Friday, December 5, 2014

How nostalgia killed Dr Singer's capacity for rational thought


Source

A Facebook friend recently has been sending around a 2013 article/toddler-grade tantrum by Dr Jeffery Singer, general surgeon/old man sitting on a park bench. Let me tell you, it's a tear-jerker:

These rules are being bred into the system. Young doctors and medical students are being trained to follow protocol. To them, command and control is normal. But to older physicians who have lived through the decline of medical culture, this only contributes to our angst.

One of my colleagues, a noted pulmonologist with over 30 years’ experience, fears that teaching young physicians to follow guidelines and practice protocols discourages creative medical thinking and may lead to a decrease in diagnostic and therapeutic excellence. He laments that “ ‘evidence-based’ means you are not interested in listening to anyone.” Another colleague, a North Phoenix orthopedist of many years, decries the “cookie-cutter” approach mandated by protocols.
God forbid you should go to work and find rules "bred into the system" that restrict your ability to do whatever the hell you feel like doing. I'm sure no one but doctors ever feels that pain.

This all seems very sad, but when a doctor opines that evidence-based medicine "means you are not interested in listening to anyone" you have to understand that as code for "You aren't listening to me." Because until startlingly recently, that's how virtually all medicine worked; the leading lights of the profession laid down the law, and that was the standard of practice.

There are now standardized grades of evidence, starting with multiple randomized controlled trials and extending downwards through observational studies and case reports and at the very bottom, "expert consensus" (i.e., this is what we believe but we have no proof of any kind.) This quality of evidence is known colloquially in the industry as GOB-SAAT. Good Old Boys -- Sitting Around A Table. Before evidence-based medicine, we had eminence-based medicine.

Dr Singer's hatred of protocols -- things like washing your hands between patients, pausing before surgery to confirm you're doing the right surgery on the right patient, or observing low-risk head injuries in children rather than irradiating their brains -- is common in physicians, but in Dr Singer, it has been amplified by ideology:
Ayn Rand’s philosophical novel Atlas Shrugged describes a dystopian near-future America. One of its characters is Dr. Thomas Hendricks, a prominent and innovative neurosurgeon who one day just disappears. He could no longer be a part of a medical system that denied him autonomy and dignity. Dr. Hendricks’ warning deserves repeating:

“Let them discover the kind of doctors that their system will now produce. Let them discover, in their operating rooms and hospital wards, that it is not safe to place their lives in the hands of a man whose life they have throttled. It is not safe, if he is the sort of man who resents it—and still less safe, if he is the sort who doesn’t.”
What are we to think of a man with a responsible job and most of his career behind him who has the intellectual interests of a socially stunted 14-year-old? These are questions I cannot answer.

To try and interest us in what is essentially an irrelevant paean to his own nostalgia, Singer's lards up his argument with predictions of disaster aplenty (All the doctors will quit! Except for the brain-dead younger generation weak enough to tolerate the intrusion of "so-called 'evidence-based medicine,'") some of which are really funny in their own right:
 In other words, we’re about to experience the two-tiered system that already exists in most parts of the world that provide “universal coverage.” Those who have the financial means will still be able to get prompt, courteous, personalized, state-of-the-art health care from providers who consider themselves professionals. But the majority can expect long lines, mediocre and impersonal care from shift-working providers, subtle but definite rationing, and slowly deteriorating outcomes.
Yes, if nothing is done, we will soon experience the universal free healthcare that delivers better health outcomes at half the cost in the UK and Canada -- a model so wildly popular in those countries that right-wind politicians get nowhere near office without swearing to protect and defend it.

But the real punchline here is his claim of "slowly deteriorating outcomes." The outcomes are better than ours, but in the confines of Dr Singer's Rand-poisoned mind, that simply means they are deteriorating too slowly for the terrible and inevitable decline to be evident to the causal observer.

Back in the reality-based community, things look rather different:
The United States health care system is the most expensive in the world, but this report and prior editions consistently show the U.S. underperforms relative to other countries on most dimensions of performance. Among the 11 nations studied in this report—Australia, Canada, France, Germany, the Netherlands, New Zealand, Norway, Sweden, Switzerland, the United Kingdom, and the United States—the U.S. ranks last, as it did in the 2010, 2007, 2006, and 2004 editions of Mirror, Mirror. Most troubling, the U.S. fails to achieve better health outcomes than the other countries, and as shown in the earlier editions, the U.S. is last or near last on dimensions of access, efficiency, and equity. In this edition of Mirror, Mirror, the United Kingdom ranks first, followed closely by Switzerland (Exhibit ES-1).
Protocols have good and bad points. I deal with frustrating and clinically counterproductive protocols every day. If you are going to spend seven years training a doctor, you should probably leave room in the system for them to exercise their clinical judgement. But many protocols are proven to improve patient outcomes. So if they make our days as doctors a little less fun, so be it. And evidence-based medicine, like free speech, is best answered by more and better evidence, not nostalgia for the days when the stand of care was "I got a hunch." Fuck GOB-SAAT.

