Archives for posts with tag: evidence-based medicine

21st Century Cures Act will distort the meaning of ‘FDA approved’

The term “FDA approved” means a lot to those of us working in health care and the patients we treat. But if the 21st Century Cures Act becomes law – the House of Representatives approved it Wednesday and the Senate will vote on it next week – this mark of trustworthy stewardship will become a shadow… (more…)

I am looking forward to participating in a TEDMED Google+ Hangout, March 25th at noon, on the topic of Health Myths. You should come participate! To quote a blogpost from the TEDMED folks:

You can get the flu – or worse — from a vaccine. Only old folks get strokes and heart attacks. Calories in, calories out. X-rays cause cancer. Sleep eight hours a night and drink eight glasses of water a day. Skip the sunscreen on a cloudy day. Take a multivitamin every day, just to be on the safe side. An apple a day keeps the doctor away. Exercise more to lose weight.

What are the most popular health myths, and how do they spread?  Does social media spread scams faster that it helps dispel them?  How can doctors help patients practice proven steps for prevention – and still keep up on current research themselves?

Great questions all. Leaving aside a quibble about X-rays and cancer (about which the jury seems to be out still, or at least conflicted), the list above points to something important about so-called health myths: they come in many different flavors. Flu is not caused by a vaccine; that’s just wrong. But older people are generally speaking indeed at higher risk of strokes and heart attacks (though of course it’s not only them). “Exercise more to lose weight” isn’t exactly true, but exercising is one way to keep off weight lost through reduced caloric intake (and – can lead to weight loss by itself, actually).

All this is just to say that one myth may be quite unlike another. They are as diverse in their way as any kind of belief. Thinking more about it, I would like to jettison the whole term “myth” altogether. We all have beliefs. Some of them hew fairly closely to the science – others don’t.

What the summary above artfully avoids is potentially the most interesting question: who believes most in scientifically unfounded assertions? Is it patients, or doctors (and nurses) themselves? I would wager that we are all in the same boat. Much as doctors don’t know statistics, much of us don’t practice according to the medical evidence either.

If you push a little bit on many medical assertions once held to be widely agreed-upon truths, you will find yourself coming away with a handful of dust — and more, if you push harder. Does everyone need screening for prostate cancer? No. What about breast cancer? Unclear. Does vitamin D help much, say, for heart disease (except in older populations at risk of fracture)? Maybe, though the evidence is thin. If we treat mild hypertension, can we expect mortality benefits? Doesn’t seem that way. Yet practices based on these suppositions persist.

In other words, doctors are as much myth-makers, and myth-peddlers, as patients. Which means we all need to reevaluate our relationship to science. If doctors can be as uncertain as patients, shouldn’t we be skeptical of the hierarchy that still obtains in certain quarters? Shouldn’t we come together to discuss our fears, preconceptions, worries, and expectations?

I covered one of the chief residents in our hospital for two days this past week, seeing more than 40 patients in total. My off-the-cuff remarks on Facebook still apply:

Twenty-two inpatients later, it is time once again to declare my awe and admiration for all who do this work daily: hospitalists, housestaff, nurses, techs, custodial staff (et al., et al.). And, of course, the patients who are – on the other side of the hospital mirror – working harder, in their way, than all of us.

What’s more, I was struck – not for the first time – by the differences between the hospital and clinic, not just in tone, atmosphere, communication, and pace, but in how those things affect medical thinking and decision making.

I am not talking about critical care or emergency medicine, nor about “codes” (cardiac resuscitation) – merely because I do not participate in such care regularly. But I think my observations might still apply.

Simply put, the hospital operates under the assumption that things have to move faster. That tempo, I believe, encourages a certain frame of mind: we should treat, do, test. People are sick, we want to make them better, and lab tests and interventions are an avenue to this.

You already see where I’m going, because – if you know me or my blog – you know my bias. If anything, I think we tend to over-order as a health care system….and the scientific literature bears me out.

What would it mean to apply evidence-based medicine to hospital care, in a thorough-going fashion? I mostly do outpatient medicine, and guarantee you that I assume no superiority for the application of EBM in that realm: no, much of what we do in clinic is still based on intuition, externalities, unsupported lore, personal preference (not the patient’s, God forbid, but our own as doctors), or some other blend of bias and conviction.

Would it be harder in the hospital to take stock? Would workflow be disrupted?

Or is it just that I am unaccustomed to the hospital, and things are really changing in that direction? What do you think?

My Facebook feed is riven in two. On the one hand, there are free-thinking, scientistic doctors, patients, and those who love them (or tolerate them enough not to hide their status updates). On the other, are Jews: and, for some reason having to do with personal curiosity and involvement with Yiddish speakers, many of them are Jews who are Chasidic, used to be, or are somewhere in between.

Both camps are in the throes of realignment. The former, because evidence-based medicine is screeching towards the end of a game of chicken, the other competitor being (unfortunately) the patient as a person, with power over her own decision making.

