I enjoyed Dr. Philip Mohler’s article “New Drugs: How to Decide Which Ones to Prescribe” [June 2006]. However, it is missing two key pieces of advice. First, a new drug should not only be at least as effective as the older drug; its efficacy must also be measured using outcomes that actually matter to patients. All too often a drug is promoted because it provides a small benefit as measured by a disease-oriented outcome such as the effect on blood sugar or peak flow. What really matters is patient-oriented outcomes – the effect on morbidity, mortality, symptoms, cost and quality of life.

Second, Dr. Mohler includes colleagues as “unbiased” sources of information. Sadly, given the pervasive reach of pharmaceutical representatives and drug company advertising, they may not be unbiased. Also, the informal and often haphazard observations of an individual physician are fraught with unintentional bias and are no substitute for a careful, well-designed study.

Instead, I encourage readers to read the STEPS (Safety, Tolerability, Efficacy, Price, Simplicity) feature in American Family Physician, where every month we carefully and objectively review a new drug using just the criteria that Dr. Mohler proposes (go to https://www.aafp.org/afp/steps).

Author’s response:

Dr. Ebell’s thoughts are germane and appreciated. Unfortunately, surrogate outcomes and comparisons with only placebo are the (sub) standard mode of bringing drugs to market. As he posits, pharmaceutical marketing often highlights the statistical significance of outcomes that have little, if any, clinical utility.

Dr. Ebell’s comments regarding our physician colleagues’ lack of unbiased medication knowledge are on target. The pharmaceutical industry, with its glitzy pens, pads and pizza, remains the primary source of drug information for most physicians. Dr. Bob Goodman’s Web site, http://www.nofreelunch.org, documents the very powerful bias that physicians expose themselves to when they see drug reps and accept “free” sample medications.