance in burnout (Nagelkerke R² = 0.036).

Table 3.

Meeting any CBI burnout criteria: regression analysis

	Univariate			Multivariable a
Variable	Odds ratio	95% CI	p-value	Odds ratio	95% CI	p-value
Age	1	0.99–1.01	0.99
Gender
Male	Reference	Reference	Reference
Female	0.98	0.79–1.21	0.84
Language
French	Reference	Reference	Reference
German	0.71	0.56–0.89	0.003**
Italian	0.74	0.49–1.13	0.16
Employment in %	1	0.99–1.01	0.84
Working hours per week	1	0.99–1.01	0.72
Night shifts per month	1.04	1.01–1.08	0.021*	1.05	1.01–1.09	0.015*
Experience	1.01	1.00–1.02	0.08
Title
BLS Driver/volunteer	0.49	0.18–1.33	0.16	0.54	0.19–1.54	0.15
EMT	0.93	0.61–1.42	0.74	0.88	0.57–1.36	0.52
Paramedic	Reference	Reference	Reference	Reference	Reference	Reference
Other	0.51	0.30–0.87	0.014*	0.51	0.29–0.90	0.020*
Canton collapsed by population size
Over One Million	Reference	Reference	Reference
500,000 to one million	1.26	0.92–1.73	0.15
250,000 to 500,000	1.2	0.91–1.58	0.20
Less than 250,000	1.15	0.85–1.55	0.36
Canton collapsed by responses
Aargau	0.88	0.50–1.55	0.66	0.95	0.53–1.68	0.85
Basel-City	0.64	0.34–1.17	0.15	0.61	0.33–1.14	0.12
Bern	0.8	0.55–1.17	0.24	0.81	0.55–1.19	0.28
Fribourg	1.22	0.68–2.17	0.51	1.2	0.67–2.15	0.54
Geneva	1.05	0.65–1.70	0.84	1.03	0.63–1.66	0.92
Grisons	0.63	0.37–1.08	0.09	0.65	0.38–1.12	0.12
Neuchâtel	2.08	1.19–3.65	0.010*	2.35	1.32–4.17	0.004**
Ticino	0.86	0.54–1.39	0.54	1.02	0.62–1.68	0.94
Valais	0.78	0.47–1.29	0.33	0.79	0.48–1.32	0.37
Vaud	1.09	0.73–1.61	0.68	1.10	0.74–1.65	0.63
Zurich	0.83	0.57–1.20	0.31	0.81	0.56–1.18	0.28
Other	Reference	Reference	Reference	Reference	Reference	Reference
Relationship status
Single	0.95	0.70–1.27	0.71
In a stable relationship	0.91	0.71–1.16	0.44
Married	Reference	Reference	Reference
Divorced	1.64	1.03–2.61	0.036*
Widowed/other	1.15	0.57–2.30	0.70
Partner working as a healthcare professional
Yes	Reference	Reference	Reference
No	0.91	0.72–1.14	0.40
Not applicable	1.06	0.74–1.51	0.76
Children
Yes	Reference	Reference	Reference	Reference	Reference	Reference
No	0.8	0.65–0.99	0.036*	0.77	0.62–0.96	0.021*

^aVariables removed at univariate stage: age, gender, employment, working hours per week, canton collapsed by population size, and partner working as a healthcare professional; variables removed during backward stepwise regression: language, experience, and relationship status

HEMS Helicopter emergency medical services, ICU Intensive care unit, BLS Basic life support, EMT Emergency medical technician

Levels of significance: * for p <0.05; ** for p <0.01; *** for p <0.001

Secondary outcomes

Patient Health Questionnaire-9

The median score for the PHQ-9 was 4.0 [quartiles 2.0–7.0]. Most participants exhibited no depressive symptoms or only mild depressive symptoms. However, 199/1485 (13.4%) participants presented moderate to severe symptoms (Fig. 3).

Fig. 3.

Classification of participants by severity of depressive symptoms, as assessed using the Patient Health Questionnaire-9

Turnover intention

Figure 4 displays the distribution of turnover intention frequency among participants, highlighting the range from never having considered leaving to weekly considerations

Fig. 4.

Turnover intention frequency distribution. The right-hand pie chart shows the distribution of turnover intention frequencies among respondents classified in the green segment on the left (those reporting at least one thought of leaving the profession)

Suicidal thoughts

Suicidal thoughts were reported by 10.2% (152/1485) of participants, with 46.7% (71/152) of those indicating such thoughts had occurred within the past 12 months.

