Prevalence of Burnout Among Surgical Residents and Surgeons in SwitzerlandBurnout in Swiss Surgical Residents and Surgeons

In the past decade, work-related stress and burnout have received growing publicity in the scientific literature.^{1 - 7} The well-being of health care professionals has been spotlighted, as they are increasingly dissatisfied with their professional lives.^{8 - 10} There are numerous work-related risk factors for burnout,^{2 ,11 - 12} and studies found that psychosocial work-related risk factors rather than personal factors are associated with burnout in physicians^{13 - 16} and in nurses.^{17 - 18}

According to Maslach et al,¹⁹ burnout represents a pathologic response to long-term occupational stressors and has the following 3 dimensions: emotional exhaustion, depersonalization, and diminished personal accomplishment. Burnout is specifically related to the context of work and has been linked to decreased productivity, lower professional satisfaction, increased amotivational behavior, and reduced commitment toward job or organization.²⁰ In contrast, major depression is not just work related and usually affects all domains of a person's life.

The aim of this study was 2-fold: first, to assess the prevalence and degree of burnout among surgical residents and surgeons in Switzerland and, second, to identify predictors of burnout in the surgical community.

In 2005, a self-administered questionnaire was sent to 281 surgical residents and to 337 surgeons. Questionnaires included demographics, job characteristics, the 22-item Maslach Burnout Inventory,²¹ and an 18-item list of potential work-related stressors. The cantonal ethical committee of Basel, Switzerland, stated that this study did not require approval. Each domain of the Maslach Burnout Inventory received a score that was classified as low, medium, or high (Table 1) based on a study²¹ of 1104 medical professionals. According to a manual by Maslach et al,²¹ we defined respondents with high scores on both the emotional exhaustion and the depersonalization subscales and with a low score on the personal accomplishment subscale as participants with a high degree of burnout and respondents with a high score on either the emotional exhaustion or the depersonalization subscale as participants with a moderate degree of burnout.

Potential professional stressors assessed in the survey were identified by reviewing the literature.^{22 - 24} Each stressor was rated on a 5-point Likert-type scale ranging from 1 (not at all) to 5 (extremely) regarding how much it contributed to experienced stress.

Mann-Whitney test was used for comparison of subgroups for continuous end points and Fisher exact test or χ² test for comparison of dichotomous or categorical end points. Predictors of burnout were identified by logistic regression analysis, and variables that were significant in univariate analyses were then included in the multivariable model. To simplify interpretation of the multivariable analyses, we dichotomized answers to the contribution of the stressors (1-3 vs 4 or 5 on the Likert-type scale). P < .05 was considered significant.

Four hundred five of 618 anonymous questionnaires (65.5%) were returned. Table 2 summarizes the respondents' characteristics.

A high degree of burnout was found in 3.7% of respondents and a moderate degree in 35.1%. The burnout rates and subscale findings among physicians in different practice settings (university vs regional hospitals), as well as with different professional status (residents vs surgeons), were not significantly different. Respondents with high and moderate degrees of burnout had higher summary scores of perceived stress (P < .001), and respondents with a high degree of burnout were less likely to live with a partner (P = .004) and worked more hours per week (P = .03).

In multiple logistic regression analysis, the strongest predictors of moderate burnout were poor interaction with nurses and disturbances due to telephone consultations, and the strongest predictors of high burnout were poor interaction with nurses and high overall workload (Table 3). The most frequently reported source of perceived excessive stress (≥4 on the Likert-type scale) was high administrative workload (256 of 405 participants).

In this study, 3.7% and 35.1% of respondents had high and moderate degrees of burnout, respectively. In multiple logistic regression analysis, the strongest predictors of burnout were poor interaction with nurses, disturbances due to telephone consultations, and high overall workload.

The prevalence of burnout among physicians is surprisingly high.^{14 ,22 ,26} Ramirez et al²⁵ found high exhaustion and depersonalization scores among 27% and 19% of hospital consultants, respectively, and a low personal accomplishment score among 32%. In an investigation among primary care practitioners, Goehring et al²² identified high exhaustion and depersonalization in 19% and 22%, respectively. The most notable difference relative to our findings was the percentage with low scores for personal accomplishment (16% among primary care practitioners vs 41.2% among surgeons in our study). Thirty-two percent and 3.5% of respondents had moderate and high degrees of burnout, respectively, in their study.²²

The finding that 41.2% of our respondents rated themselves as having poor competence is surprisingly high because surgeons are perceived as being efficiency oriented.²⁷ No significant difference was found in this regard between residents and surgeons. The high percentage of depersonalization is alarming because high scores are associated with self-reported suboptimal patient care practices among residents.² In contrast to McMurray et al²⁸ and Embriaco et al,¹⁴ who showed that women were 1.6 and 1.58 times, respectively, more likely than men to have a higher Maslach Burnout Inventory score, or to Goehring et al,²² who showed that midcareer male physicians have a 1.4-fold higher risk of reporting a moderate degree of burnout than their female counterparts, no demographic factors were implicated by the model in our study or by other authors.^{29 - 30}

We acknowledge several limitations of this study. First, the cross-sectional design precludes evaluation of burnout over time. Second, a nonresponse bias cannot be excluded; individuals with higher degree of burnout may be less willing to extend the effort needed to participate in the study. However, this would be associated with an even higher prevalence of burnout among surgeons. The participation rate of 65.5% was good compared with that in other studies,^{31 - 32} indicating that respondents were willing to express their emotions.

Faced with workforce shortages, an inflexible lifestyle, and high professional demands,³³ the surgical arena will have greater difficulty in attracting residents and retaining surgeons unless stress factors are eliminated and external resources are constructed to address and eliminate burnout. The development of new work models should be sought to decrease burnout, including reduced work intensity, stressful interaction with nurses, interrupting telephone consultations, and workload rather than work hours alone.³⁰

Correspondence: Ulrich Guller, MD, MHS, Division of Visceral Surgery and Transplantation, Department of Surgery, University Hospital Bern, CH-3010 Bern, Switzerland (ulrich.guller@gmail.com).

Accepted for Publication: June 13, 2010.

Author Contributions: Dr Businger had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Businger and Guller. Acquisition of data: Businger. Analysis and interpretation of data: Businger and Guller. Drafting of the manuscript: Businger and Guller. Critical revision of the manuscript for important intellectual content: Businger and Guller. Statistical analysis: Businger and Stefenelli. Administrative, technical and material support: Businger. Study supervision: Guller.

Financial Disclosure: None reported.

Previous Presentation: This study was presented at the 94th Annual Congress of the Swiss Society of Surgery; June 13, 2007; Lausanne, Switzerland.

Additional Contributions: Daniel Oertli, MD, provided technical and material support; Bernadette von Felten assisted with data acquisition; and Phillip Hendrickson, PhD, critically read the manuscript. We thank the residents and surgeons who took part in the survey.

The Editors welcome contributions to the Resident's Forum. Manuscripts should be submitted via our online submission and review system (http://manuscripts.archsurg.com). Manuscript criteria and information are per the Instructions for Authors for Archives of Surgery (http://archsurg.ama-assn.org/misc/ifora.dtl). A narrative abstract of no more than 135 words should be included, and the manuscript should be limited to 1000 words with no more than 3 figures. We will consider small case series, historical reviews, summaries of recent developments in surgery, and laboratory studies. Please note that we are no longer accepting single case reports.