Magic Weight Loss Fruit: Three Reasons for Doubt

In the past week, headlines have been full of claims that flavonoids in fruit such as strawberries and blueberries are “the secret to losing weight.” They stop just short of announcing the discovery of magic weight loss fruit.

The root of this nonsense is a Harvard study that proves no such thing. In fact, what the study found is a small statistical association between self-reports of consuming fruits containing flavonoids and a lower probability of weight gain in three studies over 25 years of observations.

As usual, health reporters have taken the bait from a tabloid medical journal — the BMJ — and wildly exaggerated the significance of an already marginal finding. Why would one question this research? Here are three reasons:

1. The potential effect is small. The difference in weight gain for people who were thought to consume more flavonoids (one standard deviation) was less than a fourth of a pound. Professor Andrew McDougall of Montclair State University explains:

   For the sample sizes used in the BMJ article, statistical significance is almost certain with an effect size (in SD units) of 0.02 at N > 20,000. This is a pragmatic issue as to the relevance of the authors’ finding, not a statistical issue.

2. The methods have questionable accuracy. Estimates of how much flavonoids these subjects consumed were derived from dietary self-reports, which have serious issues and require cautious interpretation. People systematically shade the truth when reporting what they’ve eaten. Adding to that problem, we have the additional problem of investigators making assumptions about flavonoid content in the foods reportedly eaten. So it’s hard to argue that we really know how much of these substances have been consumed.
    
    
3. Correlation proves nothing about causation. All this study found was a weak association. It’s entirely possible that people who report eating more of these fruits have other characteristics that lead to less weight gain. It also could be that resistance to gaining weight over time causes a natural preference for these fruits.

 
None of this means that there’s anything wrong with eating blueberries, strawberries, or any other fruit with lots of flavonoids. On the contrary, they are delicious. But if you use this kind of pseudo-science to justify going on a berry binge, the effect you get might be something other than weight maintenance.

“If you torture the data long enough, it will confess to anything.” — Ronald Coase

Click here to read the study and here to read more from MedPage Today.

Blueberries, photograph © Maira Gall / flickr

Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.

February 1, 2016