In Medicine, Don't Believe Everything You Know

By Eve Harris

How do you know your doctor is right? Ideally you and your doctor have a relationship based on trust. That is, you believe she knows the best options to recommend to you. You may think your doctor is right, but -- how does your doctor know she's right? We'd like to think physicians are relying on the latest evidence of medical practice. But not all physicians do that.

I recently joined in a robust, four-day discussion designed to address this issue at the 14th Rocky Mountain Workshop on How to Practice Evidence-Based Health Care. Doctors, policy makers and yes, journalists gathered to explore what many patients might have thought they were already getting: evidence-based health care, also called evidence based medicine.

In evidence based medicine, a hierarchy of evidence guides decisions about patient care. But at the same time, evidence based medicine recognizes that evidence alone is not sufficient. That's because treatment options come with risks, and different patients will react differently to different risks. It's not a simple matter of "Drug X" or "Treatment Y" has a five percent higher likelihood of success. If "Treatment Y" involves a risk or side effect a patient finds unacceptable, then this patient's preference is part of the decision process.

Decision makers must always acknowledge these trade offs.

The emphasis on patients’ considerations makes evidence based medicine a patient-centered practice, ideal for the coming age of healthcare reform. It’s imperative that both providers and patients learn to properly evaluate their sources of medical information. Even tougher -- old habits die hard, as we all know. Earlier this year in the Journal of the American Medical Association, Stanford professor John Ioannidis (and others) wrote of just how hard it is to reverse established medical practice:

Ideally, good medical practices are replaced by better ones, based on robust comparative trials in which new interventions outperform older ones and establish new standards of care. Often, however … what was thought to be beneficial was not. In these cases, it becomes apparent that clinicians, encouraged by professional societies and guidelines, have been using medications, procedures, or preventive measures in vain.

Factors beyond science -- including bias -- can alter the outcome of a study. They can even affect what research is conducted and published. Ioannidis is well known for exposing ways this happens and why it is harmful to patients and physicians.

One take-away from the workshop I particularly endorse came from the U.S. Preventive Services Task Force chair Virginia Moyer: researchers should stop designing research around outcomes that don't matter to patients! One example we discussed was bone density. What patients care about is preventing fractures. But studies looked at bone density, believing bone density was linked to fractures. Turns out it's not. In this case, studies that look at fractures are what's needed.

But overall it appears that patient perspectives and values are increasingly part of research design. The Patient Centered Outcomes Research Institute (PCORI) is funding innovative new studies. In a recent article, PCORI explains "Why Methods Matter." Today, patients have more treatment options than ever, and changes in the way we receive our healthcare today – shorter hospital stays, for example -- may affect outcomes.  So although difficult, it’s crucial for patients and providers to understand and use the most relevant information when making decisions.

Doctors and patients are making progress in defining and trying out new ways of communicating about treatment decisions. Achieving the best outcomes via this newer, more participatory practice of medicine requires attention not only to the evidence provided by high quality studies but also clinical judgment and the bottom line: patient values.

Eve Harris is a Bay Area writer. Check out her health blog, A Healthy Piece of My Mind.