From Denmark comes the tip from former Reuters Health journalist Frederik Joelving, who saw a BBC piece and Tweeted: “Someone confusing correlation with causation here.”
— Frederik Joelving (@joelving) April 10, 2015
The headline of the BBC story inappropriately uses a verb that implies causation.
The subhead – or the first line of bold copy – does so, too: “Being overweight cuts the risk of dementia…”
A few lines down, the story makes a huge leap of faith:
“There is no cure or treatment, and the mainstay of advice has been to reduce risk by maintaining a healthy lifestyle. Yet it might be misguided.”
Misguided? Based on this observational data? That’s too big a leap, even when tempered with a weak-kneed “might be.”
There is an attempt to turn the ship around with an “Analysis” section. But the entire story should be analysis and this may be too little too late for some readers. It was for us.
At least the story ends on a strong note of caution:
Prof Deborah Gustafson, of SUNY Downstate Medical Center in New York, argued: “To understand the association between body mass index and late-onset dementia should sober us as to the complexity of identifying risk and protective factors for dementia.
“The report by Qizilbash and colleagues is not the final word on this controversial topic.”
Dr Qizilbash (lead researcher) said: “We would agree with that entirely.”
The BBC wasn’t alone in this. The Washington Post got it wrong: “Being fat in middle age reduces risk of developing dementia.”
HealthDay’s story (here seen on WebMD) required only 25 words to deliver the important message: “However, the retrospective study was only able to show an association between obesity and a reduced risk of dementia, not a cause-and-effect relationship.” This is the second time this week that we have applauded HealthDay’s language on observational studies.
A reminder: we offer a primer for journalists and the general public to better understand that the words matter when describing the results of observational studies. See “Does the Language Fit The Evidence? – Association Versus Causation.”
It is precisely these kinds of stories about observational studies – one day suggesting that overweight raises risk…another day suggesting that overweight lowers risk – that contribute to the decline in public confidence in journalism and in science.
Follow us on Twitter: