Keywords
orthopedics and traumatology - surgeons - burnout - surgical Specialty
Introduction
Symptoms of depression, dissatisfaction, an imbalance between work and personal life,
exhaustion, and burnout occur throughout all stages of medical education and different
specialties.[1] The burnout syndrome was described at the beginning of the 1980s as a set of symptoms
affecting 3 mental health areas or domains: emotional exhaustion, depersonalization,
and lack of personal accomplishment.[2] Burnout increases medical errors, hinders relationships with peers, alters the work
environment, and decreases productivity. All these changes negatively affect the health
of the patients.[3]
[4]
The prevalence of burnout among orthopedic and traumatology specialists ranges from
4% to 59.4% in different studies, depending on the population analyzed.[5] In teaching environments, burnout is more frequent in residents than in clinical
specialists, students, or both.[6]
[7] Emotional exhaustion and depersonalization are the most common symptoms in the orthopedic
and traumatology, and their occurrence is also higher compared with other surgical
specialties.[5] Factors such as anxiety concerning clinical competence, the growing number of orthopedic
surgeons, financial obligations, and imbalance between work and personal life contribute
to burnout.[7]
[8]
Knowing the prevalence of burnout is critical because of the possibility of intervening
in modifiable related factors. In Chile, the burnout rate among orthopedists and traumatologists
is 53.7% per self-reporting with nonstandardized tools.[9] The present study aims to evaluate the prevalence of burnout in its domains and
associated factors in orthopedics and traumatology specialists in Chile.
Material and Methods
This is an analytical cross-sectional study. All orthopedics and traumatology specialists
registered at the 2016 Chilean Congress of Orthopedics and Traumatology were invited
to participate. We excluded general practitioners, residents, and other healthcare
professionals. We sent an online survey via institutional e-mail for anonymous responses.
A multiple-choice questionnaire assessed sociodemographic data, including age, gender,
body mass index (BMI), drug use, modafinil use, geographic region of origin, physical
activity, the weekly number of working hours, marital status, number of children,
and weekly time spent with a partner. The validated questionnaire Maslach Burnout
Inventory: Human Services Survey for Medical Personnel (MBI-HSS [MP]) was used to
evaluate the burnout syndrome.
MBI-SS Instrument
The MBI-HSS (MP) instrument includes 22 Likert-type response questions (scored from
0 to 6). It measures emotional exhaustion (EE), depersonalization (DP), and personal
accomplishment (PA), classifying each domain as low, medium, or high per score. The
diagnosis of burnout requires an EE score > 26 points (9 questions, maximum score
of 54), a DP score > 9 points (5 questions, maximum score of 30), or a PA score < 34
points (eight questions, maximum score of 48) ([Table 1]).
Table 1
|
Domain
|
Low
|
Medium
|
High
|
|
Emotional exhaustion
|
0–18
|
19–26
|
27–54
|
|
Depersonalization
|
0–5
|
6–9
|
10–30
|
|
Personal accomplishment
|
0–33
|
34–39
|
40–48
|
Statistical analysis
The description of numerical variables used central tendency and dispersion with median
and range values. Categorical variables were described as relative and absolute frequencies
and were analyzed using the chi-squared test or the Fisher exact test, as appropriate.
The Mann-Whitney and Kruskal-Wallis tests determined the association between numerical
and categorical variables. Statistical significance was set as a p-value < 0.05. Statistical
analysis was performed with STATA version 14 software (StataCorp LLC, College Station,
TX, United States).
The Scientific Ethics Committee of the Faculdade de Medicina of the Pontificia Universidad
Católica de Chile (no. 16-226) and the director board of the Chilean Society of Orthopedics
and Traumatology approved the present study.
Results
General features
The survey was filled out by 99 traumatologists. The median age was of 45 (range:
29–76) years, and 92% (n = 85) of the participants were males. All (100%) were graduated specialists, actively
working, with a median of 50 (range: 11–80) working hours per week. Among the participants,
58% were overweight and 11% were obese. The prevalence of overweight was significantly
different according to gender (61.5% of overweight males versus 12.5% of overweight
females; p < 0.05). Eighty-five percent stated they were married; 7% were separated or divorced,
and 92% (91) had ≥ 1 child. In total, 63% performed physical activity regularly, with
significant differences according to age (median ages for inactive and active subjects
were 50 and 44 years, respectively; p = 0.006) ([Table 2])
Table 2
|
Variable
|
n (%)
|
|
Gender
|
|
|
Female
|
8 (8%)
|
|
Male
|
91 (92%)
|
|
Age in years: median (range)
|
45 (29–76)
|
|
Body mass index
|
|
|
Normal
|
31 (31%)
|
|
Overweight
|
57 (58%)
|
|
Obesity
|
11 (11%)
|
|
Marital status
|
|
|
Single
|
8 (8%)
|
|
Married
|
84 (84%)
|
|
Divorced/separated
|
7 (7%)
|
|
Children
|
|
|
Yes
|
91 (92%)
|
|
No
|
8 (8%)
|
|
Weekly working hours: median (range)
|
50 (11–80)
|
|
Daily hours of sleep: median (range)
|
6 (3–8)
|
|
Modafinil use
|
|
|
Yes
|
3 (3%)
|
|
No
|
96 (97%)
|
|
Physical activity
|
|
|
Yes
|
62 (63%)
|
|
No
|
37 (37%)
|
Burnout and associated variables
We categorized the proportion of burnout in its three domains according to the cutoff
score described. In total, 21% presented a high EE score, 20% had a high DP score,
and 6% reported a low PA score ([Figure 1]). Burnout was present in 35% of the total sample (alteration in at least 1 domain),
and 6% had abnormalities in all 3 domains. The variables significantly associated
with burnout were lack of sleep (56% of burnout in those sleeping < 5 hours a day
versus 30% in those sleeping more; p = 0.03), and modafinil use (100% in those who use it versus 32% in those who did
not use it; p = 0.04). There was no association of burnout with the other variables studied ([Table 3]).
