A great discussion. I have the utmost respect for both sides.
1. Each side has it's own concerns, and feels that their concerns are dominant to the other's. This is very similar to other contentious debates -- liberal vs conservative, abortion vs prolife, "tastes great" vs "less filling". The point is that, from the trenches it can be hard to see the other's viewpoint has merit.
2. There is clear evidence that sleep deprivation increases errors. This has been tested in multiple venues and is pretty well established. It has not been proven for surgery clearly. Two studies recently (
here and
here) looked at simulator functioning post call. These were simulator based, and hence it's possible that the functional deficits seen in the simulator are statistically significant but not clinically significant. Older studies have looked at laproscpic trainers with similar results. In fact, the only study I could find looking at clinical outcomes demonstrated
no difference.
3. On the other hand, surgery is somewhat different from other medical fields. If a patient who has been recently operated upon needs a second / repeat / rescue procedure, the surgeon that has recently operated on that patient MIGHT be the best person for the job.
Supporting this theory:
- This surgeon knows this patient's anatomy best.
- This surgeon knows exactly what happened in the OR the last time.
- This surgeon probably has an established relationship with the patient and family.
Opposing this theory:
- This surgeon might be too tired to perform at their best.
- This surgeon may have misinterpreted abnormal anatomy the first time. In a second operation, they may do so again. Sometimes a fresh set of eyes is the best medicine
- This surgeon may be less-than-fully-competent. If so, it will be hard to assess this unless another surgeon evaluates patients with adverse outcomes.
Whether or not the original surgeon is the best person to perform the second operation depends entirely on how you weigh the above factors (and I'm sure there are more that I have missed). For example, there may be other ways to pick up surgeons with skill issues -- surgical M&M, or some sort of standardized outcome assessments, for example.
4. The notion that you can learn to "deal with" sleep deprivation via practice is controversial. Most "sleep experts" feel this is simply not true -- much like people who drink before driving stating that because they do it all the time, they can drive fine while intoxicated. Most studies of this effect (none done in medicine, to my knowledge) do suggest that sleep deprivation causes a deterioration in cognitive functioning, which is physiologic and cannot be mitigated by practice.
5. However, WS does raise an important point. Some people likely need less sleep than others. Perhaps the average person needs 8 hours, but WS really needs 4.5. This raises a really thorny question: Perhaps we should limit surgery training to those who function well on less sleep. As WS points out, this happens somewhat already via "natural selection", but what if someone who needs 8 hours is simply "dying" to be a surgeon? Also, what happens if WS (or anyone else) finds that as they grow older, they can't do well on 4.5 hours. Is someone really watching?
6. The legal liability of sleep deprivation cannot be ignored. The example suggested by IUSM is completely germain to this discussion. Here's the scoop: if anyone (in any field) is up for 24-30 hours and then drives home, and gets in an accident, it is likely that both they and their emplyer will be liable for damages. There are several states that have laws specifically addressing MVA's and drivers with 24+ hours without sleep. More will follow. Even if surgeons are able to function well in the high stress venue of the OR after little sleep, they are almost certain to not do as well when driving a car tired.
7. As WS has mentioned, there is a flip side. Less hours working = less time in the OR. There is no way around that, other than lengthening training which seems impossible unless we change the way training works (i.e. you get a "junior position" after a number of years that pays much better than a resident, or some other similar change). Less experience might cause graduating surgeons to be less skilled. There is no evidence that this is true, but many senior surgeons fear it.
8. There is also a "resource utilization" issue here. Let's say you work in an inner city hospital. There are certain to be 2 major traumas every night. In such a case, it would make sense to have a trauma surgeon on call at night, doing shift work. It would be crazy to have someone be "on call", be certain that they would get no sleep, and then assume that they could work the next day without a problem. It would make more sense to work a week of nights, and have the rest of your schedule be empty. ON the other hand, if you work in a rural hospital where trauma's are uncommon, it would make no sense to have someone "in house, well rested, at the ready" all the time for something that rarely happens. But, when it does happen, you either need a contingency plan for the next day or just "live with it".
... unless there simply aren't enough trauma surgeons to go around. Sure, it would be great to have someone covering the night shifts, and the day shifts, but that assumes you have enough people to do it all, and you're willing to pay the cost.
So, where are we?
A. I do not doubt that more tired surgeons make more mistakes. In fact, there are several recent articles (here and here) that suggest this is true.
B. If we want less tired surgeons (because we think they give better care), then we will need to be willing to pay more for that privilege.
C. If we want less tired surgical trainees, then 1) we need to be aware that this may cause them to have less surgical skills upon graduation which could overwhelm any improvements from a lack of sleep deprivation, or 2) we need to make large changes to the way we train surgeons (by somehow getting them more OR experience while hours are reduced). I see no easy way to do that, unless teaching programs currently have many cases where only faculty scrub in.
D. If we do want to fix this, the problem is not the 80 hour work week. The problem is 24+6 shifts. We could consider limiting shift length, with 16 hours often suggested as the maximum. It might be possible for surgical trainees to work multiple 12-16 hour shifts weekly and still get 80 hours of experience and plenty of cases. Whether this will actually lead to more well rested surgical residents is not clear. If this results in more cases not having residents present (because their shifts are over and there is no way to cover all shifts with the residents available) then it will lead to less OR experience for residents. Note that increasing residency program sizes, although it will help cover shifts, will still dilute the training experience amongst more residents (again, unless there are a significant number of "resident-less" cases already)
Wow, that ended up being much longer than I expected. Looks like an IM admission note, doesn't it?
I'll stop now.
