On July 1, 2003, the Accreditation Council on Graduate Medical Education
(ACGME) implemented the eighty-hour workweek rules, which set limits on the
number of hours that residents in the United States are allowed to
work1. Specifically,
the ACGME rules mandate that residents are not allowed to be on duty for more
than eighty hours per week averaged over a four-week period, and they may not
take in-house call more frequently than every third night. Additionally, the
regulations state that residents must have ten hours off between shifts and
must have one twenty-four-hour period per week free from clinical duties.
Finally, the maximum duration of a work shift is twenty-four hours, with an
optional extension of six extra hours for other educational activities.
This new policy was the culmination of years of media attention and
academic debate regarding the effects of excessive work hours on the
performance and well-being of house staff, most specifically the risks that
sleep-deprived residents posed to the patients for whom they were caring. The
1984 case of Libby Zion and the subsequent findings and recommendations of the
Bell Commission did much to inspire the eighty-hour duty-week movement in New
York and then
nationally2. In this
unfortunate circumstance, the patient died while under the care of a fatigued
member of the house staff, but the case and controversy that followed focused
more on resident supervision than on the issue of fatigue.
In addition to the national exposure received by the Libby Zion story,
studies pertaining to house-staff fatigue began to appear
nationally3-5.
A 1991 article in the Journal of the American Medical Association
reported that 41% of the 114 residents surveyed felt that they had made their
most serious medical mistakes secondary to fatigue, and that a third of these
errors resulted in the demise of a
patient6.
Following the implementation of the ACGME duty rules, many residency
training programs went through a period of adjustment, altering call schedules
and hiring additional ancillary staff to fill the service gap created by the
decreased availability of residents. Despite these changes, few medical and
surgical specialties have conducted widespread surveys of their own programs
to determine the effects of the new rules on residents within that particular
specialty. In the current study, we report the results of a large, nationwide
survey of orthopaedic residents to understand the attitudes and compliance of
these house-staff members toward the new rules. Strategies used by training
programs in response to these new mandates were also identified.
The Academic Advocacy Committee of the American Academy of Orthopaedic
Surgeons (AAOS) created a survey on resident duty-hour issues (see Appendix).
Committee membership included junior and senior orthopaedic residents,
attending physicians from academic orthopaedic surgery training programs, and
staff members of the AAOS. The survey was created with several objectives in
mind: (1) to determine the compliance of house staff with the new rules, (2)
to identify methods by which training programs have measured the number of
hours that residents were on duty, (3) to measure the attitudes of residents
toward the new duty rules, (4) to understand how programs have dealt with the
decreased availability of residents as a result of the rules, (5) to assess
the perceived impact of the rules on resident quality of life, well-being, and
educational experience, and (6) to record the perceived effects of the work
rules by residents on the quality and continuity of patient care. When it was
possible, the responses of junior and senior residents were compared. The
surveys also contained open-ended sections that allowed respondents to enter
additional comments regarding the new regulations.
In order to compare the responses of house staff at different stages of
training, residents were categorized as junior residents if they were at the
postgraduate year (PGY)-1 through PGY-3 level and as senior residents if they
were at the PGY-4 through PGY-6 level.
Data analyzed in the survey were collected in two rounds. On July 22 and
23, 2004, the survey was distributed to 1955 orthopaedic residents from all
postgraduate-year levels for whom a mail or fax address was available. Data
collection from this group was halted on August 13, 2004, and the data were
compiled. On November 1, 2004, an additional 2252 residents for whom an e-mail
address was available and who had not responded to the initial round of data
collection were contacted by e-mail with a link to an Internet site for
completion of the survey. On December 1, 2004, data collection for this second
round was discontinued. The responses from the two data collection rounds were
combined and electronically tabulated. At all times in the process, the survey
results were maintained with strict confidentiality. Survey data were analyzed
for validity and accuracy by the Department of Research and Scientific Affairs
of the AAOS.
When necessary, statistical analysis was performed with use of a paired
Student t test and a t test for probabilities. Significance was set at a
= 0.05. All statistics were performed with use of Stata software (Stata,
College Station, Texas).
The survey was distributed to a total of 4207 orthopaedic trainees. The
aggregated response rate of both rounds of survey distribution was 13.2% (554
respondents). The responses of these 554 residents were analyzed. Of note,
fifty-nine responses were from orthopaedic fellows or residents of unknown
postgraduate-year status. This left 495 residents whose postgraduate-year
status was known. When analyzing the responses to survey questions that
involved stratification by postgraduate-year level, the responses of only
those 495 residents whose postgraduate-year level was known were analyzed.
However, in all other survey questions that did not involve stratification by
postgraduate-year level, the responses of all 554 respondents were
analyzed.