Table 3
|
Variable
|
Burnout rate (%)*
|
Subjects (n)
|
p-value
|
|
Gender
|
Female
|
25
|
2
|
0.71
|
|
Male
|
35
|
32
|
|
Children
|
No
|
62
|
5
|
0.12
|
|
Yes
|
32
|
29
|
|
Daily hours of sleep
|
> 5
|
30
|
24
|
0.03
|
|
≤ 5
|
56
|
10
|
|
Weekly working hours
|
≤ 44
|
23
|
7
|
0.13
|
|
> 44
|
39
|
27
|
|
Modafinil use
|
No
|
32
|
31
|
0.04
|
|
Yes
|
100
|
3
|
|
Physical activity
|
No
|
35
|
13
|
0.89
|
|
Yes
|
33
|
21
|
|
Body mass index
|
Normal
|
23
|
7
|
0.07
|
|
Overweight
|
44
|
25
|
|
Obesity
|
18
|
2
|
|
Marital status
|
Single
|
50
|
4
|
0.61
|
|
Married
|
33
|
28
|
|
Separated/ Divorced
|
29
|
2
|
Fig. 1
Emotional exhaustion
Overall, 21% (21 subjects) of the sample presented a high EE score, with no differences
according to age or gender. There was an association with the hours of sleep, with
a four-fold chance of EE in those sleeping ≤ 5 hours (odds ratio [OR]: 4.2; 95% confidence
interval [95%CI]: 1.3–12.6). The time spent with a partner was associated with a low
EE score (median: 2 [range: 1–4] hours in those with EE compared to 3 [range: 0–5]
hours in those without EE; p = 0.02). Finally, there was an association with modafinil use, with a 100% prevalence
of EE in those who consumed it versus 19% in those who did not (p < 0.01).
Depersonalization
The DP domain of burnout was present in 20% of the sample, with a higher prevalence
in younger traumatologists (median age: 39 [range: 29–63] years versus 47 [range:
31–76] years; p = 0.02). On the other hand, specialists with children had a 79% lower chance of suffering
from DP, and maternity or paternity was a protective factor (OR: 0.21 [95%CI: 0.04–0.9]).
Personal accomplishment
Low PA scores were observed in 6% of the traumatologists, while 18% had a medium PA
score. A higher number of working hours was associated with low PA (median: 57.5 [range:
50–70] hours versus 50 [range: 29–63] hours; p = 0.03). There were no differences in gender, age, or the remaining variables.
Discussion
In recent years, the medical literature reported high rates of burnout.[10] In our study, 35% of the sample presented burnout (alteration in at least 1 of 3
domains). Shanafelt et al.[4] reported a burnout rate of 45.8% in United States physicians in 2012, a percentage
that increased to 54.5% in 2014. For Orthopedics and Traumatology, the burnout prevalence
varies according to the population, the study methodology, and the measurement tools
used. Regarding its characteristic symptoms, the reported prevalence ranges from 16.2%
to 50.7% for EE, 11.4% to 59.4% for DP, and 10% to 33.3% for PA.[5]
[6]
Different variables are associated with burnout. In the present study, we found that
modafinil use and a decrease in sleeping hours were the main factors for altering
at least one domain. It has been reported that sleep deprivation[7] and long working hours,[11] both for residents and specialists, are associated with burnout. These variables
had the same association with EE and PA in the present study. On the other hand, three
specialists reported using modafinil, and 100% had burnout. A study carried out in
Chile with medical students showed that the use of this drug also predisposed to a
worse quality of life.[12]
In contrast, burnout prevalence is higher in women, either orthopedic surgeons or
from other surgical specialties.[13]
[14]
[15]
[16] However, our study found no significant differences according to gender. It is important
to consider that only 8% of the sample (8 subjects) were female, so there is probably
a lack of sample size to associate gender with burnout and other variables (the post
hoc analysis yielded a low value, of 6.6%, for this association).
As for protective factors, time spent with a partner and parenthood were relevant
in the literature.[8]
[17] We found that they were independently associated with EE and DP, respectively, reflecting
the importance of social factors for welfare protection.
One of the main limitations of the present study is the response rate for this questionnaire,
of 30% (99 responses). This rate could induce a measurement bias, over or underestimating
the proportion of burnout. In contrast, when measuring variables such as physical
activity and BMI, we did not use standardized tools but open questions (Do you practice any sport or physical activity? What is your height? What is your
weight?), maybe resulting in information bias.
On the other hand, the main strength of the study is the completion of the MBI-HSS
(MP) questionnaire for burnout evaluation. This questionnaire has been validated in
different populations and contexts in Chile. It is widely used today in the medical
field.[18]
[19]
[20]
[21]
[22] Previous publications in Chile were based on the report of burnout from independent
questions[9] (frequency of burnout sensation and insensitivity with third parties). The use of
an unvalidated tool can lead to erroneous conclusions in the measurement of a phenomenon.
Burnout prevention and treatment remain a challenge for the medical community. Although
recent studies have described effective therapies, the focus must continue on preventive
strategies, especially risk-increasing modifiable factors and the promotion of protective
factors, and the early detection of the condition. We need to promote and implement
these strategies at individual and institutional levels.
Conclusion
The present study found a burnout prevalence of 35%, defined as the alteration of
at least 1 domain (EE, DP, and PA) of the MBI-HSS (MP) questionnaire. Sleep deprivation
and modafinil use were significantly associated with an increased risk of burnout.
Future studies should investigate prevention based on modifiable risk factors.
