Work Hours Restrictions

Every so often, the ponderings of the medical blogosphere (I hate that word!) turns to thoughts of residents/trainees work hours. Most recently, here and here. There usually seems to be a split between the people who are training now, and the people who trained in the "bad old days", with those who are currently in training saying, "Look, 24-30 hours at a time and 80-88 hours should be enough!" and the "old guys" saying, "Suck it up, it doesn't get any better, you need the experience, besides, we did it, so should you, it's good for you".

So, what are the rules these days?

1) Duty hours are defined as all clinical activity related to the training program (both inpatient and outpatient), administrative duties related to patient care, transfer of care, transfer of patient care, and academic requirements such as conferences. Duty hours do not include hours spent reading and studying away from the site in preparation.

2) Duty hours must be limited to 80 hours/week averaged over a four week period, inclusive of in-house call activity.

3) Trainees must be provided with 1 day in 7 free from all clinical and educational opportunities, averaged over a 4 week period. One day is defined as a 24 hour period free from all clinical, administrative, and educational responsibilities.

4) Time for adequate rest should be allowed between shifts. Trainees must be allowed a period of 10 hours free from clinical, administrative, and educational responsibilities between duty periods.

5) Excessive time spent on duty should be avoided. This means that trainees should work no more than 24 hours in a row, with an additional 6 hours for transfer of care activities. No new patient care should be assigned after 24 hours of continuous duty.

There's also a mechanism for programs to apply for an exemption for "sound educational reasons". (note: There's no reason to apply for an exemption for "sound economical reasons")

Now, truthfully, we have good evidence that acute, chronic, and especially acute on chronic sleep deprivation can impair people's performance on a variety of simple and complex tasks. We even have fairly good evidence that sleepy doctors, particularly sleepy residents are more likely to make mistakes. The Sleep Research Society has collected a handful of studies and citations on that very topic. Sadly, I'm at home, so I don't have my super-duper full-text access, but briefly, their review found evidence that sleepy trainees make 5 times as many diagnostic errors, suffer a 1.5 to 2 standard-deviation deterioration in skills level, and make fatigue related errors that result in patients' deaths. Their findngs are appropriately cited.

What has been more difficult to show has been that reducing work hours actually improves patient outcomes. The standard excuse has been that transfer of care of patients actually increases errors. While I don't doubt that that's true, or, at least, I don't doubt that it has the potential to do so, the fact is that eventually everybody must go home. Also, "signout", the transfer of information at the end of a shift, is a skill that can be taught. As is taking care of patients you're not entirely familiar with. Face it, when I was an intern in charge of 20 postpartum patients at night, it's not like I knew them all. I had to look at their charts to manage their complications. And let's face it, if you're a resident managing a team of 5-10 patients, you may know your patients well. But once you have more than 10 patients you're responsible for, your likelihood of really knowing the patient well is really small. Not only that, but the studies also haven't shown that resident work hour limitations actually endanger patients. One of the theories of Dr. Czeisler (a guy from Harvard who's done a lot of the studies on sleep and physician performance) is that the work-hours limitations are not enough. That you really have to get under 16-18 hours of continuous duty in order to reduce errors, so that we're not seeing the beneficial effects of well rested residents yet. Although people equate the lack of proof of improved patient outcomes with a proof of more errors as a result of turnover, there isn't evidence that resident work-hours restrictions has actually caused patient care to deteriorate, either.

I had a friend who was a general surgery resident. When she ran the Neurosurgical unit, she was in charge of 50+ patients on any one day. Did she know those patients? Not even a little bit. All she could do was to do her best to avoid drowning. She got to work every morning at 330 am, and on a good day, she left the hospital at 830pm.

Was she learning? Not really. She was dictating discharge/transfer summaries, and writing prescriptions all day long. She never entered the operating room, and any patient who got sick was seen by her 2nd year resident and transferred to the ICU. These were all patients waiting for various transfers to long-term care facilities. This was a terrible abuse of her time. And she's now an anesthesiologist because of that. (Those of you who are concerned about the recruitment and retention of general surgeons should think about this).

The Libby Zion case is often cited as a catalyst for work hours restrictions, the Libby Zion case was more a mistake of judgement rather than an exhausted resident. The truth is, hospitals like residents to work these extended hours doing scut work so that they don't have to hire nurse practitioners/midleves who actually need to be paid (Residents are paid by Medicare funds, and the hospital gets ~US$100,000 per resident, of which the resident gets ~$40,000; the rest goes to fund educational programs, malpractice insurance, and other costs of running a training program.) Hospitals may not be getting rich on resident labor, but, especially in large expensive cities, residents represent a huge cost savings when compared to what it would cost to hire phlebotomists, transporters (it still is not uncommon for residents to transport their patients for surgery or tests at the hospital I work at, and I view this as a HUGE abuse of their time), and nurse practitioners.

Another reason for reducing resident work hours from the traditional 100 hours+/week is that of safety of the actual residents. The Sleep Research Society in their summary includes the fact that tired residents are much more likely to have needlestick injuries and car accidents. OSHA makes rules to protect the health of workers all the time. Why isn't it valid to protect the health of residents? I certainly have given a lot of my life to medicine. I'm not willing to risk death at the wheel for the sake of not turning over patient care to another fellow.

Are there arguments for extended shifts? There is at least one reasonable educational argument. If you're not around for a long period of time, I think that it is true that you simply don't get the same appreciation for labor, for the course of disease (like a bowel obstruction) that you do if you follow the same patient for 24 hours. So, the 24 hour limit may end up being a reasonable compromise, long enough of a time to see disease/physiology evolve but not so long that the resident becomes a menace to himself and to others.

Having said all of this, I was a med student in pre-work hours restriction days, a resident in a very well-intentioned but not completely compliant program and now I'm a fellow in a less well-intentioned but more compliant program (for the residents); I'm at the point where I make my own decisions about how to structure my schedule to comply (or not) with the rules. And because of some health stuff, I mostly try to work less than 100 hours a week, no more than 24 hours at a time and take at least 1 day off each week. I don't have much to complain about. We were not compliant with the rules when I was a resident, but my program was cognizant of the need for rest, and when we were exhausted, we often got sent home (particularly if we were exhausted because we stayed late for interesting cases/continuity patients/etc.)

I will also say, that although some in the old-guard will rail against the new restrictions, they're in place because of examples like my friend above, where the old system was so abusive that it could no longer stand. In a way, the work hours restrictions that are in place now that people who trained in other eras are so against, are the result of the (formerly much more) abusive system that let work-hours be so abused in the first place. If attendings would have stuck up for their residents and said, "You know, my intern's too busy learning to be running my patients to x-ray, you need to hire some transporters" or said, "You know, my intern might be able to scrub on a case if she weren't writing discharge prescriptions for 50 patients, you need to hire a midlevel", things may not have changed this drastically. Things are how they are, and they are how they are for many reasons; and you know, I think that although there are still problems, it's a change for the better.

And for the record, I'll take the resident who's not so tired he's envying the patients under general anesthesia, and the resident who's not so tired he resents it when my nurse calls for an anti-nausea or pain med order over the resident who's been in the hospital with me every day and night for a week, if it ever comes to that. I also do not expect that resident to risk injury/death due to fatigue in order to "take care of me". Besides, what kind of a society are we when we expect a class of essentially indentured servants (admittedly, indentured servants who can expect a healthy reward at the end of their servitude) to take such risks on our behalfs?

(image credit)