Seventy-six percent of the 554 respondents were enrolled in academic
residency programs, and 16% were in community hospital-based programs. Five
percent of the respondents were enrolled in military-sponsored programs, and
3% failed to indicate the type of program in which they were participating. Of
the 495 respondents whose postgraduate-year level was known, 158 (32%) were
junior residents, while 337 (68%) were senior residents. Sixty-six percent of
the 554 respondents indicated that their residency programs consisted of
between ten and twenty-five residents.
Compliance with Duty-Hour Restrictions
Eighty-five percent (469) of the respondents indicated that they
consistently worked eighty hours or less, averaged over four weeks, after the
implementation of the new rules. Eighty percent responded that they worked
between sixty and eighty hours. However, 33% of the respondents reported
working greater than eighty hours during at least one single one-week period
after July 1, 2003; this occurred more commonly among PGY-3 and more junior
residents.
The responses regarding the procedures for reporting hours on duty at the
respondents' institutions demonstrated that, overall, 33% of all residents had
intentionally underreported their work hours during a week in which they
exceeded eighty hours (Fig. 1).
Forty-two percent (sixty-six) of the 158 junior residents had intentionally
underreported the number of hours on duty at least once, while 29%
(ninety-eight) of the 337 senior residents had submitted similar
misrepresentations. Only 14% of all 554 respondents indicated that they always
reported rule infractions to appropriate authorities, while 32% indicated that
they never reported such violations. The reasons why the respondents
intentionally misrepresented their duty hours were varied. The reasons cited
most often included concerns that their programs would be penalized for
violations (28% of the respondents), concern about personal citations (14%),
and disagreement with the importance of reporting and complying with the
eighty-hour rules (14%).
Reporting Methodology
Data were also collected on the method by which the home institutions of
the respondents monitored duty hours. Fifty-eight percent (324) of the
residents reported that their home institution utilized some form of a
self-reporting system to monitor hours. Under such a system, residents
themselves were responsible for calculating and reporting duty hours. Other
slightly more rigorous methods used for reporting duty hours included written
logs (28% of the residents) and electronic logs (22%). Only 1% of the
residents reported the use of more rigorous systems of logging hours, such as
electronic swipe cards or outside, external monitoring systems.
The quality of monitoring hours on duty was rated on a 5-point scale (with
1 indicating that the monitoring was extremely poor; 2, poor; 3, average; 4,
good; and 5, extremely well done); the average response was 4.2 points. The
residents' perception of the accuracy of their parent institution in
monitoring hours on duty demonstrated that 77% thought that this reflected a
"good" job. Forty-four percent believed that monitoring of duty
hours by their institution was done "extremely well," while <2%
thought that their institution was "extremely poor" in this
respect. Interestingly, 48% of the 337 senior residents indicated that their
institutions monitored hours "extremely well," while only 37% of
the 158 junior residents felt this way (p = 0.017).
Attitudes
Only twenty-three percent of the 554 respondents felt that eighty hours
constituted an appropriate number of duty hours per week; 41% thought that a
shorter duty schedule was appropriate, while 34% preferred the opportunity to
be on duty more than eighty hours per week. More specifically, 37.2% felt that
a sixty to seventy-nine-hour workweek was optimal, while 31.5% were of the
opinion that an eighty to ninety-nine-hour week was ideal. Twenty-seven
percent of the 158 junior residents and 24% of the 337 senior residents felt
that eighty hours constituted an optimal work week. Forty percent of the 554
respondents believed that the duty rules resulted in favorable changes in
their training programs (Table
I). However, 56% of the junior residents felt that the training
program had improved after implementation of the duty rules, while only 38% of
the senior residents expressed similar sentiments (p = 0.0002). Overall, 48%
of all residents thought that they had personally benefited from the duty
rules; 74% of the junior residents expressed this opinion compared with only
42% of the senior residents (p = 0.00002). The respondents indicated that, on
the average, 70% of duty time was educationally valuable, while 24% of the
hours consisted of "scut" work. More detailed analysis of this
question according to the year in training revealed little difference in
responses. Other attitudes regarding the eighty-hour workweek included the
fact that only 24% of the respondents felt that the duty rules had caused
deficits in learning and training experiences, and only 16% of the respondents
believed that continuity of care had suffered. Ten percent of the respondents
felt that the rules required senior residents to increase their workload to
compensate for reduced hours by junior residents. Similarly, 10% of the
respondents felt that the changes resulted in a "shift-worker"
mentality, primarily among the junior residents. Overall, only 13% of the
respondents felt that the changes in the duty rules had caused no new problems
to the residency training program.
Adaptive Strategies
Information concerning changes made by residency programs in their
respective duty schedules to comply with the eighty-hour rules was also
solicited. Eighty-two percent of the residents reported that their program had
made at least one change to their rotation schedules to be in compliance.
