Factors at medical school and work related to exhaustion among physicians in their first postgraduate year

Abstract

Burnout and stress is frequently reported in young physicians but longitudinal studies are sparse. Exhaustion is a core facet of burnout.
To study individual and environmental medical school predictors and associated working conditions of postgraduate exhaustion, with a reference to gender.
Two cohorts of junior doctors (n=253, 58% women) graduated from Karolinska Institutet were assessed in medical school (2002 and 2005) and in their first postgraduate year (2003 and 2006). Baseline measures were: Performance-based self-esteem (PBSE), study conditions (Higher Education Stress Inventory, HESI) and exhaustion, and at follow-up exhaustion (Oldenburg Burnout Inventory, OLBI) and Learning climate in the clinic. Regression analyses on postgraduate exhaustion (OLBI) were performed in four steps. First PBSE gender and age was entered, second study conditions (HESI), third working conditions (Learning climate in the clinic), and finally we controlled for exhaustion at final year of medical school.
Response rate was 73%. Worries about future endurance/capacity (WFEC; HESI) predicted postgraduate exhaustion, but not PBSE, when baseline exhaustion was controlled for. Women's higher exhaustion scores were explained by their higher WFEC. A positive Learning climate was negatively associated with exhaustion.
High WFEC was a risk factor of exhaustion to which women were more subjected. Students with high doubts of themselves may benefit from specific programmes in medical school, addressing this risk. A positive Learning climate at follow-up seemed protective, although no conclusions on direction of causality can be made. The effect of PBSE needs further study.

Full text

Factors at medical school and work related to
exhaustion among physicians in their fi rst
postgraduate year
MARIE DAHLIN , JENNY FJELL , BO RUNESON
Dahlin M, Fjell J, Runeson B. Factors at medical school and work related to exhaustion among
physicians in their fi rst postgraduate year. Nord J Psychiatry 2010;64:402–408.
Background: Burnout and stress is frequently reported in young physicians but longitudinal
studies are sparse. Exhaustion is a core facet of burnout. Aims: To study individual and envi-
ronmental medical school predictors and associated working conditions of postgraduate exhaus-
tion, with a reference to gender. Methods: Two cohorts of junior doctors ( n  253, 58% women)
graduated from Karolinska Institutet were assessed in medical school (2002 and 2005) and in
their fi rst postgraduate year (2003 and 2006). Baseline measures were: Performance-based self-
esteem (PBSE), study conditions (Higher Education Stress Inventory, HESI) and exhaustion, and
at follow-up exhaustion (Oldenburg Burnout Inventory, OLBI) and Learning climate in the
clinic. Regression analyses on postgraduate exhaustion (OLBI) were performed in four steps.
First PBSE gender and age was entered, second study conditions (HESI), third working condi-
tions (Learning climate in the clinic), and fi nally we controlled for exhaustion at fi nal year of
medical school. Results: Response rate was 73%. Worries about future endurance/capacity
(WFEC; HESI) predicted postgraduate exhaustion, but not PBSE, when baseline exhaustion was
controlled for. Women ’ s higher exhaustion scores were explained by their higher WFEC. A
positive Learning climate was negatively associated with exhaustion. Conclusions: High WFEC
was a risk factor of exhaustion to which women were more subjected. Students with high
doubts of themselves may benefi t from specifi c programmes in medical school, addressing this
risk. A positive Learning climate at follow-up seemed protective, although no conclusions on
direction of causality can be made. The effect of PBSE needs further study.
• Exhaustion , Learning climate , Medical students , Personality , Physicians , Stress , Working conditions.
Marie Dahlin, M.D., Ph.D., Department of Clinical Neuroscience, Karolinska Institutet, Division
of Psychiatry St. Goran, SE-112 81 Stockholm, Sweden, E-mail: marie.dahlin@ki.se; Accepted
7 March 2010.
B urnout is an adverse reaction specifi cally associated
with working conditions, that at fi rst was acknowl-
edged in human service occupations, e.g. health staff.
Burnout has later been shown to be present in any kind
of working population, and is foremost related to high
demands in the working situation (1). Professionals are
often affected early in their careers (2), although women
may in addition be more at risk late in working life (3).
