Particular thanks to both Mark and Stephan for copies of the diazoxide paper by Astrup's group. Looking forward to posting on this one but not sure how time will pan out in the next couple of days! But just goes to show, you REALLY do need the full text, otherwise you get the impression that the two studies conflict each other, and that Astrup is, well, I'd better not put down my opinion of Astrup!
Look here:
The Danish diazoxide group: Can't write a discussion or understand their own results-
Due A, Flint A, Eriksen G, Møller B, Raben A, Hansen JB, Astrup A.
Location: Research Department of Human Nutrition, Royal Veterinary and Agricultural University, Frederiksberg, Denmark
Compare that group to the Danish flavanoid wash out group-who fought tooth and nail to publish a controversial result:
Young JF, Dragstedt LO, Haraldsdóttir J, Daneshvar B, Kall MA, Loft S, Nilsson L, Nielsen SE, Mayer B, Skibsted LH, Huynh-Ba T, Hermetter A, Sandström B.
Location: Department of Human Nutrition, The Royal Veterinary and Agricultural University, DK-1958 Frederiksberg C, Denmark.
Anyone spot any overlap at all between the group members? Two sorts of researchers perhaps, one with a better eye to future funding. Oh, did I say that?
Peter
This comment has been removed by the author.
ReplyDeleteLong-term studies rule. Total M & M (mortality and morbidity) are truly the only important measures, IMO. A study looking at markers won't show which group will live longer and be more free of disease. Studies don't look at long-term effects any more. They look at risk factors, markers, surrogate end-points, correlations, associations, etc. This is the only way they can play the funding game. They can't do long-term studies and show that saturated fat is harmful, unless of course they add sunflower oil to a diet containing coconut or beef fat, for example.
ReplyDeletehttp://cancerres.aacrjournals.org/cgi/reprint/45/5/1997
So they do the studies like the old milk-shake-and-carrot-cake made out of coconut or safflower oil. Such a study ignores the most vital issue: which group would live the longest? Which group would get more disease? I say the ones eating safflower oil will die faster, and more painfully than those fed coconut oil.
Studies need to start measuring the of total mortality and morbidity. I don't think anyone would appreciate less heart disease, in exchange for more strokes, cancers, suicide, and accidental deaths. But that is what many studies tacitly assume. (Ancel Keys being a good example.)
Bruce,
ReplyDeleteInteresting comment about the funding game distorting how the science is done.
The internet provides a new opportunity for online studies where motivated people (like posters here) report their diets and health markers to a freely-available database. Obviously there are many problems when a study ceases to be in a controlled setting. One would instead be trying to work with a mix of data from widely-distributed and self-selecting people. But there must be great possibilities too.
(Although even as I write this I began to wonder if it would be workable in a world with radical groups like PETA who might want to push a specific agenda about diet),
Paul.