Should Evidence-Based Medicine Be Replaced by Interpersonal Medicine?

The 22 November 2018 issue of The New England Journal of Medicine carried an opinion piece by Stacey Chang, M.S. and Thomas H. Lee, M.D. titled “Beyond Evidence-Based Medicine.” They point out what they see as the limitations of EBM and recommend an improved approach that they call interpersonal medicine. In my opinion, they are misguided.

The limitations of EBM

In 1996, Sackett et al. defined EBM as “the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients”. That is an admirable and unobjectionable goal, but it has been distorted in practice. According to Chang and Lee, “guidelines based on clinical research are being hard-wired into our operational norms, incentive programs, and information systems”. The results of clinical research are being converted into a one-size-fits-all cookbook. That was never the intention.

Note the use of the words “judicious” and “individual patients”. Most published guidelines make it clear that they are not intended to be applied to every patient as a knee-jerk prescription. They are meant to guide physicians, not to dictate to them. The cholesterol guidelines emphasize that lifestyle modifications are crucial, and that “additional factors may be considered to inform treatment decision making”. The guidelines emphasize that prior to initiating statin therapy, the clinician and the patient should have a discussion covering:

1. The potential for ASCVD risk-reduction benefits
2. Potential for adverse effects and drug-drug-interactions
3. Heart-healthy lifestyle
4. Management of other risk factors
5. Patient preferences
6. Other factors such as family history

So obviously the limitations they describe are not inherent in EBM but are features of how the guidelines are mis-interpreted by some clinicians and regulators.

They say that some quality measures have already been retired because compliance with guidelines is uniformly high, but they contradict themselves by making the claim that “even when physicians prescribe medications that have been proved beneficial in randomized trials, the chances that patients are taking them a year later are akin, at best, to a coin toss.” They offer no supporting reference for that claim, nor do they offer evidence that stopping the medications has an adverse impact on clinical outcomes. There could be many reasons for stopping the medications, from side effects to cost and availability.

They say EBM shifted the center of gravity “away from the space between clinician and patient to somewhere between research and clinician.” I think that’s a straw man argument; but even if it is perceived that way by some clinicians, it was never the intention of EBM to cut the individual patient out of the loop.

Interpersonal medicine

They say that interpersonal medicine would recognize clinicians’ influence on patients, that it would be anchored in longitudinal, multidirectional communication, that it would broach social and behavioral factors, that it would require coordination of the care team, and that it would constantly evaluate its own approach.

They say that chronic diseases are rooted in social and behavioral determinants. They say obesity, diabetes, and heart disease “cannot be addressed effectively in isolated office visits.” They claim that meaningful interactions need to be focused on “motivation, engagement, empowerment, conviction, and resilience”, and that these are not limited to provider/patient encounters but may also occur between patients and families and communities. Through such interactions, relationships are built that can endure distraction and overcome hurdles”, they say:

Interpersonal medicine demands skills that place clinicians alongside their patient as fellow travelers and experienced guides…it also requires systems that draw on those skills when they’re likely to make a difference.

I would ask how they know when they’re likely to make a difference and what can guide their actions in treating an individual patient on the basis of social and behavioral factors if there is no evidence to guide them. They are advocating a return to the “intuition-based medicine” that was all we had before EBM and that was far less reliable as a guide to the truth.

They applaud the system at Boston’s Brigham and Women’s Hospital, where Medicaid patients are screened for social needs by practice assistants, alerting the clinician and also triggering actions by community health workers and others. But they don’t provide any evidence that that system improves outcomes, and they don’t explain why it should be limited to Medicaid patients.

They stress the importance of empathy, coordination, and communication, providing references to studies showing that physician communication is associated with a 19% gain in patients’ adherence to recommended therapies and improvement in a variety of outcomes. This is old news to most of us, and they don’t provide any guidance on how specifically to achieve meaningful improvement.

They recommend the development of matchmaking protocols to pair clinicians and patients for success, but they don’t provide any guidance as to how that might be done or any evidence that benefits would ensue. They praise clinics that have no waiting rooms: patients are taken directly to a room designed to accommodate conversation among patient, family members, and clinicians; no exam table is in sight; a chair converts to an exam table when necessary. That’s all very feel-good nice, but where is the evidence that it actually improves patient outcomes?

They point out that some solutions are already available but are inconsistently applied. There is a need for “building systematic capability at a scale that mirrors our scientific effort”. Finally, they conclude:

We can pursue an empathetic version of medicine that embraces emotion and appreciates behavior if we value human nature as much as human biology.

Good medicine is already interpersonal medicine

In my opinion, good clinical medicine already does all those things. There is no need for something new called interpersonal medicine. There is only a need for good clinical medicine to do what it already knows how to do well, even if it is not currently doing a good job of it. In my opinion, it would be injudicious to institute major changes in approach willy-nilly without first testing which changes produce meaningful results.

They are essentially advocating a return to “intuition-based medicine”, which they say wasn’t wrong but had limited access to meaningful data. I would argue that it quite often was wrong. The history of medicine is full of examples where clinicians were misled by intuition.

This is reminiscent of much of alternative medicine, integrative medicine, and functional medicine. They claim to treat the whole patient, stress prevention, and consider social and behavioral factors. But just as with this new proposal for “interpersonal medicine”, there is no evidence that they are actually accomplishing what they claim or that their approaches achieve superior results. In fact, there is evidence that patients seeing alternative providers are less likely to get the immunizations and screening tests that are based on good evidence.

Something missing

Their criticism of EBM is missing a key point, the one that led us to create the Science-Based Medicine blog.

EBM tends to focus on clinical trial results to the exclusion of scientific plausibility. The focus on trial results (which, in the EBM lexicon, is what is meant by “evidence”) has its utility, but fails to properly deal with medical modalities that lie outside the scientific paradigm, or for which the scientific plausibility ranges from very little to nonexistent.

This has led to unfortunate consequences, including quackademic medicine and the infiltration of non-science-based therapies and even outright quackery into our medical schools, clinics, and hospitals. It led the American Academy of Family Physicians to provide poor quality evidence about acupuncture, dry needling, and cupping, and to make the nonsensical practice recommendation that “electroacupuncture should be considered for patients with fibromyalgia.”

EBM is bad enough when legitimate research is misinterpreted and turned into cookbook mandates, but it’s far worse when it relies on Tooth Fairy science to support improbable treatments – especially when it unquestioningly accepts positive results of studies on something like homeopathy, which is incompatible with all we know about science.

Conclusion: Good medicine is already “interpersonal medicine”

The limitations they see in EBM are not inherent in EBM itself; they are a result of individuals misunderstanding and misinterpreting how EBM should be implemented. Rather than trying to go beyond EBM, it would make more sense to correct the abuses of EBM and go back to the original concept of judicious interpretation of the evidence, (with consideration of prior plausibility and compatibility with the rest of science) and applying evidence from studies not to mandate decisions, but to help guide decisions in the context of other unique factors affecting the individual patient.

The best clinicians already practice interpersonal medicine. Admittedly there are many things wrong with the way conventional medicine is currently being practiced. We could learn a thing or two from the way alternative practitioners please their patients (with charisma, compassion, touch, TLC, long appointment times, and the nonspecific provider/patient interactions that maximize placebo responses). We could do a better job of prevention and of helping patients change their lifestyles. But that doesn’t mean we need a new system of interpersonal medicine; it just means we can do better within the existing system. And as we make changes, we still need to rely on evidence to ensure that we really are doing better.