ENDING MEDICAL REV∃RSAL
ENDING MEDICAL REV∃RSAL
:: Improving Outcomes, Saving Lives
VINAYAK K. PRASAD, MD, MPH
ADAM S. CIFU, MD
Note to the reader: This book is not meant to substitute for medical care, and neither diagnostic decisions nor treatment should be based solely on its contents. Instead, they must be developed in a dialogue between the individual and his or her physician.
Drug dosage: The authors and publisher have made reasonable efforts to determine that the selection of drugs discussed in this text conform to the practices of the general medical community. The medications described do not necessarily have specific approval by the U.S. Food and Drug Administration for use in the diseases for which they are recommended. In view of ongoing research, changes in governmental regulation, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert of each drug for any change in indications and dosage and for warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently used drug.
© 2015 Vinayak K. Prasad and Adam S. Cifu
All rights reserved. Published 2015
Printed in the United States of America on acid-free paper
9 8 7 6 5 4 3 2 1
Johns Hopkins University Press
2715 North Charles Street
Baltimore, Maryland 21218-4363
www.press.jhu.edu
Library of Congress Cataloging-in-Publication Data
Prasad, Vinayak K., 1982–, author.
Ending medical reversal : improving outcomes, saving lives / Vinayak K. Prasad and Adam S. Cifu.
p. ; cm.
Includes bibliographical references and index.
ISBN 978-1-4214-1772-1 (hardcover : alk. paper) — ISBN 1-4214-1772-3 (hardcover : alk. paper) — ISBN 978-1-4214-1773-8 (electronic) — ISBN 1-4214-1773-1 (electronic)
I. Cifu, Adam S., author. II. Title.
[DNLM: 1. Delivery of Health Care. 2. Treatment Outcome. 3. Clinical Trials as Topic.
4. Evidence-Based Medicine. 5. Therapeutics—trends. 6. Popular Works. W 84.41]
R733
615.5—dc23 2014049513
A catalog record for this book is available from the British Library.
Special discounts are available for bulk purchases of this book. For more information, please contact Special Sales at 410-516-6936 or [email protected].
Johns Hopkins University Press uses environmentally friendly book materials, including recycled text paper that is composed of at least 30 percent post-consumer waste, whenever possible.
V.K.P. For Mom, Dad, Karthik, and Nancy—the four pillars of my life
A.S.C. For Sarah, Ben, and Amelia
CONTENTS
INTRODUCTION
PART I MEDICAL REVERSAL :: Examples, Frequency, and Consequences
1 WHAT IS MEDICAL REVERSAL?
2 SUBJECTIVE OUTCOMES :: Why Feeling Better Is Often Misleading
3 SURROGATE OUTCOMES
4 SCREENING TESTS
5 SYSTEMS FAILURE
6 FINDING FLAWED THERAPIES ON OUR OWN
7 THE FREQUENCY OF MEDICAL REVERSAL
8 THE HARMS OF MEDICAL REVERSAL :: Today’s Patients, Tomorrow’s Patients, and the Health-Care Field
PART II AN INTERLUDE ON EVIDENCE
9 A PRIMER ON EVIDENCE-BASED MEDICINE :: What Is Evidence in Medicine?
10 WHAT REALLY MADE YOU BETTER :: When Evidence Gets Complicated
PART III THE ORIGINS OF REVERSAL
11 SCIENTIFIC PROGRESS, REVOLUTION, AND MEDICAL REVERSAL
12 SOURCES OF FLAWED DATA
13 WHY ARE WE SO ATTRACTED TO FLAWED THERAPIES?
PART IV BEYOND REVERSAL
14 MEDICAL EDUCATION :: A Very Good Place to Start
15 ACADEMIC MEDICINE
16 REFORMING THE SYSTEM :: The Burden of Proof and Nudging Our Way Past Reversal
17 HOW NOT TO BECOME A VICTIM OF REVERSAL
18 BEYOND DOGMA :: When Randomized Trials Are Unnecessary
ACKNOWLEDGMENTS
APPENDIX
REFERENCES
INDEX
ENDING MEDICAL REV∃RSAL
INTRODUCTION
THIS BOOK IS ABOUT HOW, despite tremendous advances in the clinical, genomic, and surgical sciences, doctors continue to use medical practices, sometimes for decades, that are later shown to be of no benefit to their patients. Over the past six years, we have researched and struggled with this topic, which we call medical reversal, and in this book we share some of our conclusions.
For those who follow the medical news, the simple fact that medical recommendations change will come as no surprise. Time and again we see enthusiasm for some medical therapy (beta carotene, vitamin E, lowfat diets) rise and fall. First there is certainty that a new practice will help extend your life, then there is equal certainty that it does not. To people outside of medicine (as well as many inside the field), this process is frustrating.
When you look at the day-to-day recommendations that doctors make— the ones that usually do not make the news—you find the story is similar. In one decade, doctors recommend an aggressive treatment of high-dose chemotherapy and stem-cell transplantation for women with breast cancer, promising that it will give a woman her best chance of a cure. Then, over the next decade, doctors report that all of that enthusiasm was misguided; the aggressive treatment was no better than a less aggressive course of therapy, which, incidentally, was what we had been doing previously.