Fifty-seven percent of the respondents reported that their program allowed
residents to assist in the design and implementation of new rotation
schedules. The most common strategies included increasing the number of
home-call assignments (28% of the respondents), the use of physician
assistants or moonlighters (25%), and the use of a night-float system (22%)
(Fig. 2). Interestingly, 60% of
the respondents indicated that the changes in rotations required for
compliance had positive effects on the residency program.
As expected, the orthopaedic service area that had the most difficulty
complying with the duty rules was orthopaedic trauma. Eighty-one percent of
the residents reported that trauma services at their institutions encountered
substantial obstacles to complying with the new rules. The two other service
areas that were identified as frequently having compliance issues were adult
reconstruction (hip and knee arthroplasty) and spine surgery, with 24% of the
residents citing these services.
Quality-of-Life Issues
Regarding quality-of-life issues, 58% of the respondents reported that,
after the implementation of the new rules, most of their fellow residents were
generally happier with their training experiences and 57% indicated that they
were getting more rest. Junior residents were generally more in favor of the
duty-hour rules and perceived more positive effects from the new restrictions
(Table I). This is evidenced by
the fact that 56% of the junior residents in our survey indicated that their
training program had improved after implementation of the hour restrictions
(compared with 38% of the senior residents), and 74% of the junior residents
believed that they had personally benefited from the new rules (compared with
42% of the senior residents). Additionally, with regard to the issue of time
for rest, 68% of the junior residents indicated they had more time for rest,
while 56% of the senior residents felt they had more time for rest (p =
0.01).
Quality of Patient Care
The opinions of the respondents regarding the effects of the duty-hour
restrictions on the quality of patient care delivered were mixed. Overall, 38%
of the respondents felt that patient care had decreased in quality in some way
after the duty restrictions were implemented, while 51% perceived no such
decline. Interestingly, there was no difference in opinion between the junior
and senior residents regarding this question of quality of care.
The most commonly cited manifestations of decreases in quality were lapses
in continuity of care (35% of the respondents), missed injuries (16%),
unnecessarily increased length of hospital stay due to limited staff (10%),
and more inpatient complications (8%).
The eighty-hour workweek rules mandated by the ACGME have already had and
will continue to have a substantial impact on the structure of residency
training programs. Residency programs have had to be innovative in devising
strategies to enable compliance with these rules. Training programs in fields
such as internal medicine, obstetrics and gynecology, and pediatrics have used
approaches to reduce resident work hours for several years prior to the
implementation of the ACGME mandates on July 1,
20037,8.
Therefore, most current studies in the literature that have examined the
effects of the duty-hour rules on residents and on their attitudes have dealt
with nonsurgical specialties. The present study represents one of the largest
nationwide surveys of residents in a surgical specialty regarding the impact
of the new rules. Despite the fact that implementation of the new rules caused
substantial infrastructure challenges to orthopaedic residency training, the
net result with regard to resident attitudes toward these challenges and
reduced work hours has been positive.
First, the data indicate that most residents believe that their training
environment has been improved by the implementation of the new duty-hour
requirements. Other surgical specialties have reported clinical psychologic
distress among their residents that has been attributed to the large number of
hours worked9. The
perception of an improved work environment by orthopaedic residents in the
current study may be a result of decreased psychologic stress concurrent with
a decrease in the number of hours worked. Particularly, our data indicate
that, among orthopaedic residents, junior residents are more in favor of the
duty-hour rules and perceive more positive effects from the new duty-hour
restrictions. Chandra, in 2004, surveyed forty-six surgical subspecialty
residents and attending physicians and similarly found that senior residents
and faculty were "less accepting" of the restrictions in
comparison with junior
residents10.
Additionally, the results demonstrated that, among orthopaedic residents,
there is no agreement on the number of hours that constitute an ideal duty
week. Only 23% of the respondents in our group felt that eighty hours was the
ideal number, while 41% thought that even eighty hours represented too many
duty hours. These findings can be compared with those of Underwood et al. who
surveyed a group of eighty-six surgeons, of whom two-thirds were senior
surgical residents and one-third were junior-level attending
physicians11. Most
respondents (74.4%) indicated that a work-week of eighty-one hours or greater
was ideal, while only 25.6% felt that a work-week of eighty hours or less was
optimal. These findings demonstrate a common theme, whereby junior surgical
residents, who are being trained in a new era of duty-hour restrictions and
concerns about lifestyle and resident well-being, tend to be more in favor of
shorter duty weeks. This is in contrast to the attitudes of their more senior
resident and attending counterparts, who are products of an era of
unrestricted duty hours that paid little regard to resident lifestyles.
Therefore, it appears that the new duty-hour rules are in agreement with the
attitudes of a younger generation of surgeons who place greater value on
personal well-being and lifestyle than did previous generations.
