The New England Journal of Medicine reports that 9 out of 10 doctors agree that 1 out of 10 doctors is an idiot. - Jay Leno

When I first joined my Radiology group after leaving Stanford, where I was on faculty, they told me that private practice was very different. Being right was less important than making sure the patients and the referring doctors were happy. In academics, we frequently challenged the indications for the studies that were ordered, antagonizing the clinicians, inconveniencing the patients, slowing patient flow through the department, and wasting large amounts of time arguing on the phones. Of course, we were following "best practices", using our up-to-the-minute knowledge of the Radiology literature to justify the carnage we were causing.

The unique characteristics of academia make such behavior possible. The doctors on staff have no choice about what group of Radiologists to send their cases. The patients have been sent from other hospitals, and have very little choice about where else to go, and are unlikely to return anyway. The residents doing the actual work change every year, and the attending doctors are often only peripherally involved in the actual care of the patients.

This is the problem with academic physicians and consultants, including he authors of the book/article referenced, telling physicians in practice about "scientific medicine". Patients come to a doctor with problems, worries and expectations which they expect to be satisfied. If the patient leaves unhappy, they will not come back, and give poor reports to their friends, their insurance companies and the online reporting forums. Although most physicians have a pretty good idea about the literature and "scientific medicine", they quickly learn that a successful practice requires some balance between the dictates of the academics and reality. It does not do much good to practice "evidence based medicine" when you have no patients to practice it on.

A parent might want the non-generic antibiotic that worked for their child last time. A women might want yearly mammograms because their sister had breast cancer. The old man with a cough might or want a chest X-ray now because he feels the same way he did when he had pneumonia previously. Giving pain medicine early may prevent calls in the middle of the night. Saying "the literature does not support this treatment" has little resonance in such situations.

A good example is the patient coming in for back pain. Although the doctor knows the pain will eventually get better by itself, the patient is desperate to get better quickly. He says "Doc, if I don't get back to work soon I will get fired and lose my house and my marriage". He tells the story of his friend who went to Dr. Sellout. That doctor did an urgent MRI which found a herniated disc. The friend was immediately referred to the orthopedist Dr. Moneybags, who told him if he did not have surgery tomorrow, he would never walk again. The friend had the surgery and was back to work in two weeks and is now doing great. He says "Doc, if you don't get an MRI today, I am going to Dr. Sellout tomorrow". The doctor orders the MRI.