We are all terrible

7 min read November 11, 2015 at 11:07pm on Health, Infection, Rant and Science

You've probably heard that the word "doctor" comes from the Latin docere - "to teach".  The original Hippocratic Oath talked about teaching the art:

I will reverence my master who taught me the art. Equally with my parents, will I allow him things necessary for his support, and will consider his sons as brothers. I will teach them my art without reward or agreement; and I will impart all my acquirement, instructions, and whatever I know, to my master's children, as to my own; and likewise to all my pupils, who shall bind and tie themselves by a professional oath, but to none else.

It's actually dropped out of the modern version (this is the one I took at my med school graduation), but vocational training is still a big part of our job.  Modern medicine realises the importance of providing education to our patients and the community.  Many specialist colleges (including mine) talk about important domains of clinical practice - including Communication, Teaching and Learning and Health Advocacy.

Well, I'm going to put it out there that I think communication of medical information to the public is broken. Broken for lots of different reasons and in lots of different ways.  

People trust their doctor (equal second with pharmacists, behind nurses) and this is great news, because hopefully it means that patients will continue to discuss important issues about their health with their doctor, take advice on board and reach a decision together. Despite the fact that it's pretty well known many doctors are less than impressed by Dr Google, this isn't one of those rants.

If you follow me on twitter, you're probably familiar with me bemoaning the state of science journalism.  And in longer than 140 characters, there's plenty wrong with it, but there's no one factor that's completely to blame.  The whole system from woe to go is terrible and in desperate need of an overhaul.

Scientific knowledge is disseminated (largely) through scientific journals.  Scientific journals are a product (and a very profitable one) and because capitalism, publishing companies want to sell more product.  This means attracting articles which are going to be cited lots.  Having lots of highly-cited articles increases the prestige of a journal, which allows them to sell more product and at higher price. This creates a system whereby researchers want to publish their articles in fancy journals ("I had a paper in ") and journals want to publish articles that make the news ("New research published in the says...").   The flip-side of this is called publication bias - articles which don't necessarily break major new ground are less likely to be picked up, and will be published in less august journals, and therefore less cited. Open-Access publishing attempts to address some of these issues, and hopefully will continue to spread (but that's another story).

The second factor impacting our imparting knowledge is in part related to this.  Because cash is tight in the research and higher education sector, researchers (and universities) need to attract funding.  This year's NHMRC funding round had a record low success rate of grant approvals - just 14.9%.  How do you improve your chances of success?  Well, you get runs on the board by getting articles published in high-impact journals (see above).  If you don't, then the uni starts giving you a hard time. Just like the journals, universities like getting publicity for the clever things their research does ("Researchers at the University of Cleverness have found...") - it helps them attract students, and researchers, and research funding.

Publication of research findings are often accompanied by media releases, which - because of the nature of public relations, tend to go for headline-grabbing articles - how many times this year have you heard a story "may lead to a cure for cancer"?  The media releases get filtered through the Media / Comms / PR team - who (doing no more than their job) jazz up the report for maximum appeal to the journos.

Then, when you reach the journalists, you find that media companies in the new media environment are no longer able to sustain specialist science journalists.  Note that there are a few exceptions in Australia (Norman Swan from the ABC and Janelle Miles (twitter)  from the Courier-Mail in Brisbane spring to mind), but by-and-large the specialists have gone, victims to the cutthroat new media environment.  The non-specialist journalists may then rely more heavily on the media release from the Uni or the researcher.  And again, the subs (doing no more than their job) pick out the juicy factoids from the article to make their attention-grabbing headline.

Of course, attention-grabbing headlines (whether you call them click-bait or "curiosity-gaps" ) are even more important in the internet media world - and the legacy media companies must keep up or continue to haemorrhage revenue on their way to obsolescence.

So is it any wonder that doctors are frustrated by what patients find in doctor Google?  There are multiple layers of people trying to spin the science to improve its appeal.  You find the exciting headline, don't read the details of the article or even if you do, there isn't any useful detail.  There's almost certainly no link to the primary article in the media story (this is a personal pet hate of mine), and even if there is, it's behind a paywall that you can't access.  Assuming that you could access the article, it's written by doctors for doctors and potentially mostly incomprehensible to a layperson anyway.

So what can the medical profession do? Well, not write rubbish clickbait headlines ourselves for a start.

This post (and the twitter rant that led to it) came about because of this story in Forbes, which was, in turn, a discussion of this article in the Journal of the American College of Cardiology.  Macrolide antibiotics are commonly used in the community for respiratory tract infections (even more so in the US) , and azithromycin was the 12th most commonly used antibiotic for inpatients in the 2014 National Antimicrobial Prescribing Survey (pdf).  The association between azithromycin and cardiovascular mortality has been around for a few years (pdf).

This particular study was an enormous meta-analysis (almost 21 million patients in 33 studies), but a retrospective one (with the usual caveats about drawing causal conclusions from retrospective studies).  The conclusions were that exposure to macrolides increased the risk of sudden cardiac death and ventricular tachyarrhythmia (this is a bad irregular heartbeat to have and one of the reasons people get defibrillated). The relative risk was almost 2.5 (ie people with macrolide exposure were more than twice as likely to have SCD or VTA).  For cardiovascular death, the relative risk was 1.31 (ie 30% more likely to die from a cardiovascular condition) - although all-cause mortality was not affected.  

This translates into 118.1 sudden cardiac deaths or tachyarrhythmiae, 36.6 sudden cardiac deaths, and 38.2 cardiac deaths per 1 million courses of antibiotics.

As the Forbes article correctly points out, this isn't exactly news if you're a doctor or a pharmacist - it's commonly known that in some people these medications alter the electrical conduction in the heart in a way that sometimes leads to these abnormal heart rhythms.  It also makes the eminently sensible suggestion that patients should discuss any history of cardiac issues with their doctor before they receive a prescription, and that doctors should consider these factors before prescribing.  

All makes perfect sense. But how is it being reported?


Azithro Google News
Screenshot of headlines via Google News


This sort of health reporting has form in Australia.  

A story on ABC's Catalyst ("the ABC's primary science journalism television series and the only science show on primetime television in Australia" - Wikipedia) in 2013 which purported to cover "debate" on cholesterol-lowering medications (although the story met the ABC's standards for accuracy, it was found to have breached impartiality guidelines). Analysis of prescribing data following this story suggested a large number of people discontinuing their medications.

So here we have an article which finds an antibiotic known to be pro-arrhythmic causing some cardiac arrythmiae, and an increase in cardiovascular mortality, but no effect on overall mortality.  It's being reported in the lay press as "doubling risk of heart deaths" - which is an accurate relative risk, but doesn't detail the magnitude of the absolute risk (ie: 36 / 1,000,000 prescriptions).  See this tweet for one of the best explanations of this important difference that I've seen.

For the reasons that I've outlined, the presentation of scientific data in the press is often less than ideal.  As doctors, we have a duty to our patients not contribute to the FUD by letting our "clarifications" be presented as sensationalist clickbait headlines that feed the confusion rather than clarify it.