The Good Doctor
Page 14
Paul Lee, “The Nature of Anecdotes,” Emner: Medical Quackery, Pseudoscience, Skepticism, September 1, 2003, http://www.skepticreport.com/sr/?p=423.
Recollection of patient encounter, KB.
Harriet Hall, “The Role of Experience in Science-Based Medicine,” Science-Based Medicine: Exploring Issues & Controversies in Science & Medicine, April 12, 2011, https://www.sciencebasedmedicine.org/the-role-of-experience-in-science-based-medicine/.
“The Plural of Anecdote is Not Data,” Skeptical Medicine, May 25, 2014, https://sites.google.com/site/skepticalmedicine//the-plural-of-anecdote-is-not-data.
“Anecdotal Evidence,” Wikipedia, http://wikipedia.org/wiki/Anecdotal_evidence.
Chauvenet’s Criterion,” Statistics How To: Statistics for the Rest of Us, http://www.statisticshowto.com/chauvenets-criterion/.
Trisha Greenhalgh and Brian Hurwitz, eds., Narrative Based Medicine, 1st ed. (London: BMJ Books, 1998).
Kathryn Montgomery, How Doctors Think: Clinical Judgment and the Practice of Medicine, 1st ed.(Oxford: Oxford University Press, 2005).
David Watts, “Perspective: Cure for the Common Cold,” New England Journal of
Medicine 367 (2012): 1184-1185.
David Hatem and Elizabeth A. Rider, “Sharing Stories: Narrative Medicine in an Evidence-Based World,” Patient Education and Counseling 54 (2004): 251-253.
Peter Tishler, “Soma Weiss, and Alfred S. Reinhart, and the care of the patient,” Perspectives in Biology and Medicine 53, no. 1 (2010).
Kenneth Brigham, Hard Bargain: Life-Lessons from Prostate Cancer . . . A Love Story (New York: Harper House, 2001)
Philip Overby, “The Moral Education of Doctors,” The New Atlantis, pp. 7-26, Fall 2005
Francis Peabody, “The Care of the Patient,” Journal of the American Medical Association 88 (1927): 877-882.
David Hatem and Elizabeth A. Rider, “Sharing Stories: Narrative Medicine in an Evidence-Based World,” Patient Education and Therapy 54 (2004): 251-253.
Vera Kalitzkus and Peter Matthiessen, “Narative-Based Medicine: Potential, Pitfalls, and Practice, Permanente Journal 13 (2009): 80-86.
Jayshil Patel, “Writing the Wrong,” Journal of the American Medical Association 314 (2015): 671-672.
Richard Byny, “The Tragedy of the Electronic Health Record,” The Pharos, pp. 2-5, Summer 2015.
Jilleen Kosko, Terry Klassen, Ted Bishop, and Lisa Hartling, “Evidence-Based Medicine and the Anecdote: Uneasy Bedfellows or Ideal Couple?,” Pediatrics Child Health 11 (2006): 665-668.
CHAPTER 13
An Expert Escort from Guidelines to Protocols
Guideline: a thing that helps someone to form an opinion
or make a decision or calculation
Protocol: a detailed plan of a scientific or medical
experiment, treatment, or procedure
—Merriam Webster Dictionary
Colin (not his real name), a thirty-seven-year-old-man, was admitted to the hospital with stomach pain, headache, vomiting, back pain, and fever. Two hours earlier, he had taken a dose of antibiotic prescribed by his dentist for a tooth infection, but he was a healthy guy, in good physical shape, and not taking any medicines regularly. He was obviously very sick when he arrived at the hospital. He was sweaty, his blood pressure was on the low side, and he had a fever and a rash on his back. His heart was racing, he was breathing too fast, and his fingers were blue. Intravenous fluids didn’t help much and he was transferred to the intensive care unit. His clinical picture looked exactly like sepsis, a blood stream infection with bacteria that is often fatal, although it wasn’t at all clear why he should have such a disorder. There is a rational formula for choosing what antibiotics to give a patient with this diagnosis while trying to identify the specific offending germ, and Colin was started on the four drugs recommended by those guidelines. Over the next two days Colin’s liver and kidneys started failing and his blood cell counts started showing an increase in the number of eosinophils, a specific kind of white cell not usually associated with sepsis. Colin’s doctors, alert to the possibility that their diagnosis might be wrong in spite of his classic signs and symptoms, concluded that he didn’t have sepsis at all, but was suffering from drug reaction eosinophilia with systemic symptoms (DRESS) syndrome, a disorder that was made worse by antibiotics. All antibiotics were discontinued and Colin eventually left the hospital with everything back to normal, a very fortunate beneficiary of a successful navigation of the bridge of maybes from population based evidence to personalized treatment.
