Last week, I encountered yet another example of why it’s so important to always read the whole study — not just the press release. In this case, it was actually a report, not a study. A press release from Alzheimer’s International with the somewhat misleading headline, “Smoking Increases Risk Of Dementia” arrived in my inbox, citing a new World Health Organization report that put smokers at a 45 percent higher risk for developing the disease than non-smokers.
When I opened the report, I learned that the “news” touted in the press release was actually just a summary of old research. There was nothing new here. Nor was there proof of causation – the cited evidence showed associations.
As I looked more closely at the report, I found an error that appeared to undermine its conclusions and suggest a sloppiness and lack of rigor.
On page two, the report authors state:
“There is evidence that current smoking is associated with an increased risk of developing dementia and AD” (6, 27–36).
Note the citation numbers here.
Two paragraphs later,
“The above findings are in contradiction with several studies that had previously led to an opposite conclusion, namely that use of tobacco was protective against dementia, particularly Alzheimer’s Disease (6, 34, 36).
Again note the citations. Three are the same.
Something obviously wasn’t right. How could the same research show that smoking may increase risk and was protective? So I retrieved these three studies and here are the actual conclusions:
“Current smoking increases risk of Alzheimer’s disease and may increase risk of other dementias. “ (6)
“Smoking does not seem protective against Alzheimer’s disease or dementia,” (34)
“The risk of dementia and AD increased with ETS exposure. Banning smoking in public areas may help reduce a dementia epidemic worldwide.” (36)
They clearly do not conclude that smoking protects against dementia.
I emailed the PR contact listed on the release and asked how three cited studies can come to opposite conclusions in the same report — and why WHO researchers or editors — or someone at Alzheimer’s International, which put out the press release — did not question this? I also identified myself as a journalist for AHCJ.
The initial response?
I have referred your technical question below to the WHO press officer with whom we are coordinating and to the principal author of thew [sic] report.
In the interim, a quick search of Twitter found posts from the WHO and others, with varying degrees of reporting accuracy. Several global media outlets ran the story, some reporting the same definitive conclusions as in the press release’s headline. At the very least, some of the media coverage appears rushed and incomplete.
After nearly a week passed, the Alzheimer’s International contact finally called me – from Denmark, the site of their annual conference. He said he had heard from the lead researchers, who thought the error must have occurred in the copy editing process. “They were horrified to learn of the mistake. They’re working to track down how the error occurred. The original report has been pulled from the WHO website and a corrected version will be posted shortly.” In reality, the original report, with the error, is still posted as of this writing.
It’s a good reminder that regardless of the reputation of the organization or institution issuing a report, study or press release, read the source information yourself. You never know what you may find.
Related
Should science journalists read the papers on which their stories are based?