It has been defi ned according to the three dimensions
emotional exhaustion, cynicism and diminished personal
accomplishment (1, 2), although some authors include only
the two fi rst (4). Exhaustion is often considered a core fea-
ture of the burnout reaction (1). From an individual per-
spective, it has been suggested that a Performance-based
self-esteem (PBSE) is a prerequisite for the development of
burnout, so only those individuals with a strong inner need
to achieve highly would press themselves hard enough to
react with burnout (1, 5). According to this theory, burnout
is regarded as a specifi c reaction in an engaged individual,
and somewhat different from pure exhaustion, which can
affect anyone in a demanding situation. PBSE is thus con-
sidered a trait-like factor.
Physicians might be especially at risk of work-related
distress such as exhaustion and burnout (6 – 10), and so
may medical students (11 – 13). While some studies have
suggested that physicians ’ burnout is primarily related to
pre-existing vulnerability (14 – 16), other studies have
focused on work-related factors, such as high demands,
lack of control and long working hours (17, 18). There
is a lack of prospective studies taking both individual
and organizational aspects into account (10).
Women and men may encounter different levels of
stress in working life and during the last few decades,
increasing numbers of women have entered medicine
© 2010 Informa H ealthcare DOI : 10.3109/ 08039 481003759219

PREDICTORS OF EXHAUSTION IN YOUNG PHYSICIANS
NORD J PSYCHIATRY·VOL 64·NO 6·2010 403
questionnaire. Participants signed a separate form, giving
informed consent. Responders received a cinema voucher.
Ethical approval was obtained from the Ethics com-
mittee at Karolinska Institutet.
Measures
The outcome variable was the exhaustion dimension of
burnout at follow-up, measured by the Oldenburg Burn-
out Inventory, OLBI (32). The OLBI has been validated
against the Maslach Burnout Inventory (33). The exhaus-
tion dimension contains eight statements, positively or
negatively worded. Items are rated on a 4-point Likert
scale; dimension scores are computed as means (range
1 – 4). We used the Swedish translation; for the T2 mea-
sures, a version adapted for students was used, where
“ work ” was replaced by “ studies ” on each item (29).
Cronbach ’ s α was 0.79 at T2 and 0.83 at T3.
Participants indicated whether they had a temporary
position, attended an internship programme or were
doing research. Activity status was indicated as “ cur-
rently working ” , “ on sick-leave ” or “ on parental leave ” .
Civil status was indicated as being married or cohabitant
(yes/no) and parenthood as having care of children in the
household (yes/no).
PBSE was measured at T1 in the 2005 cohort and at
T2 in the 2002 cohort. We used the PBSE scale of four
items, e.g. “ I think that I sometimes try to prove my
self-worth by being competent ” , rated on a 5-point scale
(1  “ Does not apply at all ” to 5  “ Applies perfectly ” ) (5,
34). Cronbach α was 0.79.
Study conditions were assessed by the Higher Educa-
tion Stress Inventory, HESI (35). Each item is rated on a
4-point Likert scale, 1 – 4, with reversed scoring for posi-
tively worded items (indicated below with “ r ” ). Seven
factors have been identifi ed, scores computed as means,
where high values are undesirable. Worries about future
endurance/competence (WFEC), three items (Cronbach ’ s
α  0.80), “ I worry about long working hours and respon-
sibilities in my future career ” , “ I am worried that I will
not acquire all the knowledge needed for my future pro-
fession ” and “ The insight I have had into my future
profession has made me worried about the stressful
internationally (19, 20). There is evidence that female
medical students and residents are more exposed to
harassment than their male peers (21). The effect of gen-
der on burnout and stress has yielded divergent results in
the literature (22 – 24).
Young physicians under training seem to be particu-
larly at risk of mental distress (23, 25 – 27), which is
important to address, as experiences and health during
the formative years might affect the physician and con-
sequently his/her patients throughout life (28). A readi-
ness within the hospital ward to adjust to the fact that
the young physician is under training seems to be pro-
tective against job stress in interns (23).
We have previously shown that exhaustion is associ-
ated with poorer self-rated health and performance based
self-esteem in medical students. Few studies of PBSE
have a prospective design; most data are cross-sectional.
Medical students have high levels of PBSE (29). It is
reasonable to believe that junior doctors, with a high
degree of perfectionism (30, 31), would be particularly
at risk of developing exhaustion, especially before they
have acquired appropriate coping strategies. Swedish
medical students may after graduation either enter the
nationally regulated internship programme of 18 months
directly, or choose to take a temporary position to gain
credits for the internship position of their choice.
