On 10 Sep 2010 at 3:08, Allen Esterson wrote: > > Query re cognitive impairment: Is it possible that previous studies > showing some correlation between brain atrophy and the likelihood of > developing Alzheimer's is relevant to the claims? > > http://brain.oxfordjournals.org/content/132/8/2026.short > > http://radiology.rsna.org/content/229/3/691.full >
Interesting. But "predictive" is not the same as "causes". And note that while they're happy to tell us about this encouraging- sounding correlation, they're somehow unable to provide a clear statement that they failed to find a difference between placebo and drug. The larger point is not that they can't be allowed some wiggle room. It may well be that their sample size was too small to detect a cognitive effect, or that they didn't continue the trial long enough (However, I understand that the prior evidence for B- vitamins for Alzheimer's is not encouraging.) No, the problem is that they instead buried this inconvenient finding. Imagine that they did find a positive effect. Would you find a similar reticence to mention it alongside the brain results? I won't hold my breath waiting for those promised negative "findings to be reported separately". But even if they do appear, how much publicity do you think they will generate split off from the rest of the study? I now see that _New Scientist_ (NS) has also expressed reservations about the study, but I'd say they got it wrong. (See http://tinyurl.com/NewScientisttake ). NS complained that the sample was too small for the brain atrophy results. Actually, getting an effect with a small sample means the finding must be robust. Also, as we're not talking about a clinical drug effect, criticizing Smith et al by saying the absolute difference in atrophy size is small is not a persuasive criticism. But NS do incorrectly say that no cognitive tests were done. This makes my point. Smith et al's write-up seems to have misled them. Bottom line: If you say in advance that cognitive changes are one of your primary outcome measures (and you should), you'd better report 'em, even if you don't like the way they turned out. And even if doing so will dampen enthusiasm for buying said vitamins. Stephen -------------------------------------------- Stephen L. Black, Ph.D. Professor of Psychology, Emeritus Bishop's University e-mail: sblack at ubishops.ca 2600 College St. Sherbrooke QC J1M 1Z7 Canada ----------------------------------------------------------------------- --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=4718 or send a blank email to leave-4718-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
