It's not clear to me what the point of this piece is. The author clearly had bad experiences with "night float", but it's not clear to me that was the problem. The hand-off doctor failed to mention an abnormal brain scan and somehow divined that the patient would be okay...Hardly seems like an effective hand-off to me. The author mentions better hand-off practices, but doesn't really go into how well they are implemented and whether they work. Also, I assume cross-coverage means that there is an overlap of time between the day and night shifts, though this is not explicitly stated. So is the author advocating further study into how to reduce errors? Or is the point that we should go back to sleep-deprived interns and residents roaming the halls? One thing that isn't mentioned is that the danger of long shifts extends outside the hospital. What are the statistics on car crashes caused by doctors who're just coming off 24 hour shifts or who have accumulated sleep debt over their 80 hour weeks?

I think it is worthwhile to discuss the ramifications of such a radical policy change as reducing the length of shifts and the total working hours by almost 20%. But I think we can do better than anecdotal stories and a brief paragraph on two existing studies.

What are the areas we should look at? The thing that stood out to me was how totally inadequate the hand-off procedures were. How do you improve that? Electronic records are expensive, take a while to learn to use, and in some cases can distract the practitioner from really picking up on the important info. Case in point, I went to a PA for a minor problem. She asked me all the "history" questions b/c her form asked me to, including whether I was on any current medications. I was taking a medication with possible side-effects, though she had never heard of it before. Though we discussed these side-effects, she got so caught up in finding it on her pull-down menu, she apparently forgot. Not 5 minutes later she told me she wanted to prescribe me medication from the same drug family. *I asked her* whether that was safe given that I was already taking a related drug. Ultimately she looked into it and decided against it. My point is that just because she took the history electronically doesn't mean it registered with her.

I'm just saying, I have to believe there's more information to draw upon than a few isolated experiences of one individual. Is that too much to ask?

I'm just saying, I have to believe there's more information to draw upon than a few isolated experiences of one individual.

As opposed to your opinion, which is based on the medical experience of zero people.

Cross-coverage is when you are covering another doctor's patient, someone who is not "your" patient. My husband is an intern at Vanderbilt, and they seem to have very fine electronic systems in place. As the article suggests with "electronic handcuffs", my husband rolls his pager to the incoming doctor or on-call team when he leaves the hospital, and all of the notes are on an electronic database, so that a scribbled-in note like the one mentioned in the article wouldn't occur.

What surprises me is that entire shifts change over at once, as if they were on an 24-hour assembly line. I'd expect a staggered schedule, more like what you find in retail, where different people start and finish at different times. This way there is significant overlap, and the doctors and nurses have time to get to know people on different shifts and talk about things beyond what gets put in the official records, and people don't get so attached to start and end times.

I'd hate to have a major complication at shift change.

One thing that I've always found odd about the discussion of shorter working hours for residents. The argument against it is that the same doctor deals with the same patient over time. But, the resident has to leave the hospital some time. He or she isn't on duty 24/7. So if a resident works an 18 hours shift, the patient could have complications in the 19th hour. If the resident works a 24 hour shift, the patient could have complications in the 25th hour.

Most seriously ill patients are going to stay in the hospital for a longer period than a single shift of a resident, no matter whether it's an 8 hour shift or 18 hours or 24 hours. So handoffs are going to happen at some point. It seems like it's important to figure out how to handle that in the best way.

To echo what some of the other commenters have said this piece seems to have no real point. Even though handoffs are inevitable and evidence has shown sleep deprived doctors are terrible, this writer seems to hint that we should go back to the old system because he and some of his colleagues felt ignorant. I would hint that the problem lies in him. To some extent, improving handoff systems is a nonbrainer but at the end of the day, he needs to step up to the plate and learn about the patients he is given, end of story. The rest of us do it (I am on night float right now) and don't whine about how in the old days it was better because we got less sleep and made more mistakes on the patients but hey, we gotta to play hero...

If this encourages more shift work mentality amongst doctors, I think it might also reduce the arrogance amongst us to a more tolerable extent as well..

You do have a point there, but the author is also correct that sometimes seeing a 24-hour cycle of progression is a very useful thing - whether from the standpoint of learning how a disease and treatment work, or from that of gathering enough information to make a correct diagnosis. There's also something to be said for the experience gained by being with a patient through the night - it's invariably 2 a.m. or thereabouts when the truly scary stuff goes down.

Ultimately, you're right that from a patient-safety standpoint, the real issue is better handoffs. I can't remember any work environment in which I received as slipshod a check-out from the person leaving duty. And while some of that is probably institutionally-based and varies from hospital to hospital, the author has to accept some blame for it - the best way to get information that might be useful later is to demand it. I learned this lesson very early in residency, as do most residents - you might annoy people by being the guy who asks too many questions when someone else is trying to get home, but asking the right questions can avoid a lot of shitstorms later in the night.

As you say, the problem of sleepy doctors does extend beyond the hospital. The stats of sleepy docs causing car crashes on the way home are not good.

<link>