British Columbia Medical Journal, March 2010 issue Please download or visit this entire issue online at http://bcmj.org/march-2010

1. Steven A. Green, Gary D. Poole, PhD
Resident work hours: Examining
attitudes toward work-hour limits
in general surgery, orthopaedics,
and internal medicine
Surgical and nonsurgical residents who were surveyed about their
work hours expressed different opinions about restricting their
hours, with surgical residents favoring fewer restrictions than non-
surgical residents.
ABSTRACT Background Reaction to work-hour regulations
Background: Residency work hours Recently the issue of resident work has been mixed. A frequently cited
are currently receiving considerable hours and, more specifically, work- divide is one between surgical and
attention. Work-hour limits have hour limits has received increased nonsurgical disciplines,7 where it is
been set in the US and the EU, and attention in medical education.1-3 Con- often noted that surgical residents will
the Professional Association of Res- cerns about the effects of resident simply not obtain the operative expe-
idents of BC has negotiated a con- sleep deprivation due to long working rience necessary for future practice if
tract stipulating a 24-hour limit to hours led the Accreditation Commit- work hours are limited.8 Meanwhile,
shift length. In surgical disciplines, tee for Graduate Medical Education in internal medicine, work-hour regu-
however, long hours are thought to (ACGME) in the United States to set lations are more frequently perceived
be necessary to learn procedures. limits on work hours. The ACGME to have a positive impact on resident
specifically limits the work week to education.9 As for whether surgical
Methods: To examine attitudes to-
80 hours (including all time in hospi- experience is diminished by work-
ward work-hour limits, a question-
tal) while requiring that residents have hour regulations, there is conflicting
naire was created and distributed to
1 day off in 7 and that no single shift evidence on the matter. 10-13 In the
residents in general surgery, ortho-
continue for longer than 24 hours.4 Netherlands, surgical residents ob-
paedics, and internal medicine at
Meanwhile, in the European Union, serving the European Working Time
the University of British Columbia in
the European Working Time Direc- Directive are quite satisfied with the
February 2009.
tive, which covers most areas of regulations and do not perceive them
Results: Survey results indicated employment, recently came to include as a threat to their training.14
that surgical residents favor fewer medical trainees. It originally man- Despite the attention being paid to
work-hour restrictions when com- dated a 56-hour work week, which resident work hours, there is very lit-
pared with nonsurgical residents. was changed to a 48-hour work week tle Canadian data concerning the atti-
on 1 August 2009.5 In Canada, there is tudes of residents toward work-hour
Conclusions: Concern about pro -
no similar legislation, but provincial regulations. This lack of data leaves
cedural competence explains some
organizations of residents have nego-
individual variability in terms of
tiated contracts with health authorities Mr Green is a third-year medical student at
these attitudes but fails to explain
to set limits. For instance, in BC the the University of British Columbia. Dr Poole
the between-group difference.
Professional Association of Residents is an associate professor in the School of
(PAR-BC) has negotiated a contract Population and Public Health at UBC and
stipulating a 24-hour limit to shift the director of UBC’s Centre for Teaching
length.6 and Academic Growth.
84 BC MEDICAL JOURNAL VOL. 52 NO. 2, MARCH 2010 www.bcmj.org

2. Resident work hours: Examining attitudes toward work-hour limits in general surgery, orthopaedics, and internal medicine
Resident Work Hours Survey Please rate how much you agree with the following statements
(Circle your answer: 1-Strongly disagree; 3-neutral; 5-strongly agree)
Conducted in association with the UBC School of Population
and Public Health General Questions:
Purpose: To examine resident work hours and attitudes towards 8. I am sleep-deprived on a regular basis.
work hours limitations 1 2 3 4 5
Strongly Disagree Neutral Agree Strongly
Instructions: There are 18 questions in this survey. Questions 3-7 Disagree Agree
require a numerical answer; question 18 is optional and intended
for you to express any comments, and the remainder of the ques- 9. I feel overworked.
tions are multiple-choice. The survey should take no more than 1 2 3 4 5
10 minutes, and your consent to participate is implied by your Strongly Disagree Neutral Agree Strongly
completion of this survey (see attached consent form). Disagree Agree
10. I feel pressured to work more.
1. Please circle the letter that best corresponds to your training
program: 1 2 3 4 5
a) Anesthesia Strongly Disagree Neutral Agree Strongly
Disagree Agree
b) Community Medicine
c) Dermatology 11. If I could extend my residency in order to work fewer hours,
d) Emergency Medicine I would.