Many people dismiss this phenomenon as the natural course of science: of course hypotheses turn out to be wrong, and we can only move forward through trial and error. Although this is certainly true in biomedical science—where there are false starts, good hypotheses that fail to live up to expectations—it is not the case in medicine. Medicine is the application of science. When a scientific theory is disproved, it should happen in a lab or in the equivalent place in clinical science, the controlled clinical trial. It should not be disproved in the world of clinical medicine, where millions of people may have already been exposed to an ineffective, or perhaps even harmful, treatment.
In this book, we hope to make the case that, when it comes to how we care for people, medicine can do a better job of recommending practices that actually work. In the long run, more and more of what doctors do can be enduring. False starts are inevitable in science, but not when we apply science to caring for people.
HISTORY
We write this book not as critical outsiders, but because, like so many other doctors, we have been there. Each of us recalls moments when we realized that what we had told our patients, or did for them, was wrong: we had promoted an accepted practice that was, at best, ineffective.
For Adam, like many doctors of his generation, the most memorable reversal was estrogen replacement for postmenopausal women. For years, he would draw the diagram in figure I.1 at least once every couple of weeks. It was a way to explain why doctors recommended that women use estrogen-replacement therapy after menopause. He would explain that the treatment would be good for the bones (decreasing the risk of osteoporosis), bad for the breasts (increasing the risk of breast cancer), and good for the heart (decreasing the risk of heart attacks). On balance, the benefits outweighed the risks. Many of his patients started the therapy.
But then a well-designed clinical trial showed that this diagram was flawed, along with the advice. Estrogen-replacement therapy was of no benefit to the heart. In fact, at least early in the course of therapy, the treatment might even carry some risk. This was a perfect example of reversal. An accepted, widely use
d therapy was found to be ineffective or possibly harmful. Doctors stopped recommending the therapy not because we discovered something better, but because we never should have used it in the first place.
I.1. Explanation of the risks and benefits of estrogen-replacement therapy (circa 1996).
For Vinayak, the experience with reversal that resonates most occurred in the cardiac intensive care unit (CCU). It was early in his training; the hours were long and the days were grueling. There were orders to be written, tests to coordinate, medications to prescribe, ventilators to adjust, and no shortage of procedures to perform. However, when he went home each day, it was not these things that occupied his thoughts, but whether a core intervention was helping people.
For many of the patients he cared for, the CCU team would recommend placement of a coronary stent, a small, expandable, metal mesh tube that can open narrowed or blocked coronary arteries. The justification for the recommendation varied widely based on the patient’s problem. For some patients, those having heart attacks, the recommendation was well-founded. Multiple studies had shown that stents saved lives. For other patients the team recommended stents where they had not been proven effective, and in some cases, stents were recommended where they had been shown not to help. For him, the situation became tragic when one patient, who had received a stent she did not need, suffered a complication of the procedure. She experienced harm without possibility of benefit. Vinayak kept thinking, What about our oath to “do no harm”? Although we preached something called “evidence-based medicine,” few of us were truly practicing it.
EVIDENCE-BASED MEDICINE AND MEDICAL REVERSAL
Twenty-five years ago, “evidence-based medicine” (EBM) was seen as the future of medicine. EBM was founded on the idea that our practice should be based on empirical evidence from studies of real patients. At its birth, EBM was revolutionary. Until that time, medical practice had been based first on hypotheses and clinical experiences and later, additionally, on our understanding of the biological underpinnings of disease. In EBM, those things were still considered important and still needed to be considered, but it was now appreciated that they should always be bolstered with information from well-done, reliable, clinical experiments.
Evidence-based medicine was accepted because of the realization that, not infrequently, practices that seemed to work and practices that the best science said should work did not. In 1981 John McKinlay, a medical sociologist, wrote an article that perfectly summarized the state of affairs in the pre-EBM world. In the article, he described the “seven stages in the career of a medical innovation.” These stages began with a “promising report” in which a medical innovation is publicized based on its promise—often a good explanation of why it should work. In the second stage, the innovation is adopted by the profession, motivated by forces as disparate as belief that the innovation will benefit patients, peer pressure, and the promise of financial gain. Stage three, in which patients and payers accept the innovation as standard, follows quickly. Only at the fourth stage in McKinlay’s analysis do “data” begin to enter the story. However, the data supporting the innovation come only from insubstantial studies that support the innovation in the most superficial way. (We spend a great deal of ink in this book discussing the uneven data that support medical innovations.)
Finally, at stage five, the randomized controlled trial, the foundation of evidence-based medicine, makes an appearance. This kind of reliable, experimental study may either support the innovation or prove (or at the very least suggest) that it is ineffective. The latter case is what we refer to as reversal. Stages six and seven see the response to the reliable, experimental data: first denial, as entrenched interests deny that the innovation may not be effective, and then finally acceptance.
