Studying the association between postgraduate trainees' work hours, stress and the use of maladaptive coping strategies.

J Ayub Med Coll Abbottabad 2007; 19(3)
37
STUDYING THE ASSOCIATION BETWEEN POSTGRADUATE
TRAINEES’ WORK HOURS, STRESS AND THE USE OF MALADAPTIVE
COPING STRATEGIES
Pashtoon Murtaza Kasi, Talha Khawar, Farooq Hasan Khan, Jawad Ghazanfar Kiani,
Umber Zaheer Khan, Hadi Mohammad Khan, Urooj Bakht Khuwaja, Musa Rahim*
Final year medical students, Aga Khan University, Karachi, Pakistan. *Department of Community Health Sciences, Aga Khan
University, Karachi, Pakistan
Background: The growing debate regarding long working hours of postgraduate trainees has been
receiving considerable attention recently. This greater workload contributes to increasing stress.
Our objective was to specifically study the association between long working hours, stress and the
greater use of ‘maladaptive’ coping strategies. Methods : A cross-sectional descriptive study was
carried out on all interns and residents at the Aga Khan University Hospital during February to
May, 2005. Level of stress was measured by use of General Health Questionnaire (GHQ-12) and
the use of maladaptive coping mechanisms through Brief Cope-28. Results: 55.1 % scored over
the threshold for mild stress i.e. GHQ > 3, while more than 46% of the trainees scored over the
threshold of more than 4 for morbid stress. Trainees under stress reported more working hours on
average as compared to those not under stress, 83.8 and 74.7 hours respectively. At the same
time, those working for longer hours were more likely to have used these negative coping
mechanisms, which would further contribute to more stress rather than relieving it. Conclusions:
Significant levels of stress have been identified. Along with this, those working for longer hours
were more likely to have used these negative coping mechanisms. Reduction of working hours is
important. Simultaneously, interventions need to be planned at imparting knowledge, awareness
and skills to cope with various kinds of stressors encountered by a trainee during his/her training.
Additionally, limits need to be devised for the working hours of the trainees.
Keywords: Postgraduate, resident, intern, working hours, stress, maladaptive coping strategies,
GHQ-12, Brief Cope, Pakistan, Medical Education.
INTRODUCTION
The growing debate regarding long working hours of
postgraduate trainees has been receiving considerable
attention recently.1 This greater workload contributes
to increasing stress and decreases the overall
performance and the quality of the life of the affected
individuals. 2,3
At the moment, there is a constant struggle
with balancing working hours of residents in
Teaching Hospitals in terms of providing appropriate
time for maintenance of continuity in patient care and
yet avoiding stress.4
The Resident Service Committee of the
Association of Program Directors in Internal
Medicine, Philadelphia, has grouped these stresses
into situational, personal and professional stressors.5
The prevalence of stress in residents, along
with the aforementioned stressors, has been studied
in depth recently at our hospital (Personal
communication Dr. Asma Usman). Using General
Health Questionnaire (GHQ-12) as the screening
tool, the author found the prevalence of stress to be
around 60%.
Our hypothesis was to see if there is any
association between long working hours, stress and
the greater use of maladaptive coping strategies.
These would have two possible implications. Firstly,
it might be important to decrease the workload to
enhance the efficiency of the residents. On the other
hand, it may be important to identify and discourage
the use of ‘negative’ coping mechanisms, which
might further contribute to the stress of these
individuals, rather than helping them in relieving it.
These measures would help improve
patient’s safety as well as in the training, quality of
life and education of the residents.6,7
METHODS
A cross-sectional descriptive study was conducted on
all the interns and residents at the Aga Khan
University Hospital (AKUH), Karachi, Pakistan.
All residency programs are overseen by the
postgraduate medical education (PGME) committee
at AKU, which sets common goals and objectives for
the trainees or residents. There are now established
programs in 16 specialties.
Permission from the PGME was obtained
before the start of the project. Additionally, verbal
informed consent from each trainee was sought.
Confidentiality of records was maintained and the
data was only accessible to the ‘Working Hours
Group’.