Please circle:
1 2 3 4 5
1) CCFP-EM
Strongly Disagree Neutral Agree Strongly
2) Royal College Disagree Agree
e) Family Medicine
f) Internal Medicine If tighter work hour restrictions were imposed (3 Questions):
General (or in Core 3-yr program)
Subspecialty: ________________________________ 12. I wouldn’t have time to master the procedures of my specialty
g) Neurology during residency.
h) Obstetrics/Gynecology 1 2 3 4 5
i) Pediatrics Strongly Disagree Neutral Agree Strongly
General (or in Core 3-yr program) Disagree Agree
Subspecialty: ________________________________
j) Pathology/Lab Medicine 13. My residency training would still adequately prepare me for prac-
k) Radiology tice.
l) Surgery: 1 2 3 4 5
Please circle: Strongly Disagree Neutral Agree Strongly
1) Cardiac Disagree Agree
2) ENT 14. I would sleep more.
3) General
4) Neurosurgery 1 2 3 4 5
5) Opthalmology Strongly Disagree Neutral Agree Strongly
Disagree Agree
6) Orthopedics
7) Plastics
If I slept more (3 questions):
8) Urology
9) Other ____________________________________ 15. I would learn more effectively.
m) Other: ______________________________________ 1 2 3 4 5
Strongly Disagree Neutral Agree Strongly
Disagree Agree
2. Program year (eg. PGY-1) (circle one): 1 2 3 4 5 6 7 8
16. I would master procedures faster.
3. Hours worked in last 7 days (including in-house call): _____ 1 2 3 4 5
Strongly Disagree Neutral Agree Strongly
Disagree Agree
4. Length of longest shift in last 7 days (including in-house call):
_____________ 17. I would commit fewer medical errors.
1 2 3 4 5
Strongly Disagree Neutral Agree Strongly
5. Hours slept last night: _____ Disagree Agree
6. Total hours slept in last 7 days: _____ 18. Optional: How do you see work hour limitations impacting your
training?
7. If I were to set a work hour limit (including in-house call), ______________________________________________________
I would set it at: _____hours/ week ______________________________________________________
Figure 1. Resident work hours survey distributed to residents in internal medicine, general surgery, and orthopaedics at the University of British
Columbia in February 2009.
www.bcmj.org VOL. 52 NO. 2, MARCH 2010 BC MEDICAL JOURNAL 85

3. Resident work hours: Examining attitudes toward work-hour limits in general surgery, orthopaedics, and internal medicine
several questions unanswered. First, tion between such concerns and atti- medicine, general surgery, and ortho-
do surgical residents favor fewer work- tudes toward work-hour restrictions? paedics at the University of British
hour restrictions than nonsurgical res- Columbia in February 2009 ( Figure 1 ).
idents? Second, do surgical residents Methods The questionnaire was distributed at
believe their procedural competency To attempt to answer these questions, academic half-days in paper form and
will be compromised by tighter an 18-item questionnaire was created was collected the same day. For the
restrictions? Third, is there a correla- and distributed to residents in internal orthopaedics and general surgery res-
idents, additional questionnaires were
given to office staff to distribute to the
100
residents missing from the half-day,
and these were then collected 1 week
90
Surgical later (orthopaedics) and 2 weeks later
80 Nonsurgical (general surgery).
70 The questionnaire included items
60
related to sleep and work hours that were
adapted from Fok and colleagues15 and
Hours*
50
items related to attitudes that were
40 similar to those asked by Morris-Stiff
30 and colleagues.5 The first part of the
20
questionnaire asked for numerical
responses concerning hours worked,
10
hours slept, and a suggested work-hour
0 limit in hours per week. The second
3. Hours worked 4. Longest shift 6. Hours of 7. If I were to set
in last 7 days in last 7 days sleep in a work-hour limit, part asked for qualitative responses to
last 7 days I would set it at statements such as, “If tighter work-
hour restrictions were imposed, I would
Figure 2. Comparison of work and sleep hours, and suggested work-hour limits, based on
responses to questions 3, 4, 6, and 7 from surgical and nonsurgical residents.
not have time to master procedures.”
Respondants used a scale of 1 to 5,
Note that surgical residents worked longer hours than nonsugical residents and recommended higher with 1 indicating “Strongly disagree”
work-hour limits. Also note that there is a remarkable similarity between the work-hour recommenda-
tions and the current conditions for each group of residents. and 5 indicating “Strongly agree.”
*Error bars show 95% confidence interval. Columns 3 and 6 show statistical significance > .05 The study was conducted with the
approval of the UBC Behavioural
Table 1. Results from numerical response questions 3 to 7 about work and sleep hours.
Research Ethics Board.
Number of Hours t-test for significance Results
Question responses by Significance Mean Survey response rates varied by disci-
group Mean SD
(2-tailed) difference pline. Of the 52 internal medicine res-
3. Hours worked in Surgical: 35 77.89 21.23 idents who received questionnaires,
0.03200 10.73 39 responded for a response rate of
last 7 days Nonsurgical: 32 67.16 18.59
75%. Of the 30 questionnaires distrib-
4. Longest shift in Surgical: 36 27.46 12.65 uted to orthopaedics residents, 20 were
0.10000 4.54
last 7 days Nonsurgical: 32 22.92 9.50 returned for a response rate of 67%.