McKinlay’s article was not only brilliant but prescient as well. In writing about medicine in 1981, he addressed issues that are still with us 35 years later and made suggestions for a way forward, many of which we endorse in this book.
EBM adopted as a way to move beyond McKinlay’s seven stages. In evidence-based medicine, innovations that seemed to work and made sense would not be adopted until they were supported by robust data. By the early 1990s, EBM was becoming gospel. The Journal of the American Medical Association published a series of articles about how to appraise reports of clinical trials. Students were taught to dissect studies published in the medical literature. Attending physicians would hand out Xeroxed copies of articles from the week’s New England Journal of Medicine. The most common question asked on physicians’ rounds was, “What is the evidence for that?”
Over the past decade, however, the EBM revolution has not been sustained. It seems counterintuitive to say this because, by many measures, EBM is as strong as ever. We now have more evidence than ever before. Our journals are full of randomized trials and meta-analyses (combinations of randomized trials), the most reliable studies on which to base practice. We have researchers devoted to studying the evidence base of medical practice. Yet, the studies we have often avoid big questions, or they are built with so much bias—favoring one side from the start—that they are not useful. And for each practice that is shown not to work, it seems as if two more dubious ones take its place. It is not that counterrevolutionaries have seized the ramparts, but the dedication to the cause, to practicing medicine from an evidence base, has waned. Some of EBM’s wane has been passive; it is hard to critically appraise every decision you, as a physician, make. Some of the wane has been less than passive, with drugmakers, device makers, and even scientists sullying the purity of the medical literature. Either way, the result has been a near epidemic of medical reversals.
What are medical reversals? We expect that medicine will progress in a generally orderly fashion, with good medical practices being replaced by better ones. We used to use cholestyramine—a horribly tolerated drug that had no effect on patients’ life expectancy—to lower cholesterol after heart attacks. Now we use atorvastatin, a well-tolerated drug backed by robust evidence that it saves lives. This is how medical practice should evolve. Reversal, however, is different. Reversal occurs when a currently accepted therapy is overturned, found to be no better than the therapy it replaced. This often occurs when a practice—a diagnostic tool, a medicine, a procedure, or a surgical technique—is adopted without a robust evidence base.
In the history of medicine, the examples of reversal are legion. Books have been written about innovations that were adopted based only on a flawed theory and ended up helping no one. From medications (calomel— mercurous chloride—for the treatment of yellow fever and arsenic therapy for syphilis) to procedures (cupping—the use of topical suction to promote cure) to surgeries (lobotomy), medicine has a rich, though ignominious, history of reversals. These examples are interesting, but they are not our focus. They are treatments adopted before we had a clear path forward. We concentrate on reversals in the age of EBM, an age when we know how medicine should safely and reliably advance.
Our interest in reversal, which began during the clinical experiences that we mentioned, grew as we read article after article that overturned accepted practices. Medications that we had prescribed (estrogen replacement and rofecoxib [Vioxx]) and procedures that we had recommended (vertebroplasty, stents for stable coronary-artery disease) were among the ones overturned. We began to research the frequency, causes, and harms of reversal. The more we learned, the more we realized that not only was this a highly prevalent problem, but it was one whose solution could go a long way toward curing the ills of our health-care system. Seeing how far our field has strayed from the evidence convinced us that this was a topic that had to leave the periphery of the medical literature and enter mainstream conversations.
AUDIENCE
We write this book for people, like us, who are confused every time they see a report that a therapy that was recommended yesterday is no longer recommended today. The therapy has fallen out of favor not because it was replaced by something better but because i
t was found not to work (or at least not to work better than the previous standard or less invasive therapy). This news comes not only in medical journals but in newspapers, radio reports, and endless Internet coverage. In this book we give dozens of examples of medical reversal, some of which might already be familiar. We show that reversals have happened in every corner of medicine. We go to great lengths to explain the causes of reversal, the harms reversal brings, and, most importantly, possible solutions.
We propose solutions that you, and we, can use. We all have been, and will be, patients. At these times we all want to know, “How can I be sure that the treatment offered to me today will not be found wanting tomorrow?” Beyond answering this, we go further and suggest more systemic solutions that can make our field less apt to adopt tests and treatments that do not work. These solutions should be of interest to students of medicine, doctors, and policymakers.
CONTROVERSY, REFERENCES, AND PATIENTS
Some of the arguments we make in this book are controversial. We state that a sizable proportion of what doctors have done has turned out to be wrong—not wrong in retrospect but unfounded when they were doing it. We also argue that much of what doctors still do is wrong. We have made every effort in the writing of this book to steer away from controversy. When discussing practices that have been reversed, we avoid areas where there is still honest debate. Even with that care, some readers are sure to disagree with some of what we say. Sometimes, the interpretation of medical practice and scientific trials is not black and white. We hope that if you disagree with one or two of our examples or arguments, you will read on. We think that on balance, the weight of our argument will convince you.
Because of the controversy of the topic, we have tried to consider (and often offer) the other side of any debate, and we always provide references to our original sources. We list these at the end of the book, grouped by chapters. We hope you will read the ones that interest you most.
Ending Medical Reversal Page 1