Data was collected using a self-administered
standardized questionnaire with four sections. It
collected information regarding basic socio-

J Ayub Med Coll Abbottabad 2007; 19(3)
38
demographic characteristics, working hours of the
residents, levels of stress by use of General Health
Questionnaire (GHQ-12) and the use of maladaptive
coping mechanisms through Brief Cope.
Questions were also asked to address sleep
and other hours (i.e. time spent apart from work or
sleep).
Estimating the working hours
Being an important variable of the study, it was
imperative that the working hours of the
residents/interns should be rightly estimated. Ideally,
maintaining daily logs or signing in/off of registers at
counters would have given us a true estimate of the
amount of time spent in the hospital. But considering
the constraints and resources available, this step
would have been impractical. Plus, it would have
been a cumbersome task for the people filling out the
forms.
Therefore, we resorted to self-reported data
on work hours. A comprehensive literature search
was done and standard questions used in previously
published studies were carefully chosen.8,9 Correlates
of long working hours identified in previous studies
were also selected and included as part of the
questionnaire.
Working hours were estimated through three
methods:
1. Firstly, a direct open ended question was asked
to write down that on an average in the past
one month of his/her training, what was the
average number of hours worked per week?
(Self-reported work hours)
2. Next, through a series of questions in the
questionnaire addressing issues such as
frequency and duration of call, average
number of hours worked per day and number
of hours slept while on call (all in the past two
weeks), the average number of hours worked
was calculated using a formula. (Working
hours calculated through formula)
3. A third rough estimate was through a table that
the person would be required to fill in to mark
the timings of his presence in the hospital.
(Working hours calculated through table)
These three values were correlated to use a
correct estimate of the working hours.
GHQ-12
The level of stress was measured through the General
Health Questionnaire (GHQ-12), which is a measure
of current mental health. The questionnaire was
originally developed as a 60-item instrument but at
present a range of shortened versions of the
questionnaire including the GHQ-30, the GHQ-28,
the GHQ-20, and the GHQ-12 are available.10
The scale asks whether the respondent has
experienced a particular symptom or behavior
recently. Each item is rated on a four-point scale (less
than usual, no more than usual, rather more than
usual, or much more than usual). A score of 0-0-1-1
for the 4 responses from left to right, was given. The
cut off score would be between 3 and 4, as described
by Goldberg. All those who had scored 4 and above
were considered to be suffering from psychological
morbidity due to stress, and were labeled as
“Stressed” and those with the score of 3 and below
were labeled as “Not stressed”.
There is evidence that the GHQ-12 is a
consistent and reliable instrument when used in
general population samples and has also been used in
similar settings to address stress in consultants and
house officers.11
Maladaptive coping mechanisms
The frequency of different coping strategies
employed by the residents in the past 2 weeks was
determined with the Brief COPE – 28.12 It has 28
items measuring 14 different coping styles (2
questions for each coping method).
The responses to these questions are
measured on a 4-point Likert-type scale with
responses ranging from 1 (“I’ve not done this at all”)
to 4 (“I’ve been doing this a lot”).
These include, for example, ‘Active Coping’
(I’ve been taking action to try to make the situation
better), ‘Religion’ (I’ve been praying or meditating),
‘Venting’ (I’ve been expressing my negative
feelings) and ‘Substance Use’ (I’ve been using
alcohol or other drugs to make myself feel better).
As evident, some of these coping
mechanisms will have a positive effect on the
individual’s life and can be termed positive. On the
other hand, others that can worsen the situation can
be termed as ‘maladaptive’.13 According to our
hypothesis, these ‘maladaptive’ coping mechanisms
can be an outcome of long working hours of the
residents and if associated with stress as well, they
can further increase it if present.
Four of the 14 Brief COPE scales were used
to test the hypotheses of this study. These included:
1. Behavioral disengagement (e.g., "I've been
giving up trying to deal with it"),
2. Substance use (e.g., "I've been using
alcohol or drugs to make myself feel
better"),
3. Denial (e.g., "I've been saying to myself,
‘this isn't real’ "), and
4. Venting (e.g., "I've been saying things to let
my unpleasant feelings escape").
In the original validation study, the internal
consistency reliability coefficients (Cronbach's alpha)

J Ayub Med Coll Abbottabad 2007; 19(3)
39
for the four scales were 0.70, 0.90, 0.50, and 0.60,
respectively; translating to the fact that the questions
are reasonably reliable.
