Now I have a bit of a moral quandry here. Normally I would be loathe to pass comment on research findings without having read the research in full but for some reason (well, I can think of lots) I have been unable to read the full JAMA article. Obviously I am not going to let this stop me though…

In the 10 Mar 07 edition of NewScientist the news section reports on a study into diets which is titled (in the magazine) The Atkins diet works – a bit. The news item begins:

Compared head-to-head against three other diet plans, the Atkins diet has come out on top. In one of the largest studies to date, overweight women lost most weight on the popular low-carbohydrate diet.

Now this seems reasonable enough. The item continues about how, during a 12 month study the sample on the Atkins diet lost more weight than those on the Zone, LEARN (low fat diet based on US government guidelines) or Ornish (lower fat) diets. 12 months is a long time for a study like this and it looked at 311 women between the ages of 20 and 50. The data should be great.

I have no intention of getting into an argument about which diet is the best, or even if the current western obsession with diet makes any sense at all (simple answer, I dont think it does). The thing which caught my eye was the science involved.

Without having read the study itself, I can only assume this was a properly constructed study to generate an unbiased result as to which diet was the most effective at weight loss. It strikes me, this is what the study found out as well.

You would think they would be happy about it…

Given the fact that the diet industry generates lots of money, even the most crackpot (“eat three ants a day”) diets will pretty much make their inventors rich (especially if a fat celeb signs up to it, gets surgery then claims it was your diet…) and you can see people will defend the cash cow.

The commentary about the study seems to think it has failed (which leads me to suspect they were trying to prove one of the other three diets was the best – I wonder who funded the study..) and Gardner (the author) is quoted in NS as saying:

“Was the slight benefit on Atkins due to the low carbs, or the high protein, or the eight glasses of water a day that may have replaced sweetened beverages? We don’t know.”

Is he saying his experiment construction is flawed? Were there so many uncontrolled variables that he can not explain the results? Was he expecting the LEAN (or Ornish or Zone) diet to come out best? (The Zone diet pretty much came out the worst, which is a blow for people who advocate the “equal proportions” approach.)

I am not convinced this is “bad science” as such. From what I can read, the study looks sound, but I am amazed at the unwillingness to accept the conclusions. Adding to the bad conclusions, if you are still curious, there is an entire website devoted to quotes about this study: “Best Quotes from Atkins, Ornish, Zone, LEARN Diet Study” and in here you can see some amazingly bad conclusions from people doing their utmost to ignore the results of this study and maintain their cash cow…

“This is the message of this article — focus on lifestyle and environmental factors and don’t worry about the macronutrient composition of the diet, particularly if you can achieve the NHLBI guidelines of a 5 to 10 percent weight loss,” says Dr. George Blackburn, chair in nutrition medicine at Harvard Medical School. “I think that was my message for the past 20 years.”

Call me old fashoned but I have no idea where he drew that conclusion from given the available information.

Still, have a look, see what you think and if anyone can get access to the full article I would love to know how it reads. (JAMA, vol 297, p969)