“Why the eosinophilia?” his doctors must have thought as they were treating what looked exactly like a straightforward case of sepsis with drugs that were killing their patient. But, sepsis doesn’t cause eosinophilia. Well, maybe in this particular case it’s something else, even though it looks for all the world like sepsis . . . except for those pesky eosinophils. Had his doctors ignored or failed to notice this inconsistency in the data and continued the antibiotics, Colin would have died. His doctors would have rationalized that sepsis is often fatal even with the best possible treatment. The lesson is that even the most carefully developed guidelines may not prescribe the best protocol for a specific case. Colin was fortunate to have doctors who never stopped looking for clues that what they were doing was wrong—pondering the maybes. That saved Colin’s life.154
Richard Parker (a physiologist friend, not the Life of Pi tiger) is fond of this joke: A general and his lieutenant are standing together viewing the battlefield. They have suffered major losses and are reduced to only a few ragtag troops. On the horizon are hundreds of enemy troops advancing rapidly toward their position. Panic in his voice, the lieutenant asks, “What should we do, general?” The general, his tone supremely confident, replies, “Lieutenant, I suggest we surround them and attack from all sides.” Exasperated, the lieutenant responds, “We have only a handful of exhausted and injured men, general, how do you propose we do that?” “Strategy,” replies the general, “is my job; tactics is yours.” Maybe there is a rough analogy in the relationship between guidelines and protocols in medicine—the experts do strategy (guidelines), the tactics (protocols) are up to you and your doctor.
The very fact that the terms guideline and protocol are often confused may be a symptom of a potentially serious problem. A guideline without conditions, i.e., absent the maybes, becomes a protocol, and that can be a fatal error. Guidelines are what you get from expert analyses of all of the available evidence from studies in groups of people, but they should not dictate the details of treatment of an individual person. Guidelines are generalities. They don’t write prescriptions or doctors’ orders. When you are ill, what you need is a protocol. Getting from a guideline to a protocol is getting from evidence-based to personalized medicine. The only path that connects the two is a bridge of maybes. Take the maybe out of guideline and the bridge is burned; your doctor is no longer treating you.
Does it have to be that way? Why isn’t it possible to write guidelines with no question marks that apply to everybody, no questions asked? After all, this isn’t rocket science! Well, that’s right, it’s not rocket science, it’s a lot more complicated than that. The good doctor knows something about how guidelines come about, and she’s not about to use them as protocols to treat you without seriously questioning whether they are the best she can do for you in your specific situation.
Here are some things that this doctor knows about guidelines. As is said of success, practice guidelines have many fathers. That is, they are generated separately by several different groups, each with their own interests: the federal government (see the National Guideline Clearinghouse); a host of medical societies; the Cochrane Collaboration; entities tha
t pay the medical bills; and various health care systems. Guidelines are influenced by the opinions and clinical experience of the experts chosen by the sponsoring organizations to write them. If the organizations are doing their job, they choose experts who are friendly to their special interests. There are a lot of special interests—controlling costs, protecting turf, favoring constituents (like doctors, risk managers, or politicians), maximizing reimbursement, etc.—that are not necessarily focused primarily on what is best for you. And then because of unconscious bias, other misinterpretations, or absent or misleading evidence, guidelines can be just plain wrong, eventually orphaned by their failure to pass muster in the real world, but leaving some damage in their wake.