Additionally, several items were modified to
refer to psychological stress rather than situations
(e.g. "I refuse to believe that it has happened" was
changed to "I refuse to believe that I have these
symptoms;" "I've been giving up trying to deal with
it" was changed to "I've been giving up trying to deal
with these symptoms").
Sampling method and sample size calculation
From the PGME, a list of all the available residents
and interns was obtained. At the time of the study,
there were 272 residents and about 60 interns. Using
relative operative curve (ROC) tables and assuming a
50% prevalence of maladaptive coping amongst
residents/interns, it was decided that the
questionnaire be administered to all residents. This
would give us a power of 80% at a 95% confidence
interval.
Data Collection and entry
The medical students went to all departments of the
hospital where residents work including the wards,
clinics, operation room, emergency room, Intensive
Care Unit, laboratory, radiology, and others, and
distributed a self-administered questionnaire among
the residents. It was first inquired if the person could
spare 10-15 minutes of time. Only then he was given
the questionnaire to administer. If not, his/her pager
number was asked and he/she was then re-
approached at a later time of his/her convenience.
A mass mail was also sent to all the
residents and notices put up on the computers of the
learning resource centers of the university
encouraging them to participate in the study.
A database was developed in Microsoft Access 2000.
The data was imported in Microsoft Excel and
Statistical Package for Social Sciences 11 (SPSS 11)
for further analysis.
Statistical analysis
Baseline information on demographic and social
characteristics was obtained using descriptive
statistics, frequencies, means for continuous variables
and percentages for categorical variables. For
statistical significance, the chi square test would be
used for categorical variables, keeping the level of
significance (p) at 0.05 and 95% confidence interval
(CI) and t-test for continuous variables.
RESULTS
We were able to administer and collect filled
questionnaires from a total of 312 postgraduate
trainees (153 (49%) males and 159(51%) females).
68.3% were residents of Karachi. 33.3% were
married. The mean number of years since graduation
of the trainees was 4.98 (SD 3.49). At the time of
interview, there were a total of 59 interns, 72 R1s, 57
R2s, 34 R3s, 40 R4s, 9 R5s and 19 senior medical
officers (SMOs). SMOs were excluded from the final
analysis.
Working hours were estimated through three
methods, as outlined earlier. Paired t-tests were then
conducted to see if there were any significant
differences between these working hours calculated.
‘Self-reported’ working hours were higher (mean -
85.6 hours), while there was no statistically
significant difference (p-value < 0.001) between the
‘Working hours calculated through the Table’ (mean
- 80.2 hours) and the ‘Working hours calculated
through the formula’ (mean - 79.5 hours). An
average of the latter two was therefore used for
further analysis.
Figure 1 summarizes the average number of
hours worked by residents in different specialties.
Data on average number of hours slept and ‘other
hours’ i.e. apart from work and sleep has been
superimposed.
Figure 1: Comparison of work, sleep and other
hours/week according to rotation/specialty
55.1 % of the trainees scored over the
threshold for mild stress i.e. GHQ > 3, while more
than 46% of the trainees scored over the tentative
threshold of more than 4 for morbid stress.
80.8
59.4
60.5
60.7
62.4
67.4
69.3
70.9
71.5
75.4
79.5
81.1
82.6
85.2
91.3
92.1
96.7
104.7
105.7
38.2
45.5
49.0
41.6
45.7
43.0
40.3
42.5
35.8
36.3
35.4
38.6
39.0
37.4
29.9
36.8
34.8
33.7
49.0
62.5
58.5
65.7
59.9
57.6
58.4
54.5
60.2
55.7
53.0
48.3
46.4
45.4
46.8
39.1
36.9
33.6
28.7
29.8
0% 20% 40% 60% 80% 100%
Total
Pathology
Ophthalmology
ER
Dentistry
Community Medicine
Radiology
Family Medicine
ENT
Anesthesia
Pediatrics
Medicine
Psychiatry
Neurology
Urology
OBGYN
General Surgery
Neurosurgery
Orthopedics
R
ot
at
io
n/
Sp
ec
ia
lty

Hours per week
Work Hours Sleep Hours Other Hours

J Ayub Med Coll Abbottabad 2007; 19(3)
40
Trainees under stress i.e. GHQ > 3 reported
more working hours on average as compared to those
not under stress, 83.8 and 74.7 hours respectively.