Aims
We wanted to study predictors of exhaustion in the fi rst
postgraduate year, with a reference to gender. We hypoth-
esized that PBSE and stressful study conditions would be
risk factors of exhaustion and that a positive Learning cli-
mate at the workplace in the clinic would be protective.
An additional aim was to study the development of exhaus-
tion over time. We followed two cohorts of medical stu-
dents from one medical school from their last undergraduate
year into their fi rst postgraduate year as junior doctors.
Materials and methods
Subjects
Postal questionnaires were distributed to two cohorts
(autumn 2002, n  113 and spring 2005, n  140) of last
semester (11th) medical students (T2) at Karolinska Insti-
tutet with follow-up 6 – 10 months after graduation (T3).
Differences in intervals were related to late response or
delay of entrance into working life of the responders. Up
to three reminders were sent to non-responders. In total,
253 individuals (58% women) were approached; mean
age (  standard deviation) at T3 was 27.4  4.1 years.
The 2005 cohort had in addition been assessed at their
third year of medical school, when data on PBSE were
collected for them (T1). See Fig. 1 for data collection
and procedure details.
Written information of the study, stating confi dentiality
and that participation was voluntary, was attached to the Fig. 1. Curricular and calendar times for data collection.

M DAHLIN ET AL.
404 NORD J PSYCHIATRY·VOL 64·NO 6·2010
exhaustion scores between T3 non-responders (2.33  0.62)
and T3 responders (Table 2, P  0.360).
Demographic characteristics
Characteristics of the responding junior physicians are
given in Table 1. Approximately half of the students had
entered their mandatory internship, while the other half
held temporary positions as junior physicians. There were
no signifi cant gender differences in any of the demo-
graphic factors shown in Table 1.
Exhaustion, PBSE, study conditions and working
conditions
In Table 2, the dependent and independent variables are
displayed. There was a slight increase in exhaustion lev-
els from fi nal year of medical school to the fi rst post-
graduate year ( P  0.031, paired samples t -test). This
effect was related to an increase of exhaustion among
women (2.28  0.40 at T2 and 2.40  0.52 at T3, P  0.043),
who also scored higher than men on both occasions
(2.10  0.44 at T2, 2.16  0.51 at T3).
PBSE was 3.41  0.83 for women and 3.37  0.78 for
men, P  0.741. Women scored higher on study condi-
tions WFEC (2.91  0.74 vs. 2.30  0.71) and Non-
supportive climate (1.77  0.50 vs. 1.47  0.40) than men.
The other study conditions did not differ regarding gen-
der, nor did recorded working conditions, Weekly work-
ing hours or Climate for learning.
Bivariate associations between independent
variables and postgraduate exhaustion
Gender (1  female), PBSE, all study conditions and T2
exhaustion had signifi cant bivariate associations with
workload ” ; Non-supportive climate , fi ve items (Cronbach ’ s
α  0.72), e.g. “ The studies have created anonymity and
isolation among students ” ; Faculty shortcomings , seven
items (Cronbach ’ s α  0.62), e.g. “ The teachers often fail
to clarify the aims of studies ” , “ I feel that the training is
preparing me well for my future profession (r) ” and “ I
am able to infl uence the studies (r) ” ; Insuffi cient feed-
back, two items (Cronbach ’ s α  0.67), e.g. “ Teachers
often give feedback on students ’ knowledge and skills
(r) ” ; Low commitment , two items (Cronbach ’ s α  0.68),
e.g. “ I am satisfi ed with my choice of career ” (r). The
factors Workload and Financial concerns were omitted
because of low internal consistency.
The working conditions assessed comprised aver-
age Weekly working hours and Learning climate in the
clinic. Learning climate was measured by a 10-item
questionnaire, focusing on the working environment ’ s
degree of adaptation to the young physician ’ s need for
continuous education, feedback and support (18). State-
ments, worded either in a positive or negative direc-
tion, are rated on a 5-point Likert scale. A sum score is
computed, Cronbach ’ s α was 0.79.
Statistical analysis
We used t -tests for paired and independent samples for
means and chi-squared tests for proportions. For the pre-
diction analysis, multiple linear regressions were run,
method Enter. The independent variables were entered in
subsequent blocks as follows: 1) gender, age and PBSE
(as considered a trait); 2) bivariately signifi cant study con-
ditions; 3) bivariately signifi cant working conditions; 4)
controlling for exhaustion at fi nal year of medical school.
For each independent variable, interaction with gender
was controlled for, by introducing interaction terms
(gender  other variable) into the model, one at a time.
