Observational study miscommunication of the week: citrus & stroke

A spin around the Web today will give you many stories about citrus fruits and women’s stroke risk.

• Citrus Fruits May Lower Women’s Stroke Risk – WebMD
• But in the body of the story there’s not a word about the limitations of such an observational study.  Only this quote: “Our study supports the conclusion that flavanones are associated with a modest reduction in stroke risk,” says researcher Kathryn M. Rexrode, MD, MPH, of Boston’s Harvard Medical School and Brigham and Women’s Hospital.  Associated with does not = “lower women’s stroke risk.” Association ≠ causation.

It got worse.

• 

  Credit: http://www.sxc.hu/profile/redzonk

  An Orange A Day Keeps Stroke Away – MedPage Today

• Somebody at the usually-exemplary MPT went nuts with that headline and it belies the caution exercised in the body of the article.  “Observational study” was mentioned in the lead sentence.  And later: “(researchers) cautioned that residual or unmeasured confounding was possible despite the detailed adjustment used in the study. Another limitation is that the actual flavonoid content in foods consumed may have differed from levels recorded because of wide variability based on where the food was grown and during what season and how it was cultivated and processed.Randomized trials are needed to test flavanone and citrus foods for reduction of ischemic stroke risk, the group concluded.”

• Oranges Are Women’s Best Friend for Stroke Prevention – EmaxHealth
• No limitations, no caveats – only this:  “Diamonds may sparkle and look good, but for better health and stroke prevention, women’s best friend may be oranges.”  Not a journalistic jewel.

• Eating citrus lowers stroke risk for women – Canada.com pickup of Agence France-Presse story 
• Headline states it as fact – “lowers stroke risk” – but body of story refers to “apparent link.”  Which is it? (Answer: the latter is about all you can say accurately.)

Two noteworthy efforts:

• What Citrus Means for Stroke Risk – ABC News.com.  Excerpt:

“This study adds absolutely nothing to the relationship between fruit and strokes,” said ABC News’ chief health and medical editor Dr. Richard Besser. “The conclusions of the study go beyond the data.”

Researchers noted that the women who consumed the most flavonoids smoked less, exercised more  and ate better, suggesting   they already had an overall healthier lifestyle.

“The things we know that are important for stroke prevention remain,” said Besser.

• Even a tiny brief on the USA Today website was able to say this in just 46 words:

A new study is prompting headlines such as “An Orange a Day Keeps Stroke Away.” But the study doesn’t make a very strong case — and it’s possible other healthy habits or diet choices accounted for the stroke differences seen among citrus lovers, some experts say.

We offer a primer on our site, “Does The Language Fit The Evidence? – Association Versus Causation.” It goes into great detail explaining how many messages about observational studies are miscommunicated, and what wording can be used to ensure accuracy.

Comments are closed.