Ending Medical Reversal: Improving Outcomes, Saving Lives
Vinayak K. Prasad and Adam S. Cifu
Every doctor should read this book.
Medical reversal is when a treatment is widely accepted by the medical profession as the gold standard of care — and later found to be useless or even harmful. Even before I read Ending Medical Reversal, I was becoming alarmed by my own experience with medical reversals as a pulmonary and critical care physician. The use of activated protein C for sepsis is one example: It was all the rage during my critical care training, but in 2012 a randomized controlled trial found no benefit. Similarly, intensive blood glucose control in the ICU, once believed to improve mortality, subsequently turned out to do the opposite. So without directly recognizing the phenomenon of medical reversals, I had already begun to change my practice style – to require two well-designed clinical trials with consistent results before jumping on the bandwagon of a new therapy.
But for me and, I suspect, many other doctors, this book serves the needed function of giving the phenomenon of medical reversal a name and dragging it into the light. And looked at squarely, medical reversal must be recognized as a major source of harm to patients, and a large-scale failure of medical science. The authors of the book first brought attention to the issue of medical reversal in a research letter published in JAMA Internal Medicine in 2011. In this book they go further: using careful quantitative analysis, they estimate that 40% of therapies may become medical reversals, and provide a dismaying appendix of 142 reversals over a ten year period.
The scientific process is by nature experimental, and there are bound to be false starts. Truth is approached asymptotically as research adds to, improves, and sometimes corrects earlier research. But the question is at what point along this continuum are therapies permitted to move out of the context of research, and into widespread use. Prasad and Cifu believe novel therapies are often adopted too early, and that a higher standard of proof should be required before the transition. The premature movement of therapies into general use, they believe, is the consequence of inadequately critical evaluation of evidence — in conjunction with the grip of industry and financial conflicts of interest on the medical profession.
The authors argue that patients, doctors, and researchers share a human tendency to be optimistic about new therapies. This hopeful tendency needs only a small push from financial self-interest to become indiscriminate – and the authors rightly point out that what it gets is something more like a shove. Physicians often stand to benefit financially from tests and procedures. And the pharmaceutical and device industries bring considerable pressure to bear through marketing and through financial relationships with physician “thought leaders” and guideline developers, researchers, academic medical centers, and the very agencies charged with regulating them. To this I would add one other factor: our medical training is so intensely socializing, and the tradition of learning from powerful mentors so strong, that as a profession we are particularly susceptible to “group-think” – we tend to march in lock-step, even when in the wrong direction.
In this environment, it is crucial that practicing physicians are well-trained in how to evaluate evidence supporting new therapies critically. But, the authors argue, this skill does not receive enough emphasis in traditional medical training. Instead, doctors are trained to think mechanistically, making them vulnerable to accepting therapies based on plausible explanations of how they might work – rather than whether they actually do work.
How can we minimize medical reversal? Prasad and Cifu suggest wide-spread changes to reduce the influence of industry and create a higher standard of evidence for new therapies. Some of these would require governmental action: for example, creating an independent agency to solicit and coordinate clinical trials (effectively taking the role away from industry), and giving the Food and Drug administration more teeth, with a higher standard for demonstrating effectiveness and a tougher requirement for confirmatory studies after accelerated approval. But many of the suggested changes could be made from within the medical profession. For example, just as the justice system places a burden of proof “beyond a reasonable doubt” on the prosecution in a criminal case, the medical profession should require a rigorous standard of evidence before adopting new therapies. Academic medical centers should promote critical literature reviews and confirmatory studies by giving these more weight in tenure and promotion considerations. In addition, they should prohibit faculty from receiving honoraria, gifts, and other enticements from industry. Concerted effort should be made to enroll patients in randomized controlled trials to expand the evidence base. And finally, medical education should be overhauled to provide extensive training in evaluation of evidence, rather than the current emphasis on mechanistic thinking.
In their emphasis on the importance of critical thinking and on the shameful grip of industry on the medical profession, Prasad and Cifu are right on the money. Readers will no doubt have concerns and questions about the specifics of some of the reforms. The suggested large changes to medical education, for example, seem to go beyond what would be necessary to increase the emphasis on critical evaluation of evidence, and will be controversial. In addition, the reforms are so wide-ranging that it is difficult to imagine a coordinated path towards their implementation. But the first step has to be recognition and understanding of the problem, and this book provides that cornerstone.
This slim, quick read is jaw dropping for its ambition and reach. It searches in every corner of the profession, and provides detailed prescriptions for reform. In simple, laymen’s terms it manages to describe complex principles in the evaluation of scientific research. I suspect that despite the fact that the book is written in language accessible to the non-medical public, the most appreciative audience may turn out to be physicians (many of whom, let’s be honest, might be grateful for an update on study design and type-I error in laymen’s terms).
But, physicians beware: this book will mess with any complacency you might still have about what you think you know.
Powered by WPeMatico