Results
Response rate and representativity of sample
Of the 253 students approached, 186 (73.5%; 111 women,
59.7%) returned completed questionnaires on both occa-
sions. There were no signifi cant difference regarding
gender and age (Table 1, P  0.203) distribution, nor T2
Table 1 . Demographic characteristics of interns at follow-up.
Age, mean  standard deviation
29.1  4.0
n (%)
Current position Internship programme 85 (45.7)
Temporary position 95 (51.1)
Research 5 (2.7)
On sick-leave 3 (1.6)
On parental leave 4 (2.2)
Cohabiting/married 105 (57.1)
Children in household 31 (16.7)
Table 2. Performance-based self-esteem (PBSE), study conditions,
working conditions and exhaustion at T2 and T3.
Scale
range Mean s
Final
undergraduate
year (T2)
PBSE 1–5 3.40 0.81
Worries about
future endurance/
capacity
1–4 2.66 0.79
Non-supportive
climate
1–4 1.65 0.48
Faculty
shortcomings
1–4 2.49 0.39
Insuffi cient
feedback
1–4 2.86 0.62
Low commitment 1–4 1.32 0.47
Exhaustion 1–4 2.21 0.49
1st postgraduate
year (T3)
Weekly working
hours
24–80
∗ 45.7 9.1
Climate for
learning
10–50 36.3 6.2
Exhaustion 1–4 2.29 0.52
∗ Actual range in sample, no scale.
s , standard deviation

PREDICTORS OF EXHAUSTION IN YOUNG PHYSICIANS
NORD J PSYCHIATRY·VOL 64·NO 6·2010 405
position), none of which infl uenced the results or turned
out signifi cant. There was no interaction between gender
and any of the predictors.
Discussion
We studied predictors of exhaustion in junior doctors in
their fi rst postgraduate year. The study condition WFEC
at fi nal year of medical school predicted postgraduate
exhaustion, while positive working conditions had a pro-
tective effect. Women were more exhausted than men,
which was explained by their higher WFEC. PBSE was
not a signifi cant predictor, when controlling for fi nal year
undergraduate exhaustion, which in itself was a strongly
signifi cant predictor.
Exhaustion
We found that exhaustion scores increased from T2 to
T3, but only among women. The fi nding of higher dis-
tress scores among women is in line with some previ-
ous studies of medical professionals, although data are
divergent on this matter (22,23). We have shown previ-
ously that Swedish female medical students had higher
exhaustion levels, regardless of stage of education (29).
A recent Swedish study of female physicians found
that 20% had sought professional help for burnout or
exhaustion at follow-up (Table 3), but not being cohabitant
( β  −0.025, 95% CI −0.180 to 0.131) or having a parental
responsibility ( β  −0.131, 95% CI −0.339 to 0.077). Since
data on PBSE was collected at different time-points in the
two cohorts, we controlled this association for cohort,
which did not change the β for PBSE more than margin-
ally. The Climate for learning at the working place was
negatively associated with exhaustion, while there was no
signifi cant relationship with working hours.
Multivariable analysis
In the fi rst step, gender and PBSE were signifi cant pre-
dictors of postgraduate exhaustion, but when study condi-
tions were added, gender did not remain so (Step 2).
Further analyses showed that the effect of gender was
mediated by WFEC, which was the only study condition
signifi cant in the model. In adding the second step, the
explained variance increased from 0.16 to 0.32 (Adj R 2 ).
Climate for learning, entered in the third step, was
also signifi cant and added considerable explanatory
effect; the Adj R 2 of the model was 0.42. WFEC and
PBSE remained signifi cant. In a fi nal model, when we
controlled for exhaustion at T2, PBSE did not remain.
The variance explained in this fi nal model was 0.47.
Additional analyses were also run to control for
cohort and type of work (intern or holding a temporary
Table 3 . Predictors and associated concurrent work-related factors of Exhaustion in the fi rst postgraduate year .
