Thanks for the clarification BMD. It confirms my skepticism. Lisa your comments add to this conversation as well, thank you very much.
Thanks for your reply, Ed. I’m very impressed with your program and your results. It gives me hope.
I’m not being scientific at all, but my first reaction to the post was, there SURELY wasn’t any Type 2 diabetes in the concentration camps. Good luck in getting people to live that way long term. Were these people hospitalized and minutely supervised in what they ate? It sounds SO much like Allen’s pre-insulin starvation diet!
My second reaction was, well, it ain’t gonna help me. Purely selfish, I know. I really WOULD like to see Type 2 cured, but this study is obviously VERY preliminary.
Thanks to you folks with good scientific minds for your observations!! 
I have a high opinion of Jenny. What she presents is consistently well researched & well written.
Excess weight is associated with insulin resistance so extreme weight loss could explain the change in insulin resistance. Its surprising that the authors claim a rejuvenated pancreas producing insulin again as the reason for the positive effect they noticed. In fact insulin production seems to have gone down, perhaps because the subjects body’s were now using insulin efficiently again.
More questions than answers at this point.
Excess fat in the liver is associated with insulin resistance so perhaps a reduction in liver fat was the reason for the return to a normal metabolism.
I beg to differ. I have a very low opinion of her writing. I think her statement about “pseudo-science” is a good example.
Carb intake or calorie intake? 500/600g of carbs a day is a whopping lot, and doesn’t make sense. On the other hand 500/600 calories a day is starvation level, and I don’t know that any dietitian would recommend that. Could you double check, please?
Lila and BMD - you should read this paper again, because you are reading things that are not in the paper.
Lila - there is no claim about “beta cell regeneration”. There is instead a claim about “beta cell function” - namely that after eight weeks the beta cell function of these subjects was restored. The author postulates that this is because pancreatic fat is reduced, and supports this with measurements of the reduction in pancreatic fat.
BMD - you call these “hand picked” subjects, but this paper is very clear about how they were picked in the Methods section where it states: "Individuals with type 2 diabetes 35-65 years, HbA1c 6.5-9.0, diabetes duration <4 years, stable BMI 25-45 were recruited. Participants were excluded if being treated with insulin, thiazolidinediones, steroids or beta-blockers, with a serum creatinine >150 …"
In other words, these are obese T2’s who have had diabetes for less than four years and who do not yet require insulin or thiazolidinediones. Apparently he is looking for T2’s who are diabetic due to being overweight (not “skinny” T2’s) and who still have sufficient beta cells which are present but not functioning correctly. It would be nice to see these restrictions tested to see which are necessary. But the groundbreaking part of this research is that it actually worked with this subgroup of T2’s - and it worked when no one would have predicted that it would.
You also complain about the sample size. But as you know, a preliminary study with eleven subjects is a very good start - I hope you aren’t saying that a preliminary study should have been done with hundreds of participants because this is never done. It would be prohibitively expensive to test every new untested idea with hundreds of subjects and would never be for a preliminary study - but would be done as a follow on study, which hopefully will be done.
As you say, more studies are needed, and I think the authors of this study would agree. But they were absolutely right to publish with the results they had - “publish or perish” in the academic field forces publication before someone else publishes first.
It’s probably 500/600 calories - not grams - of carb a day? That would be 125-150 g carb per day. But 500 g of carb would be 2000 calories just from the carb - seems unlikely unless they’re running a half marathon every day …
Jag1, when I state that the subjects recruited for the study were “hand picked,” I am referring to the very selective parameters that were used to define which T2s were to be included in the study. The parameters indeed are outlined in the methods section of the article. However, when such narrow criteria are used to define a subgroup, the findings should not be presented in such a way as to generalize to a larger group or subgroup. That’s precisely what the authors do. In other words, the results from the selective subgroup of T2s who participated in the study should not be presented in such as way that appears to apply to a broader group of T2s.
Another problem with the preliminary study is that the findings were not presented as such. The conclusions should fit the methodology in terms of the sample size, the statistical power, and the study design. They don’t. Not only is the sample size too small for the control and intervention groups of this quasi-experimental nonequivalent groups design, the nonequivalent groups (control and intervention) are too nonequivalent in terms of sample size for each group. To explain, the intervention group began with 15 participants and ended with 11, and the control group had 9 participants, all of whom continued with the study. The smaller the sample size per group, the greater the differences between those groups will be amplified. For example, the intervention group is over 20% larger than the control group. The statistical power for this study is very small, too small in my estimation for the authors to draw such strong conclusions.