A second conclusion to be drawn from the data in the present study is that
most orthopaedic surgery residents are using self-reporting systems to log
duty hours. The respondents in our survey reported a high rate of intentional
underreporting of duty hours. These findings call into question the accuracy
and usefulness of such self-reporting systems. A report by Saunders et al., in
2005, indicated a high rate of discrepancy in the number of work hours
documented when resident self-reporting systems were compared with a more
rigorous, electronic swipe-card
system12. In a
recent study conducted by the American Medical Association, 69% of the 1010
residents reported that they knew how to report excessive hours, but only
about half stated that they would feel comfortable doing
so13. In the
current study, one-third of the respondents had intentionally underreported
duty hours on at least one occasion, and only 1% reported the use of more
rigorous hour-log systems such as swipe cards. Despite this, more than
three-fourths of the respondents felt that their home institutions were
utilizing effective duty-hour measurement systems. The current regulations
allow the eighty hours of duty per week to be averaged over a four-week
period. Therefore, although residents on occasion may be on duty for longer
hours, their averages may fall within the compliance range. However, these
findings should point to the possibility that residents may not be comfortable
reporting excessive duty hours and that a number of orthopaedic training
programs are not yet in compliance with the eighty-hour rules. This survey
also raises concern that a detailed review of residency programs may reveal
duty-hour violations that could result in penalties or citations.
The third major conclusion that can be drawn from our data is that
orthopaedic residencies, similar to other training programs, are predominantly
using physician extenders, moonlighters, night-float systems, and increased
home-call assignments to comply with the new rules. Several other reports in
the literature have documented that surgical and nonsurgical disciplines alike
have utilized these same strategies to comply with the duty
rules8,10,14-16.
Although strategies such as physician's assistants and moonlighters may appear
to be obvious and easy solutions to the duty-hour rules, such ancillary
services carry with them a substantial cost burden that will be increasingly
difficult for many orthopaedic departments to absorb.
Another insight to be gained from the data in the present study is that,
among orthopaedic surgery subspecialty services, residents on an orthopaedic
trauma service experience the most difficulty in complying with the duty
rules. This is likely due to the unpredictable and complex nature of trauma
care at academic medical centers. Coordinating and administering the required
care for their patients often disrupts the ability of residents to participate
in the sign-out process and entices or compels the residents to remain in the
hospital after being on call. General surgery trauma services and
neurosurgical trauma services have also had similar difficulties complying
with duty
restrictions14,17.
The effects of duty-hour restrictions on continuity of care and, ultimately,
on patient safety are as yet unknown. However, these findings suggest that
some collaboration and cooperation among orthopaedic, general surgery, and
neurosurgical trauma services to devise efficient systems to care for the
complicated injuries of trauma patients may yield novel compliance
strategies.
The final major conclusion that can be drawn from the current study is
that, similar to other specialties, the effects of restricted duty hours on
the surgical and other educational experiences of orthopaedic residents are
currently unknown. In the present study, both junior and senior residents
indicated that approximately 70% of the average duty week consisted of
educationally valuable time. However, nearly a quarter of the orthopaedic
residents also felt that the new rules caused deficits in learning and
training experiences. The current literature contains many reports that
attempt to assess the effects of the duty rules on resident education. The
literature is mixed regarding changes imposed on surgical case volume by these
requirements18-20.
Reports have also shown that the performance of junior residents on
in-training examinations has improved after implementation of the work
rules21. Our study
does not resolve this important question, but it does provide substantially
more cross-sectional data than previous
reports10,22.
Further evaluation is needed to determine the effects of duty-hour
restrictions on resident education, both as demonstrated on examinations as
well as through the quality of patient care. It is conceivable that residents
of all specialties will have a decreased quantity of educational experiences
under the new rules; however, these same, better-rested, and more attentive
residents may have more quality experiences in conferences and during
operative cases. Future research is required to answer these questions more
definitively.
In summary, this report provides insight into the attitudes of orthopaedic
surgery residents toward the new duty rules. It also provides evidence that
many years and much additional research will be required to fully delineate
the myriad effects that the new ACGME rules will have on patient care, house
staff well-being, and residency training in the United States.
A copy of the survey that was distributed to orthopaedic residents is
available with the electronic versions of this article, on our web site at
(go to
the article citation and click on "Supplementary Material") and on
our quarterly CD-ROM (call our subscription department, at 781-449-9780, to
order the CD-ROM).
Note: The authors thank Sylvia I. Watkins-Castillo, PhD, and Jim
Frankowski of the Department of Research and Scientific Affairs as well as
Jeanie Kennedy, Regulatory Affairs Manager, of the American Academy of
Orthopaedic Surgery for their help in the preparation, dissemination,
analysis, and review of the data presented in this manuscript.