Subgroup post hoc analysis

The canton of Neuchâtel stood out from other groups, showing a 2.3-fold increase in the OR of meeting at least one CBI burnout criterion. Consequently, a subgroup analysis comparing Neuchâtel to all other cantons combined was conducted. Table 4 presents socio-demographic variables, and Table 5 summarizes the findings of this analysis.

Table 4.

Subgroup analysis of socio-demographics variables

Variable	Neuchâtel (n=62)	Other cantons (n=1423)	p-value
Age	34.0 [29.0–43.5]	38.0 [33.0–46.0]	0.12
Gender			0.41
Male	41 (66.1)	846 (59.5)
Female	21 (33.9)	575 (40.4)
Non-Binary	0 (0.0)	2 (0.1)
Language			<0.001
French	62 (100.0)	428 (30.1)
German	0 (0.0)	877 (61.6)
Italian	0 (0.0)	118 (8.3)
Employment in %	100.0 [80.0–100.0]	100.0 [80.0–100.0]	0.20
Working hours per week	44.0 [40.0–44.0]	42.0 [36.0–45.0]	0.002
Night shifts per month	5.0 [3.0–7.0]	6.0 [4.0–7.0]	0.82
Experience	10.0 [4.0–20.0]	12.0 [6.0–20.0]	0.38
Title			0.002
BLS Driver/volunteer	1 (1.6)	21 (1.5)
EMT	8 (12.9)	95 (6.7)
Paramedic	44 (71.0)	1239 (87.1)
Other	9 (14.5)	68 (4.8)
Relationship status			0.19
Single	15 (24.2)	259 (18.2)
In a stable relationship	16 (25.8)	457 (32.1)
Married	23 (37.1)	598 (42.0)
Divorced	7 (11.3)	75 (5.3)
Widowed/other	1 (1.6)	34 (2.4)
Partner working as a healthcare professional			0.41
Yes	19 (30.6)	552 (38.8)
No	36 (58.1)	713 (50.1)
Not applicable	7 (11.3)	158 (11.1)
Children			0.90
Yes	32 (51.6)	723 (50.8)
No	30 (48.4)	700 (49.2)

Data expressed as median [quartiles] or n (%)

BLS Basic life support, EMT Emergency medical technician

Table 5.

Subgroup analysis for the Copenhagen Burnout Inventory and the Maslach Burnout Inventory

Outcome	Neuchâtel (n=62)	Other cantons (n=1423)	p-value
Meets at least one criterion on MBI	35 (56.5)	609 (42.8)	0.034
Emotional exhaustion	16.5 [10.0–23.3]	12.0 [7.0–18.0]	0.021
Total meeting criteria (≥27)	9 (14.5)	128 (9.0)	0.14
Depersonalization	8.5 [3.0–12.3]	6.0 [3.0–10.0]	0.09
Total meeting criteria (≥10)	27 (43.6)	398 (28.0)	0.008
Personal accomplishment	35.0 [30.8–40.0]	38.0 [34.0–42.0]	0.045
Total meeting criteria (<34)	22 (35.5)	323 (22.7)	0.020
Meets at least one criterion on CBI	35 (56.5)	507 (35.6)	<0.001
Personal burnout	43.8 [33.3–58.3]	37.5 [25.0–50.0]	0.010
Total meeting criteria (≥50)	25 (40.3)	419 (29.4)	0.07
Work-related burnout	46.4 [32.1–54.5]	32.1 [21.4–42.9]	<0.001
Total meeting criteria (≥50)	28 (45.2)	255 (17.9)	<0.001
Client-related burnout	37.5 [19.8–54.2]	20.8 [8.3–37.5]	0.002
Total meeting criteria (≥50)	17 (27.4)	178 (12.5)	<0.001

Data expressed as median [quartiles] or n (%)

BLS Basic life support, EMT Emergency medical technician

Discussion

Burnout prevalence and predictors: main findings

This study, the first of its kind in Switzerland, assessed the prevalence of occupational burnout among EMS providers. With a response rate of 40.5%, it demonstrated that mental health issues are a significant national concern, with 43.4% and 36.5% of respondents meeting at least one criterion for burnout based on the MBI and CBI, respectively. Although these differences could potentially reflect inconsistencies in the data, our interpretation is that they more likely highlight the multidimensional nature of burnout. Using both tools therefore provides a more nuanced understanding of how various facets of burnout manifest among EMS professionals. Multivariable logistic regression identified several predictors of burnout. A higher number of night shifts per month was consistently associated with increased risk across both MBI and CBI. According to the MBI, increasing age and female gender were independently associated with a reduced risk of burnout. According to the CBI, the canton of employment was independently associated with burnout risk, while being childless and holding a professional title classified as “Other” were protective. These findings underscore the complex interplay of occupational and demographic factors in the development of burnout among EMS providers and raise important questions about potential protective mechanisms within this professional group.

