Thanks for this very stimulating post.
You wrote, "Also, perhaps over time there will be selection into the health work force of individuals more responsive to extrinsic motivation and more resilient to lower perceived autonomy in the work setting."
This is a really interesting point, but it seems to touch on an area that opponents of extrinsic incentives (in health, or education in the U.S. debate for example) are concerned about.
The outcomes incentivized are subject to a measurement problem as in all principal-agent settings. There might be adverse selection into these jobs (of those who are not motivated for the "right reasons"). These may be individuals who are more willing to game the system; more subject to multitasking problems (will only work on the rewarded measures because they don't have a strong intrinsic incentive to work on other areas).
This suggests that the planned RCTs of these schemes should gather detailed information on what kinds of people enter and exit the jobs. This is in addition to other longer-term effects like learning to game the incentive.
I look forward to hearing more about ongoing work in this area.