Four points that I'd make on this subject. The first is that 16 hrs is too long of a shift to be working throughout, as is 24 hrs. What I'd tell you is that hardly ever would a doctor be working during the entirety of either of those shifts. In fact, in many cases the doc would have more of a chance to get a nap in on the 24 hr shift than the 16 hr. Basically, I'm suggesting (with zero evidence except my own experience), that I'm not sure there's a significant difference between 16 and 24 hr shifts' effects on a doc's ability to function well--both can be fine under the right circumstances and both can be equally terrible. Also related to this is the fact that nurses, doctors, respiratory therapists work as a team in hospitals ... most critical decisions are checked on a number of levels.
Secondly, do not underestimate the danger of hand-offs. This is the main reason that I'm in favor of this change. Any time there is major information exchange in a hospital environment, there is increased room for error due to misunderstandings, forgotten information, etc. There have been many efforts to standardize information exchanges and they have become better in many circumstances. But I'll tell you that, in my experience, more hand-offs increased the risk of errors in patients and if I were in the hospital on a medical floor, I'd prefer single physicians for 24 hr shifts rather than 2 or 3 handing off to each other per 24 hr shift.
The third point is about how healthcare functions in the country, specifically around large hospitals. Residents and fellows get their training in these places, but hospitals also rely on these trainees to staff their enormous floors. If we were to change the shifts, let's say to a maximum of 12 hrs each, we would likely need significantly more residents in the system overall. Not only does that volume need to come from increased sizes in medical school classes, but the hospital systems would need to be paying the salaries of these trainees. Hospitals with trainees receive some amount of this salary money from the federal government, and some comes directly from the hospital bottom line. Either way, the total cost burden to the system increases, meaning that we will all be paying more in an already overpriced system.
The fourth point is that these work hours rules truly impact the training of residents. There are surgeons graduating after 5 years of ENT training who have not had enough experience to become attendings at the places where they are interviewing. I have spoken with a number of senior attendings about this issue and they all say that they are more and more concerned about whether the trainees coming through these programs are getting enough training. So the option to deal with that would probably be to add on another year of training to fields where competency is not being reached, again increasing the cost to the system and to the overall financial picture of the trainees (many are barely making the equivalent of minimum wage during training and need to start paying on a quarter million USD worth of debt).
Basically, I'm not disputing your concerns, just pointing out that our system will likely cost much more if we move the other direction with hrs rules. I also would not jump to the conclusion that 24hrs is truly worse than 16hrs for patient outcomes. Both are terrible, and maybe not significantly different.