Monday, October 20, 2014

Thoughts on Placebos

CarbonKyle, scourge of alternative medicine and pseudoscience of all sorts, has some negative things to say about placebos in the comments here. As a frequent user of what one might term semi-placebos (things I hope will help a little, but are primarily recommended to give a non-harmful treatment whilst the body heals itself) I felt I had to speak up for the humble placebo:

I recommend "The House of God" for a medical perspective on the value of, as Shem puts it "Doing as much nothing as possible."

Allopathic medicine prescribes a lot of ineffective or unnecessary medicine -- these are really placebos by another name. The difference is that while a good placebo is harmless, a bad placebo -- like an antibiotic for a viral pharyngitis or a muscle relaxant for back pain, or (a horror that takes place 250k times a year in the US) an unnecessary heart cath for stable angina -- may have some serious negative health consequences.

Placebos are necessary because treatment is part of the ritual of healthcare. Listening, helping the patient make sense of the problem, and offering treatment all have a value in themselves, apart from the biomechanical efficacy of the treatment (or even whether the diagnosis is correct.)

It's true that many people find the explicit use of placebos unethical, and it may be. But if you were to ask a hundred doctors "Have you ever prescribed a treatment you didn't believe would cure the patient, in order to give a patient something, knowing that the disease was going to get better in time regardless?" 90% of them would say yes (and the other 10% are lying.)

The physician has a different role than the scientist. The purpose of the scientist is to seek empirical truth; the purpose of the physician is to seek healing. And to do that, our mandate is broad:

"Life is short, and Art long; the crisis fleeting; experience perilous, and decision difficult. The physician must not only be prepared to do what is right himself, but also to make the patient, the attendants, and external circumstances cooperate."

Thursday, October 9, 2014

"Informed" "consent"

Dr Orthochick over at Fizzy's place is talking consents.

It's a tangent, but I find myself a huge skeptic/cynic of/about the whole concept of informed consent. This is of course supposed to be our antidote to the bad old paternalistic days when you would want some information about what was going to happen with your condition and your care and your wise old doctor would pat you on the head, take a long drag of his unfiltered Camel, and tell you not to worry your pretty head about it.

But the concept of informed consent does not make very much sense, philosophically, nor is it particularly practical. Informed consent is supposed to include "PARQ," shorthand for "procedures, risks, alternatives, and questions." The big daddy here is risks; that's what we talk most about. The risks of doing a thing and the risks of not doing it. And then you get to chose. Sounds great! But how am I supposed to give you an accurate account of the possible risks of something I want/need to do for you? There are always dozens of things that can go wrong, some minor, some major.

Obviously the significance of each particular risk depends on how likely it is to come to pass, which is a statistical question. But most people (including most physicians) are not good with statistics. We tend to overrate rare risks, underrate things we haven't seen personally, make big changes in response to one bad experience, etc.

Practically speaking, in the best case I tell you and you understand whether a risk is rare or common. But if it's rare, you probably don't need to worry about it, and if it's common, it's probably minor. (Informed consent would be really useful if doctors commonly performed unnecessary procedures that killed or crippled half the people who had them, but for obvious reasons, we don't do that.)

So what you're left with are rare serious risks, to which it will be difficult to assign significance to because they are rare, and common minor risks, which neither you or I care much about.

Sometimes people have strong opinions about their care, and sometimes they have a much greater tolerance for one kind of risk (kidney failure from a coronary angiogram) versus another (a heart attack like the one that killed their father.) And sometime people are done with medical care, and they want to slow down or stop -- that's a really important discussion to have.

Most of the time, though, people don't have the medical or statistical expertise to weigh all the risks and benefits of a particular course of action -- that's what they pay me for. And they know that. Most people are going to do whatever it is you recommend as their doctor (at least in the emergency setting, where I work). And that's really the only rational course, when you hire a professional to help you do something you don't know how to do yourself.

It's not about the doctor thing. If I hire an IT person to fix my computer, and they tell me I have too many zeroes and I need to buy more ones, then that's what I'm going to do.

So at the end of the day, the real "informed consent" boils down to this: Do you trust me? Do you feel like I care about you? Do you think I am careful and competent?

If you do, and I recommend something, you should do it. If you don't, then regardless of what I say, you probably shouldn't.

You can't take the trust out of the doctor-patient relationship. Every patient (with decisional capacity) should have all their questions answered, know what's happening, and have veto over any and all things being done to their bodies. But on a daily basis we assault patients' ears and brains with a mountain of complications and terrible possibilities that they can't possibly weigh and measure, even well, and we badger them to make important decisions when they are hurt, scared, and sick, all in the name of a chimera called "informed consent."

But the only decision they are really making, most times, is: Do you trust me?