Let’s take the most recent example of this hair-raising collision: statins. Cholesterol medicines, in other words, and what to do about them. Yes, we should probably get rid of these artificial numerical targets, it seems like most people agree on that. But the new guidelines, according to which (grosso modo) everyone with an estimate 10-year heart disease risk of 7.5% “needs” to be on such a medication, are rightfully controversial. If you are the type to read JAMA articles, you will be entertained by John Ioannidis (he of the “half of all scientific findings are wrong”) on the one hand, who darkly prophesies “one of the worst disasters in medical history” due to vast overtreatment with statins. (Um. Flu? AIDS? The plague?), and, on the other hand, a professional, prudent take by some collaborators on the new guidelines themselves, who take what seems to be the best defense possible: The science is better; we are working towards incremental improvement of guidelines. In short: nothing’s perfect. This is a step forward from where we were.

If you read the articles back-to-back, it seems an awful lot like the internal breakup currently convulsing ultra-Orthodoxy. Belief and practice are at loggerheads, and tiny fish are getting squished between the logs. Are you a believer or not? Are you an incremental advancer or a revolutionary? Are you willing to listen to doctors by virtue of their social standing and hereditary place as healers, even if their advice might not be better than random chance?

Does medicine need a revolution to upend received wisdom – even the new received wisdom of our day, which is the dogma of evidence-based medicine – enthroning in its place the empowered singular patient? Sure, if you are the type to upend and revolt. But not even the most engaged of patients want to get rid of everything the medical establishment has to offer. Similarly, even those about to leave their hidebound religious communities sometimes find themselves at peace with a stable compromise. They don’t have to believe in everything.

If a revolution is possible, but we choose an incremental change, are we being sensible – or hypocritical?

I am exhausted, but before I drift off to bed here in Beijing I wanted to give an account of my first full day.

I sat in on rounds at Peking Union Medical College Hospital, my host and one of the top-ranked hospitals in China. The General Internal Medicine Division is renowned for its ability to treat the hardest cases and consistent high reputation, which becomes a self-fulfilling prophecy in certain respects (sound familiar?).

The similarities are not all that interesting: the team sits round a table and talks about the new patients, then walks through the wards seeing the old patients. Questions are asked to put medical students on the spot (in American English we have a word for that). The differences, however, are somewhat instructive.

In the United States, at least in the internal medicine programs I am familiar with, the senior resident runs rounds and the attending stands by the side to give a teaching point or a minor correction; here, it was the attending leading the discussion. In the United States, the entire team, in many hospitals, is by now acculturated to use hand sanitizer on leaving and entering every room. In the PUMC GIM ward, I was told by someone that I didn’t need to use sanitizer if I wasn’t touching the patient.

There was one similarity which was immediately evident: the hierarchy that hung over interactions between doctor and patient, and the great respect with which the patients treated the doctors every word (though the medical students I spoke to later expressed worries that patients no longer respected them). I don’t understand enough Chinese to know whether the doctors were attuned to the patients’ needs apart from their own particular workflow needs on rounds, but if these doctors are anything like many American ones, I can guess the answer…

* * *

Later, I had the great opportunity to give a presentation for medical students about bridging evidence-based medicine and patient-centered care, using localized prostate cancer as a case in point. We are trying to understand why patients with that most limited stage of cancer might leave an active surveillance (watchful waiting) program to get radiation and surgery which might not be clinically indicated.

We had a lively discussion. I fielded an expected question about what differences I noticed between the Chinese and American health care systems, after less than 48 hours of superficial experience with the former. I tried to demure, but one thing I did talk about was the overuse in the American system, over against the underuse in China which is prevalent for millions and millions of mostly rural poor. We also talked about what doctors might do when what patients want is against the best evidence.

After the lecture, I had a chat with a student of Uygur ancestry who was very interested in the role of religion in health care in the United States. I told him what I think is true: aside from end-of-life care and bioethics, the role of religion is underexamined.

* * *

Finally, I met a bioethicist, Yali Cong, from the Peking University Health Science Center (not to be confused with PUMCH, above. A city of 20 million, Beijing has a lot of hospitals!). We talked about one of the chief difficulties for those involved with clinical and research bioethics: the expectations of clinicians that bioethics will be able to give an “answer,” where in reality what a bioethicist can give is an overview of possibilities, a mapping of the territory, and – in the most lasting influence – a habit of thought that even, or especially, non-bioethicists might benefit from.

* * *

It’s been a great visit so far, and even after I leave China I hope to connect with people here through email and Weibo, my newest social media addiction made all the more interesting by the fact that my Chinese is a lot less than fluent.

I was shocked to read about what some doctors at my institution have been doing. Read the whole article, which is thoroughly researched, painstaking, and – not to mince words – damning. A group of radiologists, with a distinguished senior scholar at the head, have been interpreting X-rays and CT scans against medical and scientific consensus, sacrificing not just intellectual consistency but the fortunes of coal miners and other workers, whose diagnoses of black lung were thrown into shadow, and whose legal suits found in favor of their employers.