The same relationship held true for morbid
stress (GHQ > 4) as well, 84.8 and 75.3 hours/week
respectively.
Similarly, individuals coping negatively
reported more mean number of hours worked per
week (Table 1). At the same time, individuals
practicing maladaptive coping styles were more
likely to be stressed as well (Table 2). Substance
abuse was reported minimally.
Table 1. Coping mechanisms and the mean number of
hours worked/week
Coping
mechanism
Mean
number of
hours
worked/week
n p-value
Denial
Not at all 79.0 164 0.04*
Yes 83.3 119
Total 283
Substance Misuse
Not at all 80.9 281 0.51
Yes 72.8 2
Total 283
Venting
Not at all 74.8 106 <0.001*
Yes 84.4 177
Total 283
Behavioral Disengagement
Not at all 80.0 211 0.18
Yes 83.2 72
Total 283
Table 2. The relationship between the different coping
mechanisms and mild stress
GHQ mild
stress

No Yes Total c2 p-value
Coping mechanism
Denial
Not at all 91 78 169 14.53 < 0.001*
Yes 39 85 124
130 163 293
Substance Misuse
Not at all 129 162 291 0.026 0.872
Yes 1 1 2
130 163 293
Venting
Not at all 75 36 111 38.96 <0.001*
Yes 55 127 182
130 163 293
Behavioral Disengagement
Not at all 116 102 218 26.98 <0.001*
Yes 14 61 75
130 163 293

DISCUSSION
We expected both underreporting and especially
over-reporting of the working hours. And as shown,
the mean working hours calculated both through the
formula and the schedule were lower than those
directly reported by the residents in response to an
open-ended question. However, there was little or no
difference in the hours calculated through the formula
and the hours estimated from the table; showing a
‘true’ estimation of the working hours. Still, the
working hours were very long in a number of
departments, exceeding the arbitrary limit of 80
hours.
We found significant levels of mild as well
as morbid stress in the trainees of the hospital, with
every second individual suffering from some degree
of stress.
Trainees working for longer hours were
more likely to have resorted to using these negative
coping mechanisms. This would further add to their
stress, rather than relieving it. The reason for low
scores on the substance misuse is probably
underreporting and the fact that the use of alcohol as
such in not a culturally and socially acceptable
phenomenon in Pakistan.
As with all studies, our report does have
limitations. The most obvious is that the data is self-
reported; the figures may thus be over reported or
underreported owing to different perceptions of the
difficulties of the training years. Since this is a cross-
sectional study, we would not be able to establish any
causal relationships. Plus the study is conducted in
residents from one teaching hospital in Pakistan with
characteristics and regulations of its own. The results
may be similar to but we would not be able to
generalize them to other hospitals.
Our study has a number of implications.
Firstly, we have been able to provide estimates of the
working hours of the residents. A detailed report has
been submitted to the PGME (available on request)
regarding working hours of all interns and residents,
and it is pleasing to note that a “Working Hours Task
Force”, which had already been formed before the
start of the study, is taking a number of steps to
reduce the working hours in some of the departments.
In fact in some, actions have already been taken and
the number of hours reduced.
We, however, from the above results cannot
suggest the limit for the working hours or people who
are working ‘too much’, but taking the 80 hour limit
as suggested by the American College of Graduate
Medical Education, a lot of individuals are
overworked and working beyond this limit, as
outlined earlier.

J Ayub Med Coll Abbottabad 2007; 19(3)
41
In Pakistan, there is no rule as such to define
the upper limit of the working hours. Such data might
be useful in devising future regulations and the work
time that may need to be reduced.
Secondly, we have been able to show the
relationship between working hours, stress and the
use of maladaptive coping behaviors. Although these
associations need to be studied further, some
implications are obvious. The use of these negative
coping mechanisms is going to further increase the
levels of stress of the residents.