Night shifts

Consistent with the findings of Kansoun et al. [44], our results indicate that working night shifts is a significant risk factor, with a dose-dependent effect. According to both the MBI and the CBI, this result paves the way to a discussion as to how shift repartition should occur, as EMS need to be staffed 24/7. Night shifts disrupt normal sleep patterns [45–47], lead to chronic fatigue [48], impact quality of life [49], and intensify stress [50, 51]. The cumulative toll of these shifts can erode resilience over time [52], increasing physical and mental health vulnerability. A cyclical pattern, where inadequate recovery from night shift-related stress exacerbates burnout and undermines overall wellbeing, warrants particular attention as to how shifts are distributed and how post-shift recuperation needs to be planned.

Age

The inverse relationship observed between age and burnout risk in our cohort, as assessed using the MBI, likely suggests a survivorship bias, whereas no significant association was found when using the CBI. Indeed, only the most resilient EMS providers remain in the profession over time. This finding is consistent with previous research reporting a lower turnover intention in older participants [53], a negative correlation between age and burnout [54], and with studies identifying younger age as a prominent risk factor for burnout among nurses [55] and physicians [11]. Several factors may explain this association. Older EMS providers are more likely to possess advanced coping mechanisms, greater job stability, and improved adaptation to occupational stressors. Over time, these individuals may also benefit from increased professional recognition and autonomy, both of which have been shown to mitigate burnout risk [56, 57]. Furthermore, experience may foster a more realistic appraisal of job demands and expectations, reducing the emotional toll of high-stress situations. However, while burnout symptoms such as emotional exhaustion may decline with age, prior studies have suggested a potential increase in depersonalization among older professionals [57]. This shift may reflect a protective distancing mechanism or changes in professional engagement, and raises questions about how aging affects interpersonal aspects of care. Notably, older individuals often report a stronger sense of personal accomplishment, which may serve as a buffer against the negative effects of depersonalization. Overall, these findings suggest that while older EMS providers may be less emotionally exhausted, the nature of their engagement with work may evolve.

Gender

Female gender was associated with lower burnout risk according to the MBI, a finding that remained significant even after controlling for potential confounders such as workload (employment and weekly working hours), and night shifts; no significant association was observed using the CBI. Notably, the exclusion of gender from regression models resulted in a reduction in explained variance (R²), highlighting its significance as predictor. Previous studies have had mixed conclusions regarding the impact of gender on burnout: while some studies have also suggested a protective effect [58], others have reported the opposite [59–64], with a third category reported no consistent gender-based differences [65, 66]. One possible explanation lies in gender-based differences in coping strategies, role perception, and expectations within the EMS profession. It is also possible that female EMS providers who remain in the field over time represent a particular resilient subgroup, or that they may benefit from different support networks or interpersonal dynamics in the workplace, demonstrating a form of survivorship bias.

Family configuration

Being childless was associated with a reduced burnout risk according to the CBI, while no significant association was observed with the MBI. This result may be explained by the additional strain involved in balancing professional responsibilities with family obligations, particularly in high-demand occupations such as EMS. Work-family conflict has been consistently associated with higher burnout rates, reduced vitality, and increased turnover intentions among healthcare workers with children [67]. Contrasting with a previous study whose discussion puts forward the hypothesis that children helped focus the attention away from rumination, and created emotional buffer [11], this finding is in line with Ong et al. who found that female nurses with parental responsibilities often experience heightened role conflict, with caregiving duties frequently taking precedence over self-care and recovery time [68]. However, the negative impact of work-family conflict could potentially be mitigated with the presence of supportive organizational structures, providing opportunities such as flexible scheduling, child care assistance, or family-friendly policies [68, 69],.

