By now you've heard the news: people who eat chocolate are thinner than those who don't. It's the perfect news story, and one that a lot of outlets ran with — ourselves included — but it's also the kind of piece that science writer Deborah Blum says can contribute to public cynicism about science and what it does.
This is a good news story.
And, no, sadly it's not that scientists have finally done that holy-grail study proving that eating chocolate by the truckload induces weight loss. But a group from San Diego did publish one this week that kind of, sort of suggested that possibility.
And the good news is that, for the most part, journalists did not bite (couldn't resist), at least not entirely. That is, while a slew of stories –- more than 600 -– sprang up just about everywhere, a respectable number remained skeptical about "The Willy Wonka Diet" as Wall Street Journal blogger Laura Kruetzer called it.
The study was published as a research letter Monday in the Archives of Internal Medicine with the rather cautious title "Association Between More Frequent Chocolate Consumption and Lower Body Mass Index." The press release from the University of California-San Diego, where lead author Dr. Beatrice Golomb and her colleagues are based, took a slightly less cautious route: Regular Chocolate Eaters are Thinner!
The press release, in particular, proved an irresistible target for medical blogger Yoni Freedhoff, who promptly posted a parody: Holy awesomesauce batman! Eat chocolate, get thinner! If you're not eating it already, maybe you should! Freedhoff wasn't impressed with the actual study either.
His blog, by the way, is called Weighty Matters and he is a physician who specializes in weight loss treatments. So he took some time to deconstruct what the study itself really said -– and what it didn't say.
As Freedhoff pointed out, the San Diego researchers were not originally studying chocolate but rather non-cardiac effects on statin drugs. They asked the 1017 study participants (between the ages of 20 and 85; 70 percent male) to fill out a life-style questionnaire regarding things like diet choices and regular exercise. Participants were also weighed as part of the research and their body mass index (BMI) was recorded.
When Golomb and her colleagues analyzed the results, they were surprised to find that participants who reported more frequent consumption of chocolate had lower BMIs (a measure of proportionate fat in the body) than others in the study. This pattern held even when the chocolate-eaters didn't report extra high levels of calorie-burning exercise activities.
In other words, the questionnaire revealed –- or at least seemed to reveal — a surprising association between self-reported candy consumption and BMI. Interesting, right? And worth sharing. Worth telling the news media –- as Golomb did -– that chocolate was her new favorite vegetable? Let's call that debatable.
Because this was not, you note, a clinical or experimental study designed to establish or find a mechanism for the influence of chocolate on weight. Health bloggers were fairly irate about that, actually. And not just Freedhoff. "Chocolate makes you slim…like ergonomic chairs make you fit," summarized Blisstree blogger Briana Rognlin.
"This study has some serious limitations," noted the Happy Science blog, The Happy Scientist (an anonymous doctoral student) went on to list of them, including: 1) people lie about what they eat 2) the scientists "guestimated" actual calorie intake from the questionnaire, making it difficult to judge the chocolate effect and that 3) the study is behind a paywall, so that many journalists were unable to judge the actual data.
That doesn't mean that the UCSD results lacked fans. On her "What's Trending" blog, Shira Lazar wryly noted the "chocoholic frenzy" that brought us such coverage as "Craving Weight Loss? Eat Chocolate and Shed Pounds", an Angela Daidone story on newjerseynewsroom.com. Or how about this lead from U-T San Diego's Peter Rowe: "Chocolate: melts in your mouth and melts off your waistline."
But, for the most part, even the mainstream media approached the results warily.
"It sounds a bit too good to be true," began the story by Jennifer Corbett Dooran of the Wall Street Journal, noting that the study did not actually prove a link between chocolate consumption and weight loss.
"I take all these with a grain of salt since they're population studies that make associations but don't prove a cause and effect," wrote The Boston Globe's Deborah Kotz.
In a well-researched story, Jacey Fortin, at the International Business Times, detailed the numerous questions that other researchers had raised about the study.
And at NPR's food blog, The Salt, Allison Aubrey wrote:
The researchers found that chocolate's correlation to thinness started to melt away among the participants who consumed the most. They also didn't suss out whether the type of chocolate - white, milk, or dark, which can have varying amounts of cocoa - made a difference.
I'd like to raise a modest concern here that by over-hyping a study like this, institutions may in the end be contributing to public cynicism about science and what it does. I'm heartened by healthy skepticism in the science reporting community but I'll be more heartened when we see a comparable thoughtfulness at the distribution end as well.
To be fair, the UCSD scientists did not pretend that they'd found an obesity cure in chocolate. And although they offered up a possible explanation for their finding –- the known anti-oxidant chemistry of the candy might also have a positive effect on metabolism -– they did not claim to have proved that either. Rather they suggested that it was an idea that needed further study.
In other words, stay tuned for the next round.
SThis post by Deborah Blum originally appeared on Knight Science Journalism Tracker, a blog dedicated to the evaluation and improvement of science writing across the country.