Medicine by the numbers? Not always

Our medical group has endless charts and graphs of physician "performance," whether it be percentage of flu shots given, percentage of smokers counseled, percentage of cholesterol readings of less than 200 in patients, or number of patients whose heights are measured at least once a year. A physician can get a "grade" on more than 50 measures, and his or her yearly evaluation is largely based on the results. Even our annual quality bonus, initially intended to measure the quality of our work, now measures our numerical performances on several scales.

Drug companies know how much of one drug we use, and how much of another. They base their marketing to us on these profiles.

THE TRUE MEASURE OF A GOOD DOCTOR

But this is not the way we learned to evaluate our colleagues, nor is it the way we continue to evaluate them. We look at the font of knowledge, work ethic, coping skills, collegiality, compassion, and caring.

In our training, our mentors were not the doctors who ordered the most prostate-specific antigen tests, but the ones who did the best work, the ones who cared most about their patients. That's how I evaluated physicians in 1975, and that's how I evaluate them in 2012.

But our informatics colleagues even have an answer for that. They say that performance on these numerical measures is a "marker" for overall performance. For as many times as I have heard that said, no one has ever been able to give me any proof of that statement. I don't believe it, and I bet you don't either.

My belief is that there are some great doctors out there with poor mammogram rates and that there are some whose care I would find not as good whose mammogram rates are high. I doubt there is a true correlation.