The Associated Press' science writer, Malcolm Ritter, wrote about a study published in Wednesday's issue of Neurology
. At SFgate.com the article had the heading Big Belly Boosts Risk of Later Dementia
This is exactly the kind of headline that makes Sandy Szwarc, the Junkfood Science
blogger, see red. In an article with the headline Computer Gaming: Your waist doesn’t really go to your head
, she picks apart the study and the media coverage of it.
Szwarc starts out by discussing how to recognise when supposedly scientific findings are used for marketing purposes. Then she goes on to point out the flaws in the study. Some of these are:
Dementia was defined as corresponding to certain billing records in Kaiser Permanente's invoice system. Among the codes included was the one for "memory impairment." Needless to say, not all memory impairment is dementia, not even for the elderly. Even Wikipedia is more stringent when defining dementia
as "the progressive decline in cognitive function due to damage or disease in the brain beyond what might be expected from normal aging."
According to the Merck Manual of Geriatrics "Diagnosing dementia is extremely imprecise and requires a very thorough physical evaluation to differentiate it from benign age-related memory loss. It also needs to be differentiated from a lot of treatable conditions that mimic dementia among elderly, such as the effects of illness; prescription medications; hypothyroidism; vitamin B12 deficiency; depression and isolation; and poor oxygenation due to lung, heart or circulation problems."
75% of original cohort was excluded and there is no data on how they differed from those that were included in the study. In addition, although the 2008 study reported this week uses the same data as one from 2005, there are unexplained differences in the results.
Finally, this was a retrospective study, i.e. it looked at records of past events and tried to draw conclusions from these. Retrospective studies are considered the weakest type of observational study and should be used mainly for generating hypothesis.
Read the AP "science" article
and Szwarc critique
of the study side by side to get the full view. Szwarc also has a number of interesting articles on the so called obesity paradox