Burnout and quality of life among orthopaedic trainees in a modern educational programme

The challenges of medical training can lead to significant personal distress.¹ Burnout, poor quality of life and job dissatisfaction have been identified among physicians during their postgraduate medical training.^2-4 Poor quality of life has been reported in 14.8% to 18% of trainees in different surveys.^4,5 Using the Maslach Burnout Inventory,⁶ symptoms of burnout have been found in 21% of Dutch trainees from different specialties who have a working week which does not exceed 48 hours,⁷ and in as many as 56% of orthopaedic trainees in the United States with an 80-hour working week.⁸ Although similarly high levels of burnout have been described among general practitioners in the United Kingdom,⁹ the rate of emotional distress among British orthopaedic surgeons and trainees has not yet been studied. Because sub-optimal physician well-being is associated with negative effects on patient care, preventing stress and burnout of trainees should be a priority in every postgraduate training programme.^2,10,11

The development of distress and symptoms of burnout can result from a high level of work-home interference and high professional and educational demands.^12,13 High workloads, long and irregular working hours¹²and organisational and educational factors such as lack of autonomy, lack of social and supervisory support have also been associated with distress and burnout in trainees.¹⁴ Over the past decade, major changes have been implemented in postgraduate training programmes in several countries, including the United Kingdom and the Netherlands, to address these problems. First, educational training programmes have been gradually modernised from a process-based model focusing on completing the time period of postgraduate training and fulfilling the numerical requirements of surgical procedures to be mastered, to an outcome-based model with increased emphasis on generic competencies in addition to medical knowledge and surgical skills.^15-18 Second, strict compliance with the European Working Time Directive (EWTD) limits the working week of Dutch orthopaedic trainees to 48 hours, compared with that of 80 hours a week for United States trainees. In the United Kingdom, the same EWTD applies, but trainees can work longer hours by signing an opt-out clause. Although concern has been raised about the negative effects of reducing the working hours of postgraduate doctors on their clinical and surgical competencies, two systematic reviews of practice in the United States suggest that reducing trainees’ working hours from over 80 hours a week does not seem to adversely affect postgraduate training and patient care in surgical and nonsurgical disciplines.^19,20 However, the impact of reducing the working week to 48 hours has not yet been sufficiently evaluated.¹⁹ Third, the modernisation of postgraduate training programmes has placed increased emphasis on the importance of stimulating a safe clinical learning environment.²¹ Earlier work on Dutch trainees has shown that the risk of burnout increases if trainees perceive little reciprocity in the relationship with their supervisors.²²Conversely, adequate supervision, room for extra operating, and evaluation of and attention to the individual competence of trainees throughout their traineeship are positively influencing both trainee well-being and patient outcomes.²⁰

The aim of this study was to evaluate the emotional well-being of orthopaedic trainees in a modern competency-based training programme with strict compliance to a 48-hour working week. We hypothesised that quality of life would be better and the prevalence of burnout symptoms would be lower in trainees enrolled in such programmes than in trainees who followed a more process-based programme with an 80-hour working week, such as the orthopaedic training programme in the United States. In addition, we hypothesised that the emotional well-being of trainees would be positively related to the quality of the learning environment.

Materials and Methods

Starting in their third postgraduate year of training, and after completing their two-year general surgery rotation, all orthopaedic trainees in the Netherlands participate in an annual national compulsory course which covers the basics of orthopaedic surgery. Trainees have to participate three times in this course, in their third, fourth and fifth year of postgraduate training. During one of these courses, held in November 2011, we asked all attending trainees to complete a questionnaire which assessed their quality of life, symptoms of burnout and the quality of the clinical learning climate using validated instruments as described below. Demographic data were also recorded. The Dutch Orthopaedic Society approved this study. Because no patients were included, the study was exempt from formal ethical board review under Dutch law. Following guidelines for educational research issued by the Netherlands Society of Medical Education, anonymity was guaranteed, participation was voluntary, and informed consent was obtained.

Quality of life was evaluated using two single-item linear analogue self-assessments, which are being widely used in quality of life research and have been validated across a broad range of medical conditions.^23,24 Trainees were asked to rate their quality of life on a scale of 1 to 5, with response options ranging from “As bad as it can be” to “As good as it can be”. Satisfaction with the balance between personal and professional life was assessed on a similar five-point Likert scale.²⁵ High quality of life was defined as a response of 5 or 4 to each of these questions. Likewise, poor quality of life was defined as a response of 3, 2 or 1 to each of these questions.

To assess symptoms of burnout, we used the key items of the Maslach Burnout Inventory (MBI),⁶ which is considered to be the reference instrument for the evaluation of burnout in the medical literature.¹ Emotional exhaustion and depersonalisation are considered the cornerstone dimensions of burnout.^1,26 Therefore, we focused on items which assessed the dimensions of emotional exhaustion and depersonalisation, drawing on responses to two items from the MBI. In a large sample of more than 10 000 trainees and physicians, these two single-item measures correlated strongly with the aforementioned dimensions of burnout.²³ Emotional exhaustion was evaluated by the question, “How often do you feel burned out from your work?” and depersonalisation by the question, “How often do you feel you’ve become more callous towards people since you started working as an orthopaedic trainee?” These questions were answered on a seven-point Likert scale,^6,25 ranging from ‘never’ to ‘daily’. Symptoms of high emotional exhaustion were defined by a frequency of at least once a month on the single-item emotional exhaustion measure. Similarly, symptoms of high depersonalisation were defined by a frequency of at least once a month on the single-item depersonalisation measure.