Surgical: 36 6.00 1.44 Meanwhile, of the 45 questionnaires
5. Hours of sleep
last night
0.42000 -0.41 distributed to general surgery resi-
Nonsurgical: 32 6.41 2.62
dents, 18 were returned for a response
6. Hours of sleep in Surgical: 36 41.81 8.92 rate of 40%.
0.47000 -1.38
last 7 days Nonsurgical: 30 43.18 6.04 Notable findings included signi-
7. If I were to set a Surgical: 35 82.00 14.26
ficant differences between surgical
work-hour limit, I 0.00040 14.85 (general surgery and orthopaedics)
would set it at: Nonsurgical: 31 67.15 18.02 and nonsurgical (internal medicine)
86 BC MEDICAL JOURNAL VOL. 52 NO. 2, MARCH 2010 www.bcmj.org

4. Resident work hours: Examining attitudes toward work-hour limits in general surgery, orthopaedics, and internal medicine
residents for year of program (2.66 vs Table 2. Results from Mann-Whitney U-test for significance of qualitative responses to
1.76, P < .01), hours worked in the last questions 8 to 17.
7 days (77.89 vs 67.16, P < .05), and
Number of Signifi-
recommended work-hour limit (82.00 Question responses by
Mean Rank
U Z cance
vs 67.15, P < . 001) ( Figure 2 ). There group
rank sum
(2-tailed)
was, however, no significant differ- 8. I am sleep- Surgical 38 41.04 1559.5 663.5 0.86 0.39
ence between the hours worked in deprived on a regu-
the last 7 days by a group and the lar basis Nonsurgical 39 37.01 1443.5
work-hour limit recommended by that Surgical 38 41.30 1569.5 653.5 0.95 0.34
group. This was seen in both the sur- 9. I feel overworked
Nonsurgical 39 36.76 1433.5
gical (77.9 vs 82.0, P = .248) and the
nonsurgical (67.2 vs 67.1, P = .438) 10. I feel pressured Surgical 38 40.42 1536.0 687.0 0.58 0.56
groups ( Table 1 ). to work more Nonsurgical 39 37.62 1467.0
For the questions requiring quali-
11. I would extend Surgical 38 36.33 1380.5 639.5 1.09 0.28
tative responses, hypothesis testing residency to work
was done using nonparametric meth- less (if I could) Nonsurgical 39 41.60 1622.5
ods. In comparing responses between 12. I would not mas-
surgical and nonsurgical residents, a ter procedures Surgical 37 40.64 1503.5 642.5 0.85 0.39
Mann-Whitney U-test was used to effectively (if there
were tighter work- Nonsurgical 39
rank responses from highest to lowest hour restrictions)
36.47 1422.5
value and then compare the mean rank
13. I would be ade-
between groups. No significant dif- quately prepared for Surgical 37 36.04 1333.5 630.5 1.00 0.32
ferences were found for any of the practice (if there
questions, including whether proce- were tighter work-
Nonsurgical 39 40.83 1592.5
hour restrictions)
dural competency would be hampered
by stricter work-hour regulations 14. I would sleep Surgical 37 38.99 1442.5 703.5 0.20 0.84
( Table 2 ). more (if there were
tighter work-hour
We also calculated the correlations restrictions) Nonsurgical 39 38.04 1483.5
between responses to the question
asking for a recommended work-hour 15. I would learn Surgical 38 38.78 1473.5 732.5 0.10 0.92
more (if I slept
limit and the questions asking how more) Nonsurgical 39 39.22 1529.5
work-hour restrictions would affect
16. I would master Surgical 38 40.28 1530.5 692.5 0.52 0.60
procedural competency and prepared- procedures faster (if
ness for practice. These correlations I slept more) Nonsurgical 39 37.76 1472.5
were performed while controlling for 17. I would commit Surgical 38 37.41 1421.5 680.5 0.65 0.52
differences in program and year. It was fewer medical
found that those concerned with pro- errors (if I slept
more) Nonsurgical 39 40.55 1581.5
cedural competency showed a moder-
ate tendency to suggest higher work-
hour limits (r = 0.458, α = 0.00018),
and those who believed that they would gical residents favored fewer work- between the groups in terms of con-
still be adequately prepared for prac- hour restrictions than nonsurgical res- cern about procedural competency
tice even with tighter restrictions show- idents, and our results indicate that being compromised, and whether
ed a moderate tendency to suggest they did, with surgical residents, on such concerns predicted differences
lower work-hour limits (r = 0.506, average, suggesting a weekly limit of in attitudes toward work-hour regula-
α = 2.7E 05). 82.00 hours and nonsurgical residents, tions. Regarding the second question,
on average, suggesting a 67.15-hour the answer appears to be no, as there
Conclusions limit. The second and third questions were no significant differences be-
This study sought to address three this study sought to answer were tween surgical and nonsurgical groups
questions. The first was whether sur- whether there were differences on any of the questions about attitudes
www.bcmj.org VOL. 52 NO. 2, MARCH 2010 BC MEDICAL JOURNAL 87

5. Resident work hours: Examining attitudes toward work-hour limits in general surgery, orthopaedics, and internal medicine
6. Professional Association of Residents of
British Columbia. Collective agreement.