Professional title

Contrary to previous research, our findings did not demonstrate that holding paramedic certification was associated with an increased risk of meeting burnout criteria. This diverges from the results of Crowe et al., who reported that paramedic certification significantly elevated the odds of burnout across all domains of the CBI [14], and from Bentley et al., who observed higher odds of depression and anxiety among paramedics compared to EMTs [70]. In the Swiss context, certification levels within EMS, such as EMTs, paramedics, and BLS-level providers, are linked to differing scopes of practice and responsibility. However, international variations in the role, education, and operational duties of paramedics may explain the discrepancy between our results and those from other healthcare systems. Organizational characteristics, including workload distribution, types of missions, responsibilities or the availability of institutional mental health support, may also contribute to this difference. Additionally, the composition of our sample may have limited our ability to detect significant differences as most of the respondents were paramedics. Whilst differing in distribution from other studies, this is a good reflection of the current makeup of Swiss EMS providers [34]. Unexpectedly, multivariable regression analysis based on the CBI indicated that participants in the “Other professional title” category had lower odds of burnout, whereas no significant association was observed with the MBI. This heterogeneous group of 77 participants included emergency medical dispatchers, individuals in managerial roles, and in-training paramedics or EMTs. The diversity within this category limits interpretability, but it suggests that certain non-frontline roles may be associated with reduced exposure to stressors typically encountered in clinical EMS practice. Prior studies have established that frontline EMS roles involving high call volumes, traumatic exposure, and violence are strongly associated with burnout risk [8, 10, 14, 71]. It is plausible that some roles within the “Other” category, such as management or educational positions, have reduced exposure to such conditions, thereby offering a degree of protection. Similar findings have been reported among emergency nurses, where non-clinical responsibilities have been linked to lower burnout rates [9].

The inclusion of in-training EMS providers within this category introduces further complexity. Although such trainees often participate in frontline duties depending on their training stage and cantonal regulation, their burnout profiles may differ from those of qualified staff. While a systematic review suggested that EMS trainees are at higher risk of mental health issues than certified providers [72], our study did not specifically target this population, which may partially explain the discordance. Emergency medical dispatchers also formed a small subset of this category. While they do not engage in direct patient care, they encounter unique operational stressors. Boland et al. identified dispatchers as being at elevated risk for burnout compared to paramedics [8], yet our findings did not support this trend—likely due to the limited number of dispatchers included in our sample.

In summary, although the “Other professional title” category appeared protective against burnout, interpretation of this finding is constrained by sample heterogeneity and limited subgroup size.

Canton of work

According to the CBI, the canton of employment was independently associated with burnout risk, whereas no significant association was observed with the MBI. Despite national standards for training and qualifications required to practice, this result reflects the limitations of Switzerland’s decentralised healthcare system [73], as demonstrated by Heymann et al. [11]. Each canton autonomously regulates healthcare delivery, establishes its own health policy priorities, and manages resources, political governance and organizational structures [74–77]. These factors directly shape the working conditions of EMS providers. Given this decentralisation, the present study is best interpreted as a multicentric analysis involving 27 distinct regions (26 Swiss cantons and Liechtenstein), rather than a singular national perspective. The heterogeneity across cantons influences several aspects of the EMS occupational environment, including job characteristics (e.g., emergency response vs. patient transfer), workload, frequency of night shifts, availability of administrative support, and scope of professional responsibilities. Importantly, such variation may exist not only across cantons but also within them, between different EMS organizations. In order to identify more precisely the reasons for higher risk factors in a specific canton, it would be necessary to conduct inter-organizations research within the same canton. As such, these findings are consistent with prior research identifying work organization, job characteristics, and resource availability as key determinants of burnout among emergency care workers [9, 14].

The subgroup analysis supports this interpretation by demonstrating significant differences between working conditions. The fact that nearly all domains of the CBI and MBI were affected further strengthens the potential negative consequences of regionalization for wellbeing. On the basis of these results, targeted qualitative and quantitative studies are needed to understand the differences in working conditions between the cantons, and their role in the wellbeing of Swiss EMS providers. As a solution, implementing a national collective labour agreement could help address many of these discrepancies, providing more uniform working conditions and more equitable support, potentially mitigating burnout risk for EMS providers.