To evaluate the quality and safety of the clinical learning environment we used the Dutch Residency Educational Climate Test (D-RECT).²⁷ The D-RECT is a validated tool to measure the quality of the learning climate and consists of 50 items on 11 subscales (e.g. feedback, coaching and assessment, supervision, patient handover and professional relations between consultants).²⁸ Respondents are asked to indicate their agreement with each item using a five-point Likert scale, ranging from totally agree (5) to totally disagree (1), or to rate the item as not applicable. High scores (4 or 5) indicate a ‘good’ clinical learning climate. Scores 3 or lower are considered a cause for concern. This questionnaire also records data on gender, age, year of training and site of training.

Statistical analysis

This was performed using Statistical Package for the Social Sciences v17.0 (SPSS Inc., Chicago, Illinois). Student t-tests were used to compare means. Categorical variables were compared using Pearson’s chi-squared test. Analysis of variance (ANOVA) was used when comparing means in more than two groups. Correlation between variables was determined using Spearman’s rank coefficient. P-values < 0.05 (all two-tailed) were considered significant.

Results

Of the 112 orthopaedic trainees attending the training day, 105 (94%) responded. There were 22 (21%) female respondents. Overall, 38 (36%) worked in an academic medical centre, 66 (63%) in an affiliated general hospital, and one trainee did not provide information about their teaching hospital. Of the trainees, 37 (35%) were in their third postgraduate year, 43 (41%) in their fourth year, and 25 (24%) in their fifth year.

The mean (standard deviation (sd), range) overall quality of life score was 4.0 (sd 0.7, 2 to 5). Poor quality of life scores (score < 3) were found in 19 trainees (18.1%). Furthermore, 49 trainees (46.7%) were somewhat or very dissatisfied with the balance between their personal and professional life (score < 3). Some symptoms of burnout were seen in 29 trainees (27.6%) and of this total of 29 trainees, 17 (16.2%) reported symptoms of emotional exhaustion, and 12 (11.4%) symptoms of depersonalisation at least once a month. Poor quality of life and symptoms of burnout were equally common in men and women and in different years of postgraduate training (Table I). Although we did not find a difference in overall quality of life, satisfaction between work-life balance and symptoms of emotional exhaustion between trainees working in affiliated general hospitals and academic medical centres, monthly symptoms of depersonalisation were reported more frequently by trainees from affiliated general hospitals (n = 11, (16.7%)) than by those from university hospitals (n = 1 (2.6%), p = 0.03) (Table I).

The mean (sd, range) total clinical learning climate score was 3.8 (sd 0.4 (2.6 to 4.6), indicating an acceptable learning climate. Total scores < 3 (cause for concern) were reported by 18 trainees (17.0%). No differences were found in clinical learning climate scores between males and females (p = 0.66), university or general hospitals (p = 0.22), or year of training (p = 0.06). Higher scores on the D-RECT (indicating a better learning climate) were associated with better quality of life (r = 0.31, p = 0.001), more work-life balance satisfaction (r = 0.31, p = 0.002), fewer symptoms of emotional exhaustion (r = -0.21, p = 0.028) and fewer symptoms of depersonalisation (r = -0,28, p = 0.04). A total of seven of the 18 trainees (39.0%) rating the learning climate quality as poor (score <  3) reported poor quality of life (score < 3), compared with 12 (14%) of the 87 trainees who gave high ratings (score >  3) to the learning climate quality (p = 0.012, chi-square test). Similarly, a significantly higher rate of depersonalisation was found in trainees who rated the quality of the learning climate as poor (n = 5, 28%) compared with trainees who rated the learning climate as good (n = 7, 8.0%) (p = 0.017, chi-square test).

The quality of the learning environment was rated significantly higher by trainees with good quality of life than by those with poor quality of life (Fig. 1, Table II), and by trainees without symptoms of depersonalisation than in those with such symptoms (Fig. 2, Table II). There were no significant differences in satisfaction with work-life balance (p = 0.177, Student’s t-test) and symptoms of emotional exhaustion (p = 0.445, Student’s t-test) between trainees who rated the quality of the learning climate as ‘good’ and trainees who rated the quality of the learning climate as ‘poor’.

Trainees’ overall quality of life and satisfaction with work-life balance was significantly correlated, with work adapted to the trainee’s competence and with the role of the supervising consultant (Table III). Additional learning factors of climate associated with overall quality of life included role of the specialty tutor, who is considered the consultant responsible for the trainees’ educational programme and patient sign-out, meaning communication, discussion and making decisions in patient care. The strongest learning climate factor associated with symptoms of burnout was poor peer collaboration (Table III). It should be noted that some of the correlations in Table III are weak, with values < 0.4.