We can point fingers at these doctors. If I didn’t work at Hopkins, perhaps I would go on at greater length here about what their systematic deviation from scientific practice means for patients’ lives.

The more you think about it, though, the more you realize that we are all implicated, in greater or lesser measure, in similar activities. Our motives are perhaps not as venal; the connection to coal company’s payment not as relevant. But inconsistency of diagnostic practice, dealing out judgments, interpretations, and prescriptions not based on the best scientific evidence, and depending on pseudoscientific “lore” under the influence of economic factors are all widespread in today’s medicine.

In fact, if you consider how widespread in today’s medicine is the use of non-evidence-based treatment, you understand that this pneumoconiosis story is only the tip of a very black iceberg indeed.

Cross-posted at the  Talking To Your Doctor blog. 

1. Should I take an aspirin?

2. Should I take a medication for osteoporosis?

3. How often should I get a DEXA scan?

4. Should I see a specialist about this?

5. What’s the best test to keep me from getting cancer?

6. Can you write me a prescription for an antibiotic?

7. Can’t I just get an MRI?

8. What’s the difference between you and a family medicine doctor?

9. What diet can help me live longer?

10. Can’t you just order me some baseline bloodwork?

And a bonus:

11. Why is your hospital better than any other?

Recently I learned about a pregnancy in which the woman was having frequent painful contractions. Only after her eventual delivery was it clear that she had developed a “uterine window,” where the scar from a previous C-section was almost worn through. The woman was told, after her Caesarian, there had been a significant risk of uterine rupture.

But why then was she not delivered earlier? Because, answered the doctor, there are risks to the baby from preterm delivery.

In that case, what was the balance between the risk of uterine rupture on the one hand and the risk of preterm delivery on the other? No one seemed to consider this tradeoff, or at any rate discuss it with this woman. She kept reporting her contractions; she was reassured and pain relief was provided; but preterm delivery was never considered, as if 39 Weeks were carved somewhere in marble.

Such hidebound protocolism is the norm across medicine. The blood pressure must be lower than 140 over 90, because…, well, because guidelines! The hemoglobin A1C must be lower than 7 because the American Diabetes Association said so, because…they are the experts!

Of course, those who recommend these cutoffs are experts. But that does not mean there are never any countervailing concerns or other possible routes. Is there a way to flip the protocol switch to “flexible” in doctors heads?

The Vegas-cavorting executives of GSK recently made news for their off-label shenanigans, promoting medications for uses which had not been approved yet by the Food and Drug Administration. Bad enough.

Is it worse or better, though, to realize that the everyday practice of medicine involves an order of magnitude more questions, and more specific questions, than can possibly be addressed by the FDA? 

I want to discuss with a patient the treatment of her reflux. Sure there are a bunch of approved medications available. We are taught in medical school that the proton pump inhibitors are more effective than the H2 (histamine receptor) blockers. But which of the PPIs are more effective? Which strike the best balance between cost, side effects, and symptom relief? And how do any of them compare to non-pharmacologic treatments?

There are precious few studies to answer any of these questions. Sure, let’s blame Big Pharma. They have blood on their money. But is it any wonder that doctors are susceptible to such inappropriate influence, when the appropriate variety is so hard to come by – not due to any venality, but due to the imperfections of our research system? Studies of one drug against placebo are valued; comparative effectiveness studies that pit drugs against each other are vanishingly rare. And forget about considering cost!

So sure, let’s blame greedy pharmaceutical companies for what they’ve done, but let’s acknowledge the difficult spots patients and doctors are in when they try to find the best evidence to answer clinical questions. If we figure it all out, we can celebrate in Vegas.


Often, doctors and patients request unneeded tests out of either ignorance, inertia, or fear. Recently, I was part of the Good Stewardship Working Group of the National Physicians Alliance, which convened focus groups of different medical specialties to agree on “Top 5” lists of procedures or tests which are medically contraindicated and can cause harm. This work was published in the Archives of Internal Medicine

As a continuation to that work, we recently published a review in the Archives summarizing the reasons why requesting imaging (i.e., X-rays, CT scans, or MRIs) in routine cases of lower back pain without red flags is not a good idea, and quantifying the harm to the patient. Read it here.

Of course, the question outside the scope of the immediate article is this: what happens when the patient, against all recommendation, wants one of these imaging modalities anyway? Should the doctor do it or not?

This question attracts me because it makes things complicated. The interest of the patient comes into conflict with our interest. Do we respect the patient’s statement as it stands, do we encourage the development of a more knowledgeable approach to medical testing, or do we avoid the conversation with some appeal to insurance companies’ coverage criteria? 

In a useful review in JAMA, Brett et al. say: “Actively challenging patients’ requests for nonbeneficial interventions does not subvert properly understood respect for patient autonomy and is consistent with the professional obligation to practice high-quality, cost-effective medicine.” This is true enough. Challenging never subverts autonomy. But decision-making does. So what do we do when the rubber meets the road?

It’s not an easy question. What would you do in such a conflict, as doctor or patient?