Depressive symptoms

Approximately 13.4% of respondents reported moderate to severe depressive symptoms. However, these results should be interpreted with caution. Although depressive symptoms were measured using a validated and widely used scale, the determining factors of exposure and protection were not studied. In comparison, 27.2% of paramedics and EMS volunteers in Australia and New Zealand were found to experience high or very high psychological distress [78], and prevalence of depression in EMS providers ranged from 6.4% to 42.9% in a recent systematic review [79]. In this study, Wagner et al. highlighted that there is a lack of in-depth data regarding how depression correlates to work, specifically to trauma-related mental health disorders. Furthermore, depressive symptoms among paramedics may also be influenced by factors external to work, related for instance to disease-related intake of medication or daily sleep duration, which were not assessed in our study [80]. Considering these limitations, emergency care workers are known to be vulnerable to psychological disorders [81, 82]. In particular, they are exposed to chronic occupational stressors such as traumatic incidents, irregular and prolonged shift work, and high job demands [11, 83]. While not the focus of this study, the lasting effects of the COVID-19 pandemic may have amplified psychological stress among EMS providers [84, 85], a trend also observed in paramedic students, where elevated distress and reduced academic performance were reported [86]. Organizational support, from supervisors and peers, has been shown to play a protective role, mitigating symptoms of depression, anxiety, and stress, and influencing in turn the intention to leave the profession [87]. A better interpretation and generalizability of our findings require qualitative research to determine how depressive symptoms are associated with work conditions, and investigating the local policies in place.

Turnover intention

Burnout contributes to increased sickness-related absences and elevated turnover intentions [88]. In our study, a majority of participants reported having considered leaving the profession, with over half doing so at least annually. These findings align with Crowe et al., who demonstrated a strong association between burnout and turnover intention [14]. Moreover, the expressed intention is a predictor of actual attrition, with 19.6% of those expressing a desire to leave ultimately doing so [89]. Despite this, the association between burnout and turnover intention remains underexplored in EMS providers, and is likely influenced by complex, multidimensional factors [88]. Additional contributors identified in the literature include limited career development opportunities and unfavourable work schedules [90], individual psychological traits [15], and reduced job satisfaction [87, 91]. In Switzerland in 2023, EMS organizations reported a shortfall of 181 professionals [34] and a 2022 survey found that 43% of EMS providers anticipated leaving the profession within the next decade [90]. Longitudinal and qualitative research are required to elucidate the mechanisms underlying turnover intention as well as to interpret short- and long-term attractiveness of the paramedic profession.

Suicidal thoughts

One out of ten respondents reported having considered suicide during their career, underscoring the presence of suicidal ideation within this professional group. In 2017, a Swiss Health Survey reported a 7.8% prevalence of suicidal thoughts within the previous two weeks in the general population [92]. Methodological differences between both studies do not allow for direct comparison, and, hence, interpretation of the findings is limited. Nevertheless, it has been demonstrated that emergency care workers suffered higher rates of psychiatric disorders, including suicidal ideation [11, 93–95]. In a US survey, 37% of EMS providers reported having contemplated suicide, and 6.6% had attempted it - figures significantly exceeding those of the general population [96]. Similar data from Arizona [97], and from Martin et al. [98], emphasized the role of depression and PTSD symptom severity as key predictors of suicidality among EMS professionals. The proportion of EMS providers reporting suicidal thoughts in our study is less alarming, and the relationship between suicidal thoughts among Swiss EMS providers and their occupational activity remain unclear, as they were not examined. Nevertheless, considering the worrying signals in other countries, early prevention and interventions are required in relation to the known risk factors among EMS providers. Among them, shift work disorder might be a specific target, as poor sleep quality is pervasive among EMS providers [99] and sleep disorder has been shown to be closely linked with suicidal ideation [92].

Strengths and limitations

This study has several limitations that should be considered when interpreting the results. First, although the survey was distributed uniformly across Cantons, the number of respondents varied beyond what would be expected from population size alone, resulting in an overrepresentation of some smaller Cantons and an underrepresentation of larger ones. Additionally, because participants were nested within cantons, the assumption of independent observations may not fully hold, and unmeasured canton-level factors may have influenced response patterns. Although unequal group sizes limited the usefulness of intraclass correlation or multilevel analyses, future research could further explore these contextual differences. Relatedly, because participation relied on voluntary response through multiple channels and no comprehensive national roster of EMS providers was available, the study employed a non-probability convenience sampling strategy. This may limit the generalizability of the findings and introduce selection bias. Second, an inclusion bias is possible, as only active EMS providers were contactable. This survivorship bias may have led to an underestimation of burnout and depressive symptoms, and may explain the inverse association between experience and burnout. Furthermore, the survey did not capture whether respondents were undergoing any form of treatment (e.g., therapy, medication, or organizational support), limiting insights into how ongoing care might influence reported outcomes. Although the multivariable models used were statistically significant, they explained only a small portion of the variance in burnout scores (low R²), reflecting the multifactorial nature of burnout. All but the night shift variables were associated with only one of the two burnout scales, which assess different dimensions of burnout. The CBI may potentially overestimate burnout prevalence [100]; however, including both instruments offers a broader perspective and allows for the identification of different patterns, which mitigates this concern. Key variables known to influence wellbeing, such as physical activity [101], sleep quality, dietary behaviours [102], and access to social or therapeutic resources, were not measured. This highlights the importance of future research incorporating mixed-method approaches, including qualitative analysis of the open-ended responses collected in this study. Thematic analysis could offer novel insights into individual experiences and help identify additional predictors of burnout, informing future cohort-based studies. Finally, while self-report tools are widely used in occupational health research, they are inherently subjective and may be affected by recall or response biases.