Predictors/associated factors
Bivariate analyses Step 1 Step 2 Step 3 Step 4
Unadj B 95% CI B 95% CI B 95% CI B 95% CI B 95% CI
Gender 0.241
∗ ∗ 0.089 to
0.393
0.233
∗ ∗ −0.086 to
0.380
0.028 −0.123 to
0.179
0.088 −0.052 to
0.229
0.080 −0.056 to
0.215
Age 0.012 −0.007 to
0.030
0.012 −0.006 to
0.031
0.009 −0.008 to
0.025
0.008 −0.007 to
0.023
0.005 −0.009 to
0.020
Performance-based
self-esteem
0.192
∗ ∗ 0.099 to
0.284
0.206
∗ 0.117 to
0.296
0.137
∗ ∗ 0.054 to
0.221
0.100
∗ ∗ 0.022 to
0.178
0.072 −0.004 to
0.149
Worries about future
endurance/capacity
0.351
∗ ∗ 0.268 to
0.434
– – 0.259 ∗ 0.159 to
0.360
0.208
∗ 0.114 to
0.303
0.165
∗ ∗ 0.071 to
0.259
Non-supportive climate 0.376
∗ ∗ 0.228 to
0.525
– – 0.120 – 069 to
0.308
0.116 −0.037 to
0.189
0.048 −0.122 to
0.219
Faculty shortcomings 0.327
∗ ∗ 0.136 to
0.518
– – −0.105 −0.334 to
0.124
−0.153 −0.364 to
0.059
−0.176 −0.380 to
0.027
Insuffi cient feedback 0.219
∗ ∗ 0.100 to
0.338
– – 0.103 −0.020 to
0.226
0.076 −0.037 to
0.189
0.067 −0.043 to
0.176
Low commitment 0.210
∗ 0.048 to
0.371
– – 0.037 −0.118 to
0.192
0.038 −0.105 to
0.180
0.010 −0.128 to
0.148
Exhaustion at fi nal
undergrad year
0.545
∗ ∗ 0.406 to
0.683
– – – – – – 0.282
∗ 0.134 to
0.430
Weekly working hours 0.008 −0.001 to
0.017
– – – – – – – –
Climate for learning −0.043
∗ ∗ −0.054 to
−0.032
– – – – −0.029 ∗ −0.040 to
−0.019
−0.029
∗ −0.039 to
−0.018
Linear regression, method Enter.
Step 1: Demographic and PBSE.
Step 2: Bivariately signifi cant study conditions added.
Step 3: Bivariately signifi cant working conditions added.
Step 4: Controlling for Exhaustion at fi nal year of medical school.
* P  0.05, * * P  0.001.

M DAHLIN ET AL.
406 NORD J PSYCHIATRY·VOL 64·NO 6·2010
sample had been assessed for PBSE at third year of medi-
cal school, and it may have had an effect on fi nal year
exhaustion, which swallowed much of the variation while
entered in the regression analysis. Since the study was not
designed to explore this specifi cally, no subgroup analyses
were made. There is a lack of longitudinal studies on
PBSE and its predictive potential deserves further study.
There were no interactions between gender and any
of the predictors, thus the same risk factors operated for
men and women.
Working conditions
Working hours was not associated with higher exhaustion,
as has been noted in some previous studies (23, 43). Work-
ing hours for young Swedish physicians are comparable
with other Scandinavian countries (23), although consider-
ably lower than for UK or US interns. Our results on Cli-
mate for learning confi rm previous fi ndings, showing a
negative association with job stress among interns (23). No
conclusions regarding causality can be made, but a positive
Learning climate may be a protective factor. It may also be
considered a job resource, which counteracts exhaustion
and burnout (44). Thus the pedagogical aspects of the
working environment may be an area to acknowledge when
aiming at improved conditions for young physicians.
Limitations
This study relies on self-report measures; observational
data or objective measures of the working climate would
have allowed for a strict demarcation line between indi-
vidual characteristics and working conditions, and the
associations found should be interpreted with caution.
Slightly different measures were used to assess exhaus-
tion at T2 and T3, because of the adaptation to studies in
the student version. The authors of the OLBI gave their
permission for this revision, but no specifi c validation
process has been undertaken.
The response rate of 73.5% was satisfactory, and the
sample representative of the study population with regard
to gender, age and T2 exhaustion scores. Data were col-
lected from two cohorts, graduating with a 2.5-year
interval, but results were stable when controlling for this.
The participants had all graduated from one medical
school, which limits generalization of the results. Finally,
we did not control for actual academic skills. Thus any
association between a tendency to worry about ones
capacity and actual defi ciencies in knowledge and com-
petence cannot be ruled out.
