Physicians need to be better judges of science

acp new logoA guest column by the American College of Physicians, exclusive to

The aim of medical education is not only to transfer scientific knowledge, but also impart the wisdom accumulated over the long history of medicine that equips us to apply this knowledge for the betterment of our patients.

Although much of medical wisdom transmitted in pre-scientific times was based on ideas that were thought logical and therefore likely effective (think bloodletting), we now have a tremendous amount of information to guide us in caring for patients.  However, I’ve always been impressed with how much we still don’t know about routine patient care, the way we compensate for these gaps in knowledge, and how the resulting dogmas that guide much of our practice are perpetuated over time even when new data contradicts them.

As an example, I was taught that recommending avoidance of dietary nuts and seeds was an essential preventive measure for those with diverticular disease, based on the logical assumption that such small objects could become lodged in diverticula causing inflammation or bleeding.  This has since been disproven, ironically with an inverse association between nut and popcorn consumption and risk of complications being shown.  Yet to this day, many of my patients still report receiving this advice from their caregivers.

Another, more personal example was the advice for patients with supraventricular tachycardia (SVT) to avoid caffeine intake based on the assumption that its stimulant properties could trigger or worsen their arrhythmia.   However, it has been shown that moderate caffeine intake in those with SVT is not associated with significant electrophysiological cardiac effects.  The personal aspect of this was that my father’s cardiologist ordered strict caffeine avoidance for his rate-controlled and anticoagulated atrial fibrillation long after this information was known.  So he spent his last years dutifully avoiding coffee or tea, several of the true pleasures he had enjoyed over the course of his long life.

However, beyond lifestyle considerations, many of these persistent teachings can pose a real risk to patients.  An example is the commonly related rule-of-thumb that if a patient requires transfusion of a unit of blood, that two units should be given at a minimum.  This historical practice (which I was also taught) was predicated on the idea that a more liberal transfusion approach would improve patient outcomes by providing more than the minimal oxygen-carrying capacity needed by a sick patient.  However, despite the now well-established benefits of a more restrictive approach to transfusion, this teaching and practice persist, exposing patients to potential harms without clear benefit.

Our tendency to approach our gaps in knowledge in this way is captured nicely in the concept of the medical ignorome (ignorance + -ome) as outlined by Dr. Tom Finucane.  He defines the medical ignorome as the space within medicine where there are certain things associated with caring for patients that are not known to us, with the result that we develop theories and explanations based on the things we do know and base our clinical practice upon them.  However, despite the assumed logic that may underlie them, they often lead to the transmission of bad ideas and often become deeply embedded in the culture and routine practice of medicine as they are passed down through generations of physicians.  They, therefore, become very difficult to overcome even in the light of new evidence being available to inform our practice.

A related concept is “medical reversal” as elucidated by Dr. Vinay Prasad and others, when a current medical practice is found to be inferior to a lesser or prior standard of care, usually occurring when newer studies than those upon which an intervention was initially based contradict current practice. It resembles the medical ignorome in that we tend to implement certain practices with inadequate evidence of their effectiveness or safety.   For anyone who has been in practice for a while, it is clear that medical reversals are actually quite common.  I have personally advised women to take postmenopausal estrogens for cardiovascular protection, switched many of my type 2 diabetic patients to first-generation thiazolidinediones, and liberally prescribed rofecoxib (Vioxx), all practices that have since been abandoned due to their ineffectiveness or outright risk to patients.  Much as with the medical ignorome, once these practices have become established, they are difficult to change; deadoption is always slower than adoption, with some estimates suggesting it may take ten years or more to significantly change entrenched medical practice.

So why do we do these things?  I believe it’s because doctors truly want to do what is best for patients, and when we encounter clinical situations in which our knowledge of what works is limited, we turn to our best understanding of physiology and pathophysiology to develop explanations that make sense to us and act upon them, believing that doing so will benefit our patients.  Similarly, when a new drug or treatment approach comes along, we often jump at the chance to apply them to our patients with the hope that the latest scientific innovation will improve their health outcomes.  Subsequently, through a combination of expert opinion, professional and patient expectations, and (unfortunately) sometimes financial factors, they become embedded in our medical culture regardless of their true effect on patients.

Unfortunately, basing our practice on unproven assumptions or intervening with treatments with limited supporting evidence comes with a huge cost by jeopardizing our patients’ health, wasting valuable resources, and potentially undermining the public trust in medicine.  Patients expect us to be either correct or at least transparent about uncertainty, and it is always a very uncomfortable conversation telling someone that they need to change a treatment that I previously assured them was necessary and safe – “seesaw science” is frustrating for all.

What can we do to address these problems?  Here are some thoughts:

We need to take a more deliberate approach to practice by thinking about why we do the things we do routinely and examine the evidence base upon which we do them.

We need to be better judges of science.  When we intervene, we need to be sure that the science backs us up.  Our interventions should be based on our understanding of the evidence, and not solely on expert opinion or the influence of others.

We need to be healthily skeptical by understanding that despite the remarkable promise of advancing medical science, not every new thing is necessarily a better thing.

And perhaps most importantly, we need to be honest with ourselves and our patients about what we know and what we do not, and embrace the uncertainty inherent in medical practice.  Only then can we have meaningful conversations with our patients about the best things to do (or not do) for them.

To paraphrase Confucius, if “real knowledge is to know the extent of one’s ignorance,” then we need to better understand the medical ignorome and how it influences our individual practice.

Philip A. Masters is vice-president, Membership and International Programs, American College of Physicians. His statements do not necessarily reflect official policies of ACP.

Image credit:

View 3 Comments >

Most Popular

✓ Join 150,000+ subscribers
✓ Get KevinMD's most popular stories