There does appear to be an inverse Article 20: Scheduling. www.heabc.bc
.ca/public/CAs/RCA2006-2010.pdf7
relationship between concerns about (accessed 13 January 2010).
restricted work hours compromising 7. MacLellan AM. Residents’ duty hours in
the province of Quebec, Canada. Acad
procedural competency (and competency in Med 2003;78:11-13.
general) and suggested work-hour limits. 8. Urowitz M, Crescenzi AM, Muharuma L.
Residents’ duty hours in the province of
Ontario, Canada. Acad Med 2003;78:9-
10.
9. West CP, Cook RJ, Popkave C, et al. Per-
ceived impact of duty hours regulations:
to work hours. This may, however, attitudes are assumed or referred to A survey of residents and program direc-
reflect a small sample size; perhaps a anecdotally.7 The medical community tors. Am J Med 2007;120:644-648.
difference exists but it could not be needs to obtain data from more resi- 10. Kairys JC, McGuire K, Crawford AG, et
detected in a survey of fewer than 80 dents in a range of programs before al. Cumulative operative experience is
residents from three programs in one proclaiming that a group of residents decreasing during general surgery resi-
city. As for the third question, there feels one way or another about work- dency: A worrisome trend for surgical
does appear to be an inverse relation- hour limits. trainees? J Am Coll Surg 2008;206:804-
ship between concerns about restrict- 813.
ed work hours compromising proce- Competing interests 11. Damadi A, Davis AT, Saxe A, et al.
dural competency (and competency in None declared. ACGME duty-hour restrictions decrease
general) and suggested work-hour resident operative volume: A 5-year com-
limits. While this explains some of the References parison at an ACGME-accredited univer-
variability within any given group of 1. Bashir MR. Changes to resident call and sity general surgery residency. J Surg
residents, it fails to account for the dif- the dilution of education. J Am Coll Radi- Educ 2007;64:256-259.
ference between surgical and nonsur- ol 2009;6:277-278. 12. Schneider JR, Coyle JJ, Ryan ER, et al.
gical groups. 2. Grady MS, Batjer HH, Dacey RG. Resi- Implementation and evaluation of a new
Thus, further explanations must be dent duty hour regulation and patient surgical residency model. J Am Coll Surg
hypothesized and tested concerning safety: Establishing a balance between 2007;205:393-404.
the difference between surgical and concerns about resident fatigue and ade- 13. Romanchuk K. The effect of limiting res-
nonsurgical residents’ attitudes toward quate training in neurosurgery. J Neuro- idents’ work hours on their surgical train-
work-hour restrictions. One possibili- surg 2009;100:828-836. ing: A Canadian perspective. Acad Med
ty is illustrated by the striking corre- 3. Jagannathan J, Vates GE, Puratian N, et 2004;79:384-385.
spondence between the average work al. Impact of the Accreditation Council for 14. Wijnhoven BP, Watson DI, van den Ende
week for each group and the suggest- Graduate Medical Education work-hour ED. Current status and future perspec-
ed work-hour limit. Perhaps residents regulations on neurosurgical resident tive of general surgical trainees in the
in the surgical group are simply used education and productivity. J Neurosurg Netherlands. World J Surg 2008;32:119-
to working more hours on a regular 2009;110:820-827. 124.
basis, and thus when asked to set a rea- 4. Accreditation Council for Graduate 15. Fok MC, Townson A, Hughes B, et al.
sonable work-hour limit, they set it at Medical Education. Frequently asked Work hours, sleep deprivation, and
a level that reflects the status quo, questions about the ACGME common fatigue: A British Columbia snapshot.
while nonsurgical residents, who are duty hour standards. www.acgme.org/ BCMJ 2007;49:387-392.
not working as many hours on a regu- acWebsite/dutyHours/dh_faqs.pdf
lar basis, do the same. (accessed 13 January 2010).
Finally, it is clear that more re- 5. Villaneuva T. European Working Time
search remains to be conducted in Directive faces challenges. CMAJ 2010;
this field. In many cases, residents’ 182:E39.
88 BC MEDICAL JOURNAL VOL. 52 NO. 2, MARCH 2010 www.bcmj.org