Conclusions
Postgraduate exhaustion was predicted by the study condition
WFEC at fi nal year of medical school, which particularly
affected female medical students. The effect of PBSE should
not be ruled out, since it may increase the risk of being
depression (36). Further, as the fi nal step in the regres-
sion model showed, exhaustion at fi nal year of medical
school was itself a strong and signifi cant predictor of
postgraduate exhaustion, indicating that the exhaustion
reaction may also be persistent.
Individual and environmental predictors
Gender in itself was not predictive, when study condi-
tions were controlled for, nor was having a parental
responsibility, which might have been an expected gender-
related demographic factor, e.g. by work – home
interference (22, 37).
Among study conditions, only WFEC was a risk factor
of exhaustion in our study, and mediated the effect of
gender. The propensity of worrying about the future may
refl ect a tendency of brooding or rumination, known to be
associated with stress and depression (38, 39). It has been
suggested that personality and trait-like factors explain the
development of burnout and attitudes towards work in
physicians (14, 16). WFEC may in itself be affected by
both personality traits and environmental aspects. Women
are known to be more prone to worrisome thinking and
brooding than men (40), hence there is support for a gen-
der association that is not specifi c to medical students.
However, a previous study of our group on medical stu-
dents showed that WFEC increased from fi rst to third year
of medical school for women only (13). Study-related and
curricular conditions, and possibly harassment and belittle-
ment in clinical training (21) may thus infl uence the ten-
dency to worry, and specifi cally undermine the self-
conscience of women in medicine. Still, students of both
sexes are sensitive to the signals given by teachers and
senior colleagues; by constant reminders of the “ life-
staking ” importance of each and every detail of the cur-
riculum in their efforts to motivate students, medical
teachers may increase counterproductive anxiety. WFEC in
fi nal year students ’ may indicate a lack of confi dence in
their preparation for medicine and the fear of making
medical mistakes. A recent study showed that a growing
proportion of newly qualifi ed British doctors felt well pre-
pared for their profession, which the authors partly attrib-
uted to modernized curricula and pedagogies (41). Among
Norwegian medical students, “ brooders ” were at risk of
stress later in medical school (42). It is an important task
for medical educators not only to enhance students ’ learn-
ing of medical knowledge, but also to help students to
achieve a basic sense that they can acquire suffi cient qual-
ifi cations to enter the profession.
When controlling for T2 exhaustion, PBSE was no lon-
ger a signifi cant predictor in the model. In a previous
cross-sectional study of ours, PBSE was associated with
exhaustion (29). Although the theoretical assumption
behind PBSE and burnout is that pure exhaustion would
not necessarily entail a high PBSE (5), our fi ndings sug-
gest that there is a relevant association. A part of the

PREDICTORS OF EXHAUSTION IN YOUNG PHYSICIANS
NORD J PSYCHIATRY·VOL 64·NO 6·2010 407
Dahlin ME, Runeson B. Burnout and psychiatric morbidity among 13.
medical students entering clinical training: A three year prospective
questionnaire and interview-based study. BMC Med Educ
2007;7:6.
McCranie EW, Brandsma JM. Personality antecedents of burnout 14.
among middle-aged physicians. Behav Med 1988;14:30 – 6.
Lamberg L. “ If I work hard(er), I will be loved. ” Roots of physician 15.
stress explored. JAMA 1999;282:13 – 4.
McManus IC, Keeling A, Paice E. Stress, burnout and doctors ’ 16.
attitudes to work are determined by personality and learning style:
A twelve year longitudinal study of UK medical graduates
(Electronic). BMC Med 2004;2.
Riley GJ. Understanding the stresses and strains of being a doctor. 17.
Med J Aust 2004;181:350 – 3.
Tyssen R, Vaglum P, Gronvold NT, Ekeberg O. The impact of job 18.
stress and working conditions on mental health problems among
junior house offi cers. A nationwide Norwegian prospective cohort
study. Med Educ 2000;34:374 – 84.
Gjerberg E, Kjolsrod L. The doctor – nurse relationship: How easy is 19.
it to be a female doctor co-operating with a female nurse? Soc Sci
Med 2001;52:189 – 202.
Kilminster S, Downes J, Gough B, Murdoch-Eaton D, Roberts T. 20.
Women in medicine — Is there a problem? A literature review of the
changing gender composition, structures and occupational cultures
in medicine. Med Educ 2007;41:39 – 49.
Nora LM, McLaughlin MA, Fosson SE, Stratton TD, 21.
Murphy-Spencer A, Fincher RM, et al. Gender discrimination and
sexual harassment in medical education: Perspectives gained by a
14-school study. Acad Med 2002;77(12 Pt 1):1226 – 34.