Discussion

We found a significant number of the Dutch orthopaedic trainees reported poor quality of life and symptoms of burnout in a modern educational programme with strict compliance to a 48-hour working week. Even in a programme with such strictly reduced working hours, dissatisfaction with work-life balance was found. A better quality of the clinical learning climate was associated with better emotional well-being and fewer symptoms of burnout in orthopaedic trainees, suggesting that further improvements in the learning environment should be able to reduce the undesirably high level of burnout and poor quality of life among orthopaedic trainees.

When comparing our results with those reported in recent literature, it should be borne in mind that the prevalence of burnout symptoms varies greatly between studies, ranging from 18% to 82%.^1,12,23,28 This large variation can partly be explained by differences in work conditions, duration of the working week, specialty, culture, and response rate to the questionnaire. These differences in study methods and the large variability of burnout symptoms between studies hamper the comparison of our results to those found in other groups. Sargent et al⁸ evaluated the rate of burnout symptoms in orthopaedic trainees in the United States. They found the rate of burnout symptoms to be 56%, which considerably exceeds the 28% seen in our study. Although a range of other factors, as listed above, may also help to explain the difference in burnout symptom rates between Dutch trainees and their colleagues in the United States, the possibility of the striking difference in working hours between the two countries might be a contributory factor. Previous studies have shown a consistent relationship between lengthy working hours and sleep deprivation and low quality of life and burnout.^13,28 International comparative data are needed to further elucidate differences in burnout symptoms between trainees from different countries and with different educational programmes and working hours regulations.

We found a consistent and significant association between the clinical learning climate and the overall well-being of orthopaedic trainees. This clinical climate of learning, which constitutes the specific context in which trainees learn during their daily activities, represents a major indicator of the educational quality of a medical department that provides clinical training.²⁷ Our results suggest that improving the quality of the learning environment may contribute to an improvement in the quality of life and a reduction in the risk of burnout for trainees. Improving the climate of learning can be achieved by the appropriate integration of work and training and by tailoring the education to the learning needs of the individual trainee.²⁹ The quality of the climate of learning is of major importance because the quality of training programmes is related to patient care.²⁰ The results of our study suggest that improvements in the climate of learning should include both the role of the consultant in the organisation of the learning environment and peer collaboration among trainees, as these factors were significantly associated with overall quality of life, satisfaction with work-life balance, and symptoms of burnout, respectively (Table III).

We believe this is the first study to investigate the relationship between the clinical climate of learning and trainee well-being in a modern educational programme with strict compliance to a 48-hour working week. A strength of the study is the high response rate (94%), whereas other studies have shown response rates ranging from 27%²⁸ to 41%.⁷

We acknowledge the following limitations. First, the cross-sectional and correlational design of the study precludes inference of causation. We are now following orthopaedic trainees longitudinally to investigate whether the association between the quality of the learning environment and the prevalence of burnout symptoms and quality of life in these individuals is consistent over time, and throughout their training programme. Second, only the most important domains of well-being were assessed, meaning that our results do not provide information about depression, job satisfaction, or other factors which determine the overall well-being of trainees. We deliberately limited our survey for practical reasons, to ensure a higher response rate than have been achieved by previous physician surveys.³⁰ Further qualitative or in-depth studies are needed to provide more details about the extent and nature of poor quality of life in orthopaedic trainees in modern educational programmes with limited working hours. Thirdly, only selected demographic variables were evaluated, excluding other personal factors which may have an impact on well-being, such as socioeconomic status, marital status and children, and educational debt. A final limitation is that our study only included Dutch orthopaedic trainees, which limits the generalisation of our results to foreign training systems. Finally, it should be noted that we used only the key items from the MBI, instead of the whole instrument. Although this may have influenced the observed prevalence of burnout in our sample to some degree, it is unlikely to have had a major influence on our results as previous work has shown that responses to these key questions are a good reflection of the responses to the corresponding key dimensions of the MBI.²³ Although we acknowledge that future studies are needed to compare burnout and quality of life in orthopaedic trainees from different countries, we feel that our study provides an important insight into the well-being of trainees enrolled in a modern training programme with a strictly limited 48-hour working week.

In conclusion, we found a significant proportion of Dutch orthopaedic trainees enrolled in a modern postgraduate educational programme, with strict 48-hour workweek regulations, still reported poor quality of life and symptoms of burnout. The rate of burnout found in this study was lower than the rate of burnout found among orthopaedic trainees in the United States but we accept that they pursued a more traditional educational programme as well as being employed for an 80-hour working week. A better clinical climate of learning was associated with a better quality of life of orthopaedic trainees and fewer symptoms of burnout. This suggests that further improvements in the clinical climate of learning may help to improve the well-being of trainees and could, in the end, improve patient care.

No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

This article was primary edited by G. Scott and first proof edited by A. Ross.