So I’ve been seeing headlines like this today: Physicians Recommend Different Treatments for Patients Than They Choose for Themselves, Study Finds. Here are the numbers:

A total of 242 physicians returned the colon cancer questionnaire (response rate of 48.4 percent), and when asked to imagine they had received the cancer diagnosis, 37.8 percent of physicians chose the surgical procedure with a higher rate of death, but a lower rate of adverse effects. Conversely, when asked to make a recommendation for a patient, only 24.5 percent of physicians chose this option.

The second scenario asked 1,600 physicians to imagine that a new strain of avian influenza had just arrived in the U.S. One group of physicians were asked to imagine they had been infected, and the other group was asked to imagine that his or her patient was infected. One treatment was available for this strain of influenza: an immunoglobulin treatment, without which persons who contract flu have a 10 percent death rate and a 30 percent hospitalization rate with an average stay of one week. The treatment would reduce the rate of adverse events by half, however it also causes death in 1 percent of patients and permanent neurological paralysis in ...