Firth-Cozens J. Emotional distress in junior house offi cers. Br Med 22.
J (Clin Res Ed) 1987;295:533 – 6.
Tyssen R, Vaglum P, Gronvold NT, Ekeberg O. The relative impor-23.
tance of individual and organizational factors for the prevention of
job stress during internship: A nationwide and prospective study.
Med Teach 2005;27:726 – 31.
Fahrenkopf AM, Sectish TC, Barger LK, Sharek PJ, Lewin D, 24.
Chiang VW, et al. Rates of medication errors among depressed and
burnt out residents: Prospective cohort study. BMJ 2008;336:
488 – 91.
Clark DC, Salazar-Grueso E, Grabler P, Fawcett J. Predictors of 25.
depression during the fi rst 6 months of internship. Am J Psychiatry
1984;141:1095 – 8.
Hsu K, Marshall V. Prevalence of depression and distress in a large 26.
sample of Canadian residents, interns, and fellows. Am J Psychiatry
1987;144:1561 – 6.
Martini S, Arfken CL, Churchill A, Balon R. Burnout comparison 27.
among residents in different medical specialties. Acad Psychiatry
2004;28:240 – 2.
Firth-Cozens J, Lema VC, Firth RA. Specialty choice, stress and 28.
personality: Their relationships over time. Hosp Med 1999;60:
751 – 5.
Dahlin M, Joneborg N, Runeson B. Performance-based self-esteem 29.
and burnout in a cross-sectional study of medical students. Med
Teach 2007;29:43 – 8.
Henning K, Ey S, Shaw D. Perfectionism, the imposter phenomenon 30.
and psychological adjustment in medical, dental, nursing and
pharmacy students. Med Educ 1998;32:456 – 64.
Enns MW, Cox BJ, Sareen J, Freeman P. Adaptive and maladaptive 31.
perfectionism in medical students: A longitudinal investigation. Med
Educ 2001;35:1034 – 42.
Demerouti E, Bakker AB, de Jonge J, Janssen PP, Schaufeli WB. 32.
Burnout and engagement at work as a function of demands and
control. Scand J Work Environ Health 2001;27:279 – 86.
Demerouti E, Bakker AB, Vardakou I, Kantas A. The convergent 33.
validity of two burnout instruments: A multitrait – multimethod
analysis. Eur J Psychol Assess 2003;19:12 – 23.
Hallsten L, Bellaagh K, Gustafsson K. Utbr ä nning i Sverige- en 34.
populationsstudie (Burnout in Sweden — A population study). Arbete
och h ä lsa 2002:6. Stockholm: National Institute for Working Life
2002. Report No.: 91-7045-639-9.
Dahlin M, Joneborg N, Runeson B. Stress and depression among 35.
medical students: A cross-sectional study. Med Educ 2005;39:594 – 604.
exhausted already at the end of medical school. The strong
association with exhaustion at fi nal year indicates that the
burnout process may begin already in medical school. A
positive Learning climate towards newly graduated doctors
had a possible protective effect against exhaustion. Medical
educators and clinical preceptors in undergraduate and post-
graduate training should support and encourage students and
young colleagues in their professional role, and be aware of
possible gender patterns. Students with high doubts about
themselves may benefi t from specifi c programmes in medi-
cal school, with focus on self-criticism and possibly exag-
gerated perceptions of demands. Clear learning objectives
and expectations adapted to ideas of an appropriate progres-
sion of knowledge and skills through under- and postgradu-
ate medical education may be helpful.
Acknowledgements — We thank all participators. Professor Reidar Tyssen
gave valuable comments on the manuscript. Funding was received from AFA
Insurances and from the Karolinska Board of Education. Ethical approval
was obtained from the Ethics committee at Karolinska Institutet (00-403,
Dec 2000).
Disclosure of interest : MD and BR teach psychiatry at
Karolinska Institutet Medical School.
References
1. Shaufeli W, Enzmann D. The burnout companion to study
and practice — A critical analysis. London: Taylor and Francis;
1998.
Maslach C, Schaufeli WB, Leiter MP. Job burnout. Annu Rev 2.
Psychol 2001;52:397 – 422.
Ahola K, Honkonen T, Virtanen M, Aromaa A, Lonnqvist J. 3.
Burnout in relation to age in the adult working population.
J Occup Health 2008;50:362 – 5.
