These two studies were interpreted by their authors as meaning that reductions in work hours did not clearly do harm.
As MedInformaticsMD noted in a an earlier post, however, a major limitation of the two studies was that they only assessed mortality. While mortality is obviously an extremely important outcome of hospitalization, it is not the only one. Fatigue is likely to cause many errors that lead to bad outcomes short of mortality. Yet the current studies were not designed to determine if the reduction in work hours could have improved patient outcomes other than survival.
Another issue is the nature of the 2003 work hour limitations:
No more than 80 hours per week, with 1 day in 7 free of all duties, averaged over 4 weeks; no more than 24 continuous hours with an additional 6 hours for education and transfer of care; in-house call no more frequently than every third night, and at least 10 hours of rest between duty periods.
Consider the lot of house-staff under these new rules. It is still perfectly legal for house-staff to:
- Work continuously for 30 hours (24 "work" hours plus 6 hours for "education and transfer of care")
- Work continuously for 30 hours, get 10 hours off, and then work for another 30 hours (see above, plus provision of "at least 10 hours of rest between duty periods)
- Work 24 days during a four-week, 28 day period
My still vivid memories of internship and residency were of being pretty fuzzy after working more than 16 hours, and of being a walking zombie after working for more than 24 hours.
And, as noted in the discussion section of the two articles, there is some evidence that limiting shifts to 16 hours might be better justified than allowing 24 or 30 hour shifts.
Yet the reformed rules still allow 24 hours of continuous, intense work, and then another 6 hours of "education" and for hand-off. Then imagine, after such a 30-hour experience, having all of 10 hours to sleep, eat, catch up on life, before it starts all over again....
So another obvious explanation for the failure of these two studies to find much effect of house-staff work hour reforms were that these reforms were woefully insufficient to prevent major sleep-deprivation and fatigue.
Another question is whether the reforms shifted work to those better able to perform it, or merely re-arrange the deck chairs? So who did the work that would have previously been done by house staff working more hours than are now allowed?
My guess is that most teaching hospitals did not rush out to hire more people to take on this responsibility. After all, the house-staff work hour reforms amounted to an "unfunded mandate." Nobody gave the hospitals more money after the reforms went into effect to hire people to do this work. (If I am wrong about this, someone please correct me.)
One possibility is that the same house-staff did the same amount of work, just in fewer hours, i.e., that the reforms resulted in greater intensity of work for house-staff on clinical rotations, and most likely less opportunity to sleep even a few hours when on-call. Another possibility is that the work was given to house-staff who previously were on lighter or elective rotations.
If the reforms did not lessen the overall workload of house-staff, it is doubtful that house-staff as a group are any less fatigued on average after the reforms.
Another possibility is that the work was given to already over-worked and increasingly scarce nurses, and to the faculty in charge of teaching the house-staff, thus reducing the time available to supervise house-staff and teach. These changes could have adverse effects on quality of care that might offset any small improvements made by slightly reducing house-staff fatigue and sleeplessness.
Because of extreme sleep deprivation and fatigue, my internship year was the worst year of my life so far. One reason I went into academic medicine was the hope that I could help reform the system to make it more humane for house-staff, and hence provide them better education, and patients better quality of care.
The gradual introduction of "night-float" systems into many internal and family medicine residencies did substantially reduce sleep deprivation and fatigue, although the work load shifts they caused lessened the possibility that house-staff would have any rotations that were not intense.
However, I am afraid there are still far to many "opportunities" for house-staff to be over-worked and sleep deprived.
In a two-trillion dollar plus US health care system, there ought to be some money to pay for the work that would allow house-staff to get sufficient sleep. But unless the money is forced to follow the mandate, to quote Over My Med Body,
interns on little sleep who’ve admitted patients all night, who are writing orders at 2 or 3 in the morning is, in the best of worlds, sub-optimal, and in the worst of worlds, dangerous.
ADDENDUM (6 September, 2007) - See also comments on the Medical Humanities Blog.
1. Volpp KG, Rosen AK, Rosenbaum PR, Romano PS, Even-Shosan O et al. Mortality among hospitalized Medicare beneficiaires in the first 2 years following ACGME resident duty hour reform. JAMA 2007; 298: 975-983. (link here)
2. Volpp KG, Rosen AK, Rosenbaum PR, Romano PS, Even-Shosan O et al. Mortality among patients in VA hospitals in the first 2 years following ACGME resident duty hour reform. JAMA 2007; 298: 984.992. (link here)