Harvard Work Hours Health and Safety Group
A Scheduling Toolkit for Medical Professionals

HOME          MANUSCRIPTS         SAFETY RESOURCES        SLEEP RESOURCES         REPORT AN INCIDENT        CONTACT US
 
 
The Harvard Work Hours Health and Safety Group Intern ICU Intervention Schedule
Despite ACGME regulations limiting resident work hours to less than 80 hours per week and less than 30 consecutive hours in a single shift, few rigorously tested work hour reduction strategies exist.  In 2001, the HWHHSG received grants from AHRQ and NIOSH to conduct a demonstration project for interns working in critical care units at Brigham and Women’s Hospital in Boston. We tested the effect on intern sleep and patient safety of an intervention schedule that limited interns’ work to ≤16 consecutive hours, and reduced weekly work hours. The schedule traditionally employed in the BWH ICU’s is a “Q3” schedule (Schedule A), where residents alternate between a swing shift of about 10 hours (from about 7:00 AM until 5:00 PM) and a call shift of about 30 hours (from about 7:00 AM until 12:00 the following day).  The intervention schedule (Schedule B) employed a rotation where the call shift was split in two and residents rotated from a swing shift, to a day call shift of about 7:00 AM until 10:00 PM, to a night call shift of about 9:00 PM until 1:00 PM the following day. Senior housestaff remained on the traditional schedule and the culture of each ICU was preserved as much as possible.  For example, morning rounds, training, and admitting responsibilities were the same on the intervention schedule as on the traditional schedule. 
Raster plot of the q3 schedule and a modified schedule
*Note these schedules are presented for informational purposes only and the presentation of these schedules should not be considered an endorsement for implementation. Please contact us for more information.

In addition to limiting work shifts to 16 hours, Schedule B has several advantages from a sleep and circadian standpoint:

  1. Following the day call shift, residents are not required to return to work until late the following evening.  This allows the resident to remain on duty when necessary for a bit beyond their scheduled shift, yet still get adequate sleep upon returning home.  Also, with a shift limit set well below currently maximal shift lengths mandated by accrediting bodies, there is no concern with violating work hour standards when shifts need to be extended a bit for emergencies, or in the interest of assuring continuity of care.
  2. We encouraged residents to nap prior to reporting for the night call shift, which preemptively attenuated sleep pressure at the circadian nadir.  Most residents were able to do so.
  3. Residents slept less on duty during the nights they were working on the intervention schedule.  Having less need to sleep on duty was a consequence of an improved schedule, and is likely to improve safety by reducing the occurrence of sleep inertia experienced when awakening from deep sleep.  Sleep inertia can greatly increase the risk of provider error in the minutes after awakening while on duty.

Outcome measures for this intervention included intern sleep, fatigue and medical error rates which were compared using a randomized experimental design. Sleep and work hours were rigorously documented using daily logs validated by use of electroencephalography (EEG) and third-party confirmation of work hours. Errors were detected using a four-pronged data collection methodology that included trained physician observers continuously monitoring the performance of the interns around the clock, as well as daily medical record review. All suspected errors were rated by two independent reviewers blinded as to study condition. The intervention schedule was found to reduce weekly work hours from 85 to 65 per week, increase sleep duration by approximately one hour per night, and significantly reduce (by more than half) the occurrence of objectively documented attentional failures during night work hours. With respect to patient safety, interns made 35.9% more serious medical errors and 5.6 times as many serious diagnostic errors on the traditional schedule vs. the intervention schedule (Landrigan 2004).

The primary outcomes of this intervention schedule were very positive, but evidence of chronic sleep debt remained even in the intervention group.  Despite the fact that the extended work shift was split in half, most work shifts remained long enough to induce significant decrements in neurobehavioral performance owing to sleep deprivation (Cajochen 1999, Van Dongen 2003) It is important to recognize that a 16-hour shift limit and a 65-hour work week still exceed the limits imposed by many other safety-sensitive industries, such as transportation and nuclear power, on hours of work. Moreover, the interns often had to rise between 4 a.m. and 6 a.m., the time of maximal sleep propensity and efficiency in this age group (Dijk 1994), to review their patients' progress before morning rounds. Since nearly a third of their work hours (31 percent) were thus preceded by 6 or fewer hours of sleep in the preceding 24 hours, they continued to carry a substantial sleep debt, accounting for the high residual rate of attentional failures on both schedules, even during the day (Howard 2002). Furthermore, during both the traditional and the intervention schedule, actual work hours often exceeded scheduled work hours, owing to the interns' obligation to ensure the continued care of their patients after their own shift was over. Our data on actual work hours reveal that the maximal number of scheduled work hours must be much lower than mandated limits to allow for this inevitability.   

In addition, several cultural issues arose during our study.  Although our intervention decreased the rate of serious errors overall, our efforts to optimize the evening sign-out process were only partially successful.  Some groups of residents worked successfully as teams, while others did not.  Further, because senior residents remained on the traditional Q3 schedule while interns worked the intervention schedule, the usual mentoring bonds between the on-call resident and intern who work the nights together was impaired on the intervention schedule.  This may have interfered both with the satisfaction and learning experiences of the interns and residents.  To address these issues, we are currently implementing a revised intervention schedule.