Guidelines should be flexible enough to accommodate the idiosyncrasies of specific patients, but that doesn’t mean that guidelines shouldn’t be taken seriously. The good doctor knows that she can’t just close her eyes and do whatever the guidelines say, but she also knows that she ignores well-
reasoned guidelines at her and her patients’ peril.
For example, guidelines from the Cochrane Collaboration, the American Pain Society, the American Society of Interventional Pain Physicians, and the American Academy of Neurology all conclude that there is no evidence that injecting steroids into the spine results in prolonged relief of low back pain.155 (The Cochrane review adds the caveat that “it cannot be ruled out that specific subgroups of patients may respond to a specific type of injection therapy.” Maybe it works in the “right patient.”156)
Spinal injection of steroids is not approved by the Federal Drug Administration for low back pain. That is what is called an off-label (the drug is approved for some diseases but not this one) use of a potent steroid, methylprednisolone. In spite of the guidelines and the lack of FDA approval, nine million times every year in this country, a steroid is injected into someone’s spine to treat the incredibly common symptom of back pain.157 Some of those patients no doubt fit into one of the specific subgroups likely to respond to a specific type of injection therapy, but many do not. And there are always risks, some of which may be unsuspected even by the most alert and up-to-date doctor.
An infectious disease specialist had been puzzling over her patient with meningitis for two weeks. The patient wasn’t responding to antibiotics that should have taken care of the usual causes of the disease, and no bad bugs were growing from his body fluids that were sent to the laboratory. That is until one September day when the laboratory called with the answer. Finally, a fungus was growing from the patient’s spinal fluid. He had a rare and often fatal form of fungal meningitis. In response to detailed questioning about his health practices, the doctor discovered that her patient had recently had an injection of steroids into his spine to treat his chronic back pain. The source of the deadly fungus in his spinal fluid was finally traced to contaminated batches of the injected steroid solution that led to a major epidemic.158
The contaminated steroid story demonstrates that the consequences of ignoring guidelines aren’t just abstract medical mistakes; they can include the kinds of personal tragedies in real people that haunt a concerned doctor. Cynthia Scribe’s (not her real name) friends described her as “hilarious, beautiful, and full of life.” She was the primary caregiver for her husband, who was confined to a wheelchair with Lou
Gehrig’s disease. She lost her job and wanted to get her back pain treated while she was still covered by health insurance. A few hours after her third injection of the tainted steroid, she developed headaches. Her course was rapidly downhill and finally, after suffering several strokes, she died in a hospice surrounded by her husband and friends.159
The contaminated steroids were produced by New England Compounding Center in Massachusetts, the primary supplier of the drug (methylprednisolone acetate) intended for this use. 160The meningitis epidemic spread to nineteen states and 720 victims, 48 of whom died. The risk of something like this is no doubt low, but also most of the patients who got the drug were unlikely to benefit from it. There are always risks, some of which can’t be anticipated. The good doctor takes the Hippocratic dictum “first do no harm” seriously. If she isn’t pretty sure she can help you, she’ll at least do her best not to make you any worse.
So ignoring guidelines is risky even though the experts who develop them deliberately leave room for adapting the therapy to you; you could be one in a subgroup that is likely to respond in spite of a negative guideline recommendation. A thinking doctor understands how important uncertainties in the guidelines are in making your treatment personal, but she doesn’t take the uncertainties as a license to freelance. She won’t advise you to have a therapy based solely on her limited experience if that differs radically from the larger body of evidence. This doctor has vivid memories of her patients. She cherishes those memories, and, though they do not trump the evidence, they may help as the two of you ponder difficult options. A single experience is a dangerous basis for deciding on a treatment, but experience can be a valuable aid as you and your doctor wend your way over the bridge from guidelines to protocols.