Despite these limitations, the study presents notable strengths. The overall high national participation rate provides a good insight into the wellbeing of a substantial number of EMS providers across the country. The recruitment strategy likely contributed to strong engagement and resulted in a low dropout rate [103], supporting data completeness and reliability. While paramedics constituted the majority of the sample, the overall distribution of professional titles mirrors that of the current Swiss EMS workforce. Moreover, the use of validated PROMs, specifically the CBI, MBI and PHQ-9, bolsters the methodological rigor and accuracy of the assessment. These strengths provide a robust foundation for developing targeted mental health interventions and prevention policies to support EMS providers.

Perspectives

Targeted interventions to prevent burnout and mental health disorders among EMS providers are urgently needed and should be tailored to the specific needs of different subgroups. This study highlights the number of night shifts per month as a consistent predictor of burnout, suggesting that interventions addressing shift schedules, implementing recovery protocols, or providing targeted support to night shifts personnel could be particularly impactful. While individual-level strategies remain important, organizational and systemic approaches are also warranted. For instance, the observed regional differences in burnout risk indicate that structural factors, such as standardizing working conditions across EMS organizations, may help reduce disparities. Future research should further explore both organizational and individual determinants of burnout using mixed-method approaches to yield valuable, context-specific insights for developing effective and inclusive preventive strategies.

Conclusions

This study highlights the significant issue of occupational burnout among Swiss EMS providers, with approximately 40% of respondents meeting at least one burnout criterion. Night shifts frequency emerged as a consistent predictor of burnout across both the MBI and CBI models, showing a dose-dependent effect, underscoring its importance as a target for intervention. Other factors, including age, gender, professional title categorized as “Other”, canton of work, and childlessness, were associated with burnout in a scale-specific manner (MBI or CBI), reflecting the multidimensional of burnout. Additionally, participants reported depressive symptoms, turnover intention and suicidal thoughts, highlighting the need for targeted, evidence-based interventions to mitigate mental health risks in this population. Without such measures, there is a risk of cascading negative consequences affecting the healthcare system, the quality and safety of prehospital care, and the overall wellbeing of EMS providers in Switzerland.

Authors’ contributions

Conceptualization, L.S., T.S., E.P.H.; methodology, L.S., T.S., E.P.H.; software, L.S., T.S., E.P.H.; validation, L.S., T.S., E.P.H.; formal analysis, K.V.A., E.P.H., L.S.; investigation, L.S., T.S., E.P.H.; resources, L.S., T.S., E.P.H.; data curation, K.V.A., L.S., T.S., E.P.H.; writing—original draft preparation, L.S., T.S.; writing—review and editing, L.S., T.S., E.P.H., K.V.A.; visualization, L.S., T.S., K.V.A.; supervision, E.P.H.; project administration, L.S., T.S., E.P.H.; funding acquisition, L.S., T.S., E.P.H. All authors have read and agreed to the published version of the manuscript.

Funding

This study was supported by multiple sources. The Vocational Training College for Registered Paramedics and Emergency Care (ES ASUR) provided a part-time research position for Thierry Spichiger. Genève TEAM Ambulances and the Association of Prehospital Research Promotion funded the research position of Loric Stuby. The Swiss Paramedic Association awarded a grant to cover project-related expenses, including overtime compensation, publication fees, and dissemination. Additionally, the Swiss Society of Emergency and Rescue Medicine covered the costs associated with survey licensing and statistical analysis. None of the funding bodies had any role in the study’s design, data collection, analysis, interpretation, or the decision to submit the manuscript for publication.

Data availability

The dataset analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Participants provided informed consent electronically after reviewing the study purpose and procedures. Ethics approval was waived by the Geneva Regional Ethics Committee (CCER - Commission Cantonale d’Éthique de la Recherche sur l’Être Humain, Geneva, Switzerland) following consultations with other Swiss ethics commissions (Req-2024-00204). The study was conducted in accordance with the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.