Reduction of working hours is important. At
the same time interventions, need to be planned at
imparting knowledge, awareness and skills to cope
with various kinds of stressors encountered by a
trainee during his/her training. However, it would
probably be far from now that such an intervention
comes underway. The PGME might need to take the
initiative to help alleviate the problems of the trainees
and improve the overall quality of life of the training
resident. This would obviously in turn improve the
quality of care provided by the person. The formation
of the ‘Working Hours Task Force’ is an initial step
in the right direction.
ACKNOWLEDGEMENTS
All the authors are deeply indebted to the Community
Health Sciences Department for their constant
encouragement and support. We are also grateful to
all the interns and residents who cooperated a lot with
us and gave us time in filling of the questionnaires.
Dr. Salman Sabir and Dr. Iqbal Azam for
their guidance in statistical analysis. Dr. Asma
Usman for sharing her thesis on, “Stress and its
determinants in postgraduate trainees at a teaching
hospital in Karachi, Pakistan”. Dr. Aamir Jaffery, for
sharing his presentation on the work time of residents
rotating in surgery. Dr. Pervaiz Nayani for reviewing
the protocol and providing his comments. Mr. Saleem
Sabzali, Department of Community Health Sciences,
for his time and help for the project. Dr. Nadir Ali
Syed, Associate Dean, PGME, for his help, guidance
and encouragement for the project. Dr. Gauhar
Afshan, PGME and the working hours task force for
their cooperation. Thanks are again due to PGME for
allowing us to conduct this study.
REFERENCES
1. Howard SK, Gaba DM. Trainee fatigue: Are new limits on
work hours enough? CMAJ 2004; 170(6): 975-76.
2. Howard SK, Gaba DM, Rosekind MR, Zarcone VP. The
risks and implications of excessive daytime sleepiness in
resident physicians. Acad Med 2002; 77(10): 1019-25.
3. Parshuram CS, Dhanani S, Kirsh JA, Cox PN. Fellowship
training, workload, fatigue and physical stress: a prospective
observational study. CMAJ 2004; 170(6): 965-70.
4. Weinstein DF. Duty hours for resident physicians--tough
choices for teaching hospit als. N Engl J Med 2002;
347(16):1275-8.
5. Resident Services Committee, Association of Program
Directors in Internal Medicine. Stress and Impairment during
Residency Training: Strategies for Reduction, Identification,
and Management. Ann Int Med 1988; 109:154-61.
6. Steinbrook R. The Debate over Residents' Work Hours. N
Engl J Med. 2002; 347(16):1296-1302
7. Gaba DM, Howard SK. Patient safety: Fatigue among
clinicians and the safety of patients. N Engl J Med 2002;
347(16):1249-55.
8. Baldwin DC Jr, Daugherty SR. Sleep deprivation and fatigue
in residency training: results of a national survey of first - and
second-year residents. Sleep. 2004; 27(2):217-23.
9. Baldwin DC Jr, Daugherty SR, Tsai R, Scotti MJ Jr. A
national survey of residents' self-reported work hours:
thinkin g beyond specialty. Acad Med 2003; 78(11):1154-63.
10. Montazeri A, Harirchi AM, Shariati M, Garmaroudi G, Ebadi
M, Fateh A. The 12-item General Health Questionnaire
(GHQ-12): translation and validation study of the Iranian
version. Health Qual Life Outcomes 2003; 1(1):66.
11. Kapur N, Borrill C, Stride C. Psychological morbidity and
job satisfaction in hospital consultants and junior house
officers: multicentre, cross sectional survey. BMJ. 1998;
317(7157): 511-512.
12. Carver CS. You want to measure coping but your protocol’s
too long: consider the Brief COPE. Int J Behav Med 1997;
4(1): 92-100.
13. Vosvick M, Koopman C, Gore-Felton C, Thoresen C,
Krumboltz J, Spiegel D. Relationship of functional
quality of life to strategies for coping with the stress of
living with HIV/AIDS. Psychosomatics 2003;
44(1):51-8.
_____________________________________________________________________________________________
Address for Correspondence: Pashtoon Murtaza Kasi, 8-13/36, Kasi Road, Quetta, Pakistan. (C/O Fatima Clinic)
Phone: 0306-3737347.
E-mail:pashtoon.kasi@gmail.com