I often write about biases.  Many studies have exposed the biases of guidelines.  This study exposes the biases in meta-analysis – The interpretation of systematic reviews with meta-analyses: an objective or subjective process?

The interpretation of systematic reviews with meta-analyses is at least partially subjective. Evidence-based practitioners need to be aware that any conclusions and recommendations based on a systematic review with a meta-analysis should be read with caution even if the methodology is rigorous.

We champion evidenced based medicine, because the name has such face validity.  As a skeptic, I remain reserved whenever a new meta-analysis proclaims that we have evidence to do or not do something.  This article suggests strongly that we should be careful that meta-analyses do not have the same biases that we now know guidelines have.

We try to make decisions based on evidence, but we should not become enthralled with evidence based pronouncements unless that fit our previous understanding of the world.

Medicine remains an art informed by scientific data.  The bean counters want to make medicine a pure science, but medicine can never be a pure science.  Patients have too many variables to fit the evidence.  The evidence allows us to make informed decisions, but we must always apply the evidence to the patient, not to the disease.

The diabetes brouhaha that Dinosaur created represents a great example of this problem  Intelligent, well intentioned scientists can disagree about the management of a diabetic patient with a HgbA1c of 7.4% who is already taking 8 medications.  Those who insist that we should always treat a number without understanding the patient’s context is demonstrating a pedantic attitude.