When medical literature mislead journalists and the public

Perhaps you’ve seen some of the media coverage reporting that “an apple a day” would save as many lives as statins do, with fewer side effects. Or perhaps not — thankfully, this tongue-in-cheek “study” didn’t get the intense coverage received by the new statin guidelines (an issue I recently addressed on this blog). But I think this is a glaring and important example of irresponsible medical journalism, particularly significant because the perpetrators aren’t members of the lay press, but the staff of the British Medical Journal, one of the most respected medical journals in the world.

To briefly review the “study” in question: three bright and creative wags at Oxford came up with an amusing article claiming that you could save as many lives with an apple a day as you could with statin prescriptions.

The ironic nature of this analysis is revealed by the study’s subtitle — “Comparative Proverb Assessment Modelling Study” — and by this comment in the methods section: “No side effects were modeled for increased apple consumption; aside from the distress caused by a bruised apple, and the theoretical risk of identifying half a worm inside, apple related adverse events are not widely recognized.”

A direct look at this study reveals that it is all about statistical modeling, and while there are lots of issues worthy of scrutiny, the key question would be: on what studies were the data on the lifesaving impact of daily apple ingestion based? (Spoiler alert: it’s actually a wild projection from observational studies on “fruit and vegetable consumption,” studies that aren’t directly assessing apple ingestion and, even if they were, fall far short of the strong experimental evidence showing statins’ efficacy.) Put bluntly, the study does not use or provide any direct evidence that apples interact with the cardiovascular system. It doesn’t even come close.

Okay, this is familiar British irony. And let me be clear: I generally love that sort of stuff. (Indeed, there’s a classic BMJ-published “study” that takes a jab at evidence-based medicine, suggesting that parachutes shouldn’t be used because there are no randomized trials proving they work—a paper I routinely use in my teaching because it ironically makes an important point.)

But seriously (and this is serious): look at the press release the BMJ put out promoting this article. I defy you to see any hint of the true, ironic nature of the article in question; you’ve got to dig deeply enough to read the article itself carefully to understand what it’s about.

While it’s true that the study appears in BMJ‘s annual Christmas edition that accepts a wider range of articles for publication (some of them lighthearted) than regular editions, the typical journalist or lay person reading the press release wouldn’t know that. Many media outlets are reporting the study as if it were completely sound. Promotion like this can confuse the public.

In the world of research, particularly epidemiology, we occasionally engage in soul-searching discussions of how mindful we must be of the ways in which journalists, and the public, are likely to mis- (and generally over-) interpret our published findings. I say “particularly epidemiology,” because every week there’s some study showing that eating nuts reduces death by 20 percent, or something similar.

Sure, journalists and the public don’t really understand that this doesn’t mean everyone should start gorging on nuts, but this is part of scientific discourse, and such “association studies” are helpful in contributing to our growing understanding of disease etiology and prevention. So I feel it’s perfectly reasonable to publish such stuff, and we should do our best to educate journalists and the public about what such studies mean (and don’t mean), and let the chips fall where they may.

But when it’s the medical journals themselves that mislead (even in their press releases), we need to say, loudly and clearly: Stop!

Paul Marantz is associate dean, clinical research education and director, Center for Public Health Sciences, Albert Einstein College of Medicine. He blogs at The Doctor’s Tablet.

View 2 Comments >