Saturday, March 8, 2014

What doctors and anti-vaxxers have in common: Part 1

A few years ago, a Canadian neurologist reviewed the case of every patient admitted to 12 Ontario hospitals  with a stroke between 2003 and 2007. He only studied patients with a heart rhythm disorder called atrial fibrillation (AF) because AF is known to cause blood clots inside the heart that can get pumped out and block a brain artery causing strokes. There is an abundance of clear evidence: people who suffer from AF can dramatically reduce their risk of stroke by taking oral anticoagulants (OACs) that prevent intracardiac clots from forming. These medications are to be avoided in people with bleeding problems, but otherwise, the majority of people with AF ought to take one.

Gladstone discovered something shocking: after excluding patients with bleeding risks, only 40% of people with AF who suffered a stroke were on an OAC when all of them ought to have been, and three quarters of them were under-treated. The data was even more staggering for patients with AF who were admitted with their second stroke – a situation where it ought to be blatantly obvious that an OAC is needed: only 57% of people were being treated and two thirds of them were undertreated. Overall, only about 15% of patients with AF were appropriately treated with an OAC. The conclusion was sobering: if these patients had been appropriately treated, the majority of these strokes could have been prevented.

Of course, part of the responsibility for not being on OACs rests with patients who decide not to take one, but this is one study among many that indicate that physicians around the world routinely under-prescribe OACs to people with AF. Is it because AF is rare and doctors just don’t know about the opportunity to prevent strokes by prescribing OACs? Consider that AF is the most common arrhythmia in adults and that you and everybody you know and love has a 1 in 4 chance of developing it at some point in life. Plus, it’s responsible for 20% of all strokes. Is it because the risks and benefits are unclear? Absolutely not: as I mentioned earlier, over 10 randomized trials done around the world by different investigators involving thousands of patients have all shown the same consistent results: at the cost of a small (0.3%/yr)increase in the risk of bleeding, OACs reduce the risk of stroke by around 65%.

So why are doctors not doing the obvious and prescribing these drugs?

What's the anti-OAC body count at today?

In my next blog entry, I’ll explain the situation, but first, I want you to consider the following common scenario: 

You have a young child who has been perfectly well until he receives a combined vaccine injection for mumps, measles, and rubella (MMR). Around 3-6 days after the shot, your son develops some irritability and he feels hot; his temperature is 38.9 degrees C. You give him some acetaminophen and in an hour, he defervesces and perks up. How likely do you think it is that the vaccine injection caused his fever?

(A) > 50%
(B) 31-50%
(C) 21-30%
(D) 11-20%
(E) 1-10%
(F) < 1%?

Don't Google it. Close your eyes and visualize the situation and try to be as honest as you can. You and your spouse are probably going to bring this very question up when the fever develops. What are you going to say? Tell us by anonymously voting in the poll on the right. (Poll now down.)

(Go to Part 2, and the answer)


  1. Er, this is some bad medical advice. Studies based on very large patient databases have found that for people with CHADS2 scores of 0 (i.e., having none of several major risk factors for AF-associated stroke) the risk of having a brain bleed on warfarin all by itself significantly outweighs the likelihood of being saved from an ischemic stroke, with no other side effects, harm from dietary restrictions, or costs considered. (Younger people with lone AF do NOT have a "five times greater stroke risk" than the general population, as it turns out.) For people with a CHADS2 score of 1, the net benefit, if any, is so minute that any one person's chance of seeing it is tiny.

    Also, I would suggest you avoid the trap of giving harms in absolute numbers ("look at this tiny chance of a brain bleed [that will surely gork you]") while giving benefits in relative risk reduction ("it will cut your stroke risk by two-thirds [which for an otherwise healthy 50-year-old, means less than 1% will benefit]").

    You have a link to Dr. John Mandrola's site. He's the most honest and sensible -ologist I've encountered; I like him a lot. You might read his column on doctors' abuse of the word Need. Anyone who says "everyone with X condition should be on Y drugs" is failing to recognize that some people with X are in groups that will suffer more harm than good, and that since most people who take Y drug don't actually gain a benefit from it, those who can't bear the direct or indirect health or lifestyle effects of the drug may very reasonably refuse to take it. Negating their experience is paternalistic.

  2. Anonymous:

    I share the high esteem in which you hold Dr. Mandrola and echo the kind words with which you praise him.

    You make some points that would have been important had I been trying to write a careful patient-oriented review of oral anticoagulation for stroke prevention in AF.

    You may be interested to know that the latest guidelines of the Canadian Cardiovascular Society, the American College of Cardiology, the American Heart Association, the Heart Rhythm Society, and the European Society of Cardiology all endorse the CHA2-DS2-VASc score rather than or in addition to the CHADS2 score on the basis of high quality evidence including large population based studies. Patients with CHADS2 scores of 0 or 1 can be better risk stratified, identifying those who really should consider using an OAC and those who need not. These revised scoring recommendations categorize the majority of patients with AF as eligible for oral anticoagulation, so my statement that the majority of patients with AF ought to take an OAC is true. This is not to say that all patients ought to take one, of course, and nothing I have written endorses medical paternalism nor avoiding individualizing treatments.

    Thanks for your interest in my blog.

  3. Thanks for your follow-up. I have two concerns about the CHADS2-VASC score: First, it's used in efforts to coerce all female AF patients onto OAC though sex is the weakest of the predictors in the score and has been reported to be significant at all only in over-70s or in the first year after diagnosis. Second, it's used only to make people with such-and-such a CHADS2 score feel that their risk is "even higher than" those in the old studies of stroke rates - themselves not fully applicable to present populations, as overall risk has declined - but it's never pointed out to people with a given CHADS2 score and none of the VASC factors that their stroke risk must therefore be lower than the risk quoted from the study.

    I did not know at the time of reading this essay that you were a cardiologist; I assumed you were a philosophy type with particular health interests. I've seen a number of cardiologists in action because of a family member's former health problem - he suffered a life-altering malpractice cascade, unfortunately, and has sworn never to see another one - and some of them had a very poor grasp of statistics and a good command of you're-gonna-diiieeee doomer rhetoric that seemed, and sometimes clearly was, intentionally coercive. I must admit to great resentment of the attitude.