How much of medicine is evidence-based?

The Atlantic published a lengthy article examining the issue of over-treatment and the use of ineffective procedures. The piece cited research by Dr. David Brown that found that "stents for stable patients prevent zero heart attacks and extend the lives of patients a grand total of not at all." Interestingly, the article also cites another study asking cardiologists why they would send a patient to get a stent even if the cardiologists knew of data that suggested that such a procedure would not be helpful for that particular patient. Some responded that they wanted to avoid potential legal action while others wanted to relieve patient anxiety. Some simply did not believe the data given how much intuitive sense the procedure seems to make. The article also discusses how certain medications may improve certain measurements typically associated with better health outcomes (e.g. lowering blood pressure in hypertensive patients), but not actually improve health outcomes.

Medicine is complex, science is complex, and health policy is complex. Throwing financial incentives in the mix (e.g. pay-for-procedure) can make things even more complex.