perhaps before people start getting tooo excited and hailing this as the definitive proof that low carb diets are the answer to diabetes, maybe they ought to check out the abstract on Pubmed first...
The effect of a low-carbohydrate, ketogenic diet v...[Nutr Metab (Lond). 2008] - PubMed Result
Researchers have to be a little more circumspect when it comes to subjecting their conclusion to their peers' scrutiny, than the rather uncritical eye of a press release.
when the authors say
Quote:
|
The diet is not easy for everybody.
|
They weren't kidding. Only 58% of the study group made it to the end of the study. Bearing in mind this study only ran for 24 weeks, this does not bode well. High dropout rates do characterise ketogenic diets - they aren't tolerated particularly well. High dropout rates matter because they can bias the study's findings. What matters is why they drop out - it could be because they couldn't tolerate the diet, it could be because the diet didn't improve or maybe even worsened their control, it could be because they experienced significant weight regain. The point is that very often people drop out because the treatment does not work/makes things worse/or has unpleasant side effects. This effect can bias the results by making the treatment appear more effective because you are only measuring the people for whom the treatment work i.e. those who stay in.
The second bit concerns the weight loss - compare the 2. after 6 months the ketogenic diet group lost almost twice as much weight as the low GI group (11.1kg versus 6.9kg) There is nothing unusual in this result - low carb diets produce dramatic weight loss in the first 6 months, followed by more rapid weight regain. After 12 months however there is very little difference between a low carb diet and a standard low fat one.
The problem for the study in claiming that it is the ketogenic diet that causes the improvement in HBa1c is that weight loss is strongly correlated with improvement of symptoms for type 2. The differences in HbA1c reported by the study could be solely due to the differences in weight loss reported by the 2 groups.
Quote:
|
"It's simple," says Westman. "If you cut out the carbohydrates, your blood sugar goes down, and you lose weight which lowers your blood sugar even further. It's a one-two punch."
|
Unfortunately it isn't simple. Westman is rather overstating his case. The cochrane library reviewed the evidence on dietary interventions in type 2, and their conclusions are rather different than Westmans.
Dietary advice for treatment of type 2 diabetes mellitus in adults
When it came to comparing low-fat versus low-carb diets, they concluded that 5 out of the six studies that were eligible for inclusion were at high risk of bias - i.e. poorly designed and implemented - and therefore no firm conclusions could be drawn. If however you take time to read the results section, you will discover that long term (12+ months) there was little difference between low-fat and low-carb for either weight loss, or blood glucose control.
They did however discuss (briefly) the results achieved by Nielsen...
Low-carbohydrate diet in type 2 diabetes: stable i...[Nutr Metab (Lond). 2008] - PubMed Result
Unfortunately Nielson did not randomly select his patients for this study, meaning the study could not be included because it would be at too high a risk of bias.
Better studies are needed or as the Cochrane put it
Quote:
|
more research is required on larger populations and with a strict control group.
|
Westman's study I'm afraid is left very much wanting.