Demerouti E, Bakker AB, Nachreiner F, Schaufeli WB. The job 4.
demands-resources model of burnout. J Appl Psychol 2001;86:
499 – 512.
Hallsten L, Josephson M, Torg é n M. Performance-based self-5.
esteem. A driving force in burnout processes and its assessment.
Marklund S, editor. Stockholm: National Institute for Working
Life; 2005.
Ramirez AJ, Graham J, Richards MA, Gregory WM, Cull A. 6.
Mental health of hospital consultants: The effects of stress and
satisfaction at work. Lancet 1996;347:724 – 8.
Aasland OG, Olff M, Falkum E, Schweder T, Ursin H. Health 7.
complaints and job stress in Norwegian physicians: The use of an
overlapping questionnaire design. Soc Sci Med 1997;45:1615 – 29.
Arnetz BB. Psychosocial challenges facing physicians of today. 8.
Soc Sci Med 2001;52:203 – 13.
Firth-Cozens J. Doctors, their wellbeing, and their stress. BMJ 9.
2003;326:670 – 1.
Prins JT, Gazendam-Donofrio SM, Tubben BJ, van der Heijden 10.
FMMA, van de Wiel HBM, Hoekstra-Weebers JEHM. Burnout in
medical residents: A review. Med Educ 2007;41:788 – 800.
Guthrie E, Black D, Bagalkote H, Shaw C, Campbell M, Creed F. 11.
Psychological stress and burnout in medical students: A fi ve-year
prospective longitudinal study. J R Soc Med 1998;91:237 – 43.
Dyrbye LN, Thomas MR, Huschka MM, Lawson KL, Novotny PJ, 12.
Sloan JA, et al. A multicenter study of burnout, depression, and
quality of life in minority and nonminority US medical students.
Mayo Clin Proc 2006;81:1435 – 42.

M DAHLIN ET AL.
408 NORD J PSYCHIATRY·VOL 64·NO 6·2010
trained them well: Questionnaire surveys. BMC Med Educ
2007;7:38.
Tyssen R, Dolatowski FC, Rovik JO, Thorkildsen RF, Ekeberg O, 42.
Hem E, et al. Personality traits and types predict medical school
stress: A six-year longitudinal and nationwide study. Med Educ
2007;41:781 – 7.
Firth-Cozens J. Interventions to improve physicians ’ well-being and 43.
patient care. Soc Sci Med 2001;52:215 – 22.
Schaufeli WB, Bakker AB. Job demands, job resources, and their 44.
relationship with burnout and engagement: A multi-sample study.
J Organ Behav 2004;2:293 – 315.
Marie Dahlin, M.D., Ph.D., Department of Clinical Neuroscience,
Karolinska Institutet, Division of Psychiatry St. Goran, SE-112 81
Stockholm, Sweden.
Jenny Fjell, M.D., Ph.D., Northern Stockholm Psychiatry,
Stockholm. Sweden.
Bo Runeson, Professor, Department of Clinical Neuroscience,
Karolinska Institutet, Division of Psychiatry St. Goran, SE-112 81
Stockholm, Sweden.
Fridner A, Belkic K, Marini M, Minucci D, Pavan L, 36.
Schenck-Gustafsson K. Survey on recent suicidal ideation among
female university hospital physicians in Sweden and Italy (the
HOUPE study): Cross-sectional associations with work stressors.
Gend Med 2009;6:314 – 28.
Geurts S, Rutte C, Peeters M. Antecedents and consequences of 37.
work – home interference among medical residents. Soc Sci Med
1999;48:1135 – 48.
Denson TF, Spanovic M, Miller N. Cognitive appraisals and emo-38.
tions predict cortisol and immune responses: a meta-analysis of
acute laboratory social stressors and emotion inductions. Psychol
Bull 2009;135:823 – 53.
Morrison R, O ’ Connor RC. The role of rumination, attentional 39.
biases and stress in psychological distress. Br J Psychol 2008;99
(Pt 2):191 – 209.
Robichaud M, Dugas MJ, Conway M. Gender differences in worry 40.
and associated cognitive – behavioral variables. J Anxiety Disord
2003;17:501 – 16.
Cave J, Goldacre M, Lambert T, Woolf K, Jones A, Dacre J. Newly 41.
qualifi ed doctors ’ views about whether their medical school had

Copyright of Nordic Journal of Psychiatry is the property of Taylor & Francis Ltd and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.