Yes, it is hard to believe - sort of like the "open offices improve collaboration, it's totally not about saving money" thing.

Here's are a couple of data points though:

http://www.npr.org/sections/health-shots/2016/02/02/46524860...

Here's the original source: http://www.nejm.org/doi/full/10.1056/NEJMoa1515724?query=fea...

I think they're on call for 30 hours. I don't think they literally stay up for 30 hours routinely. They have a mess where they can sleep.

Being on call for 30 hours might as well be the same. All it takes is one patient that has a few emergencies in a short time frame for the sleep to get messed up. We have better regulations for train engineers and conductors - all for safety. My uncles do it, and my brother did for a while. Always on call, but once their shift starts, there is a strict limit on how long their shift is and a strict limit to how long they have free to sleep.

there have definitely been cold, calculating, studies done on the cost/benefit of doctor sleep to shift changeover ratios, and I don't know if it's 30 hours, but i do recall that it does not comport to your 'feel good' notions of how much a doctor should sleep.

Yeah, and when we had a newborn, we could "sleep when she sleeps". Guess how well we felt, and how capable we were of anything but the most routine tasks?

I would certainly not want my doctor or surgeon to be as “alert” as a new parent..

Just about the time I graduated (BS, not MD), the hospitals in Dunedin NZ had junior doctors on shifts of that long as a bare minimum, and they were on call for 72 hours following their shifts. They were supposed to get a few days off after that.

The 30-hour on-call shifts did start out as just on-call. However, it has long since become the case that it was effectively 30-hour straight shifts, with maybe an opportunity to catch a 20min nap.

Source: Had friends in Med School some years ago.