Problematic proxies

Ezekiel Emanuel wrote about why our intuition regarding getting the "best" doctors (specifically cardiologists) might be wrong. Apparently, research published in JAMA Internal Medicine found that certain truly sick cardiac patients fared better when senior cardiologists were not present. Emanuel speculated why this might be the case (research ability vs. clinical ability, or more unnecessary interventions), but the cause is unclear.

This is clearly a counter-intuitive finding: if your uncle were undergoing a heart-attack, would you rather have him treated by a cardiologist with decades of experience or one recently out of residency? I haven't read the actual papers themselves, but the conclusion points to why proxies are limited in their usefulness. We assume that patients will fare better under the care of experienced physicians, and therefore, we might consciously look for experienced physicians. However, it's far more informative when prospective patients can actually see risk-adjusted outcomes over a certain volume of procedures; we simply rely on experience (and other proxies) because we can't get access to the data that is closer to what we are looking for. The medical industry has been reluctant to release this type of information at the individual physician level (for some good reasons, and some bad reasons). It'll be interesting to see over the next decade if attitudes towards this type of information changes, especially since in some cases, it can literally be a matter of life or death.