> Improving the handoff seems like a much more tractable problem

You'd think, but it's a lot harder than it looks, and I'm speaking from experience.

In the meantime, they're going back to a solution that they know works, because in this case it's literally what they used to do before and what they already do for all other residents, and it has a lower failure rate.

All they're doing here is reverting a failed experiment in response to the measurable, documented problems it introduced.

From other comments here, it sounds like there really hasn't been a lot of experimentation or innovation in the "sign-off" process. It seems like things like body cameras and audio recordings could help with the transfer of knowledge between two people about a specific patient. So could telepresence, IMO, if the off shift doctors weren't trying to recover for their next marathon.

> a solution that they know works

For what definition of works? My (thankfully few) trips to the hospital have not exactly inspired a lot of confidence in the hospital system. Duplicated questions, minimized interaction with an actual doctor, being kept in the dark about what's going on, waiting for hours between human interactions, empty IV bags which hang for hours... This is questionable definition of "works".

But it goes against my uneducated intuitive beliefs! What are they thinking?