So, your doctor generates a protocol that is especially yours, but it may not be unique in the strict sense. Here is an oversimplified example that illustrates the point. Patients with cystic fibrosis (CF), a genetic disease that mostly affects the lungs, often have spells where their lung infection takes over and they have to be treated in the hospital. There is a lot of evidence that several interventions are effective—intravenous antibiotics, chest physiotherapy, inhaled medications, etc. So, based on guidelines carefully developed from the available evidence, CF doctors usually have “protocols” that include each of the evidence-based interventions. These show up automatically on the doctor’s orders in the admission part of the electronic medical record. But, isn’t that just rote adherence to the guidelines? No, it’s not, because there are options. Choice of antibiotics will depend on what bacteria are growing out of the lungs and whether the patient is allergic to anything. Frequency and dose of inhaled medications will depend on the patient’s size, ability to cooperate, how they have responded to such treatments in the past, and many other variables that are unique to a specific situation. The point is that your specific treatment comes from a careful distillation of the evidence, your idiosyncrasies, and your doctor’s and others’ experience. And your doctor doesn’t just design your protocol and go home. She pays continual attention to how you are responding, ever alert to the possibility that you could be an outlier and that your protocol might need revising at any time.
As she does her best to give you the best medical care possible, the good doctor is always aware of how important it is to recognize the difference between guidelines and protocols. Statistical data from population studies are invaluable, and the more there is of such evidence, the better. But evidence-based medicine done right uses evidence to inform care related decisions rather than as a precise prescription for what to do for an individual patient.
When deciding what to do to be as healthy as you can be or to treat your disease when you get sick, you want the best possible chance to do what is best for you. That means clearly understanding the difference between guidelines and protocols, a distinction that you can’t reliably make on your own. Far and away your best resource is a thinking doctor who listens well, knows you, and knows what she is doing—a doctor whom you trust, and with whom you’ve built a relationship that is as equal a partnership as possible.
The internet doesn’t do a very good job of even getting the facts straight and pretty much misses entirely the nuanced distinction between guidelines and protocols. If you ever feel inclined to blindly trust what you get from the public media, the independent website http://HealthNewsReview.org ought to give you pause; they critique news stories about medical claims. They reviewed around 1,800 stories from a dozen or more reputable organizations in the U.S. and found that seventy percent didn’t accurately report cost, potential for benefit, and potential fo
r harm.161 And these were stories from reputable sources. Imagine the reliability of unfiltered information from infomercials, public personalities, and the like.
“Your health is your most valuable asset” is more than a shopworn phrase, it is an essential fact of life. If you don’t believe that now, you will as soon as you have to deal with a serious disease; there is nothing like being really sick to rearrange one’s priorities. We are fortunate to live in a free country where a lot of resources are available that can help us to be as healthy as we can be. But the freedom that gives us access to good information also opens the door to misinformation that can lead us in the wrong direction. You and your doctor need to work together to parse the facts and fictions to your advantage. You need the knowing doctor’s steady hand if you’re to make it safely across the bridge of maybes.
You would be foolish to go it alone. You’ll have gut reactions, but don’t trust them. Viscera have their critical functions, but they are unreliable discerners of truth.
Philippa Horsfield, Sanjay Deshpande, and Richard Ellis, “Killing with Kindness? Drug Reaction Eosinophilia with Systemic Symptoms (Dress) Masquerading as Acute Severe Sepsis,” Journal of Antimicrobial Chemotherapy 64 (2009): 663-665.
“Epidural Steroid Injections for Back Pain: Worth a Shot, or Should You Skip It?,” Consumer Reports, March 2011.
J. B. Staal, R. de Bie, H.C. de Vet, J. Hildebrandt, and P. Nelemans, “Injection Therapy for Subacute and Chronic Low-Back Pain,” Cochrane Database of Systematic Reviews, July 16, 2008, doi: 10.1002/14651858.CD001824.pub3.
Pat Anson, “Study Questions Use of Steroids in Spinal Shots,” National Pain Report: Pain Medication, September 18, 2013, http://nationalpainreport.com/study-questions-use-steroids-spinal-shots-8821712.html.
Denise Grady, “Meningitis Cases Are Linked to Steroid Injections in Spine,” The New York Times Health, October 2, 2012.