The bulk of the data shows that errors due to sleep deprivation are less than those due to context switching to a new doctor. This is the reason for the intense schedules, not sadism. If you're aware of studies that dispute that I'd be interested in reading them.

The other day, I was speaking to a doctor friend who's in the middle of sleep-deprived training, and she defended the practice.

She said that even as a senior doctor, you are going to have to pull insanely long shifts at times. Under those conditions, you're clearly not going to be working at your best. So you're going to have to develop good instincts; more importantly, you're going to have develop trust in your own instincts, that will allow you to operate effectively despite ridiculous levels of sleep deprivation.

And the only way to do that is to go through that in a semi-controlled environment in your training.

Says my friend the doctor anyway.