As I stated in my earlier response, this study points to the need for more studies with a larger sampling, but it supports neither the broad conclusions stated by the authors, nor the sensationalized title of the article. At best, the study points to weak promise. That’s it.
BMD, the subgroup that is being selected is all T2 that develop diabetes because of excess weight. This is the vast majority of T2 diabetics. The test subjects are being tested within a few years of onset and prior to needing insulin - but all T2’s were once non-insulin using and within a few years of onset so this doesn’t really limit the results. Today there are numerous studies being performed on newly diagnosed T1’s - these won’t help me but that doesn’t mean they aren’t valuable tests. So I don’t see this as a problem either with the design of the study or with the way the results are described in the discussion or in the title.
I don’t understand your objection about “sample size”. What are you referring to? There is nothing wrong with a control group being smaller in size than the test group - in fact that is normal. Nor is there anything wrong with subjects dropping out of a study - that is also normal - especially when the intervention is so drastic.The small sample size is perfectly acceptable if the results are sufficiently large to be statistically significant - and they apparently are.
Maybe you’re right and these results won’t be confirmed in follow-up tests. But as far as I can tell the science behind this paper is strong and noteworthy. I’m glad that we at least agree that this study justifies doing further larger tests to try to replicate and broaden these results.
Personally I have no problem with using carb exchanges. That’s how they taught it when I was first diagnosed. It’s all a matter of converting from one way to measure to another. Carb exchanges can be easy to use since it is easier to memorize. Nowadays with smartphones it is possible to look everything up, though that can take a fair amount of time.
Each carb exchange is 15g carb. One “average” (e.g. Pepperidge Farm) slice of bread is one carb exchange or 15g carb. A quick look at this diet carb content (assuming “normal” portions and not the ultra-huge ones that it is possible to get if desired): 30g breakfast; 45g lunch; 45g dinner; 20g each snack; so that would be about 160g carb.
Jag1, again, the way the article is written and the way the findings are being generalized by the authors are both over the top. Therefore, it comes as no surprise to me that the study has already gained a lot of traction as being a supposed cure for T2. For example, take a look at the follow-up article found at the link albertacowpoke provided: http://hubpages.com/hub/Can-Type-2-Diabetes-Cured-with-Diet It’s sloppy editing standards and a lax acceptance policy by the journal referees that have set the stage for the sensationalized follow-up article. I don’t have a problem with a small T2 subgroup being studied, but I do have a problem with the findings not pointing more clearly that the conclusions only apply to this very select group. The way this article is written undermines the credibility of the journal.
As for the sample size, yes, I do have a problem with it being so small, and the findings subsequently being generalized. Instead, the journal article should emphasize that this is only a preliminary study. It doesn’t.
Also, when conducting a study with an intervention group and a control group, the two groups should be as similar as possible in all aspects, including the number of participants in each group. Of course, controlling group size is not always possible. Attrition can and does occur. However, the point I am making is that the smaller the size of the groups, the greater the variable differences will affect the overall results. It is important to filter as much noise as possible, especially when the study groups are very small.
Personally, had I been a referee, I would have sent the manuscript back to the authors and asked for revisions and to resubmit. These are the revisions I would have requested: (a) Rewrite the title to indicate that the study is preliminary, (b) Emphasize in the discussion area that the T2s are a very specific subgroup, and © Provide the SPSS output that is directly related to the study (I would want to see the power value. I suspect it is very weak.).
From a non-scientist here: I would be concerned because Type 2’s who become diabetic because of obesity (avg. BMI in the study was about 33) ARE a subgroup of Type 2’s; there is another group who are at most overweight, and I wonder if it would apply to them. And since different Type 2’s diabetes acts so differently, I wonder whether there aren’t a variety of abnormalities going on. And whether the findings based on this group would be generalizable. It’s too soon to proclaim it a cure, but it IS interesting.
I feel that diet and exercise can do a lot for type 2 diabetics…but not in just 1 week, or even two months. It has to be a life-long change that is made and maintained. A 600 calorie diet can not be maintained for any real length of time, not without extreme discomfort and hunger. So, although the University may have gotten great results in a short time, if these people involved in the study simply revert back to old habits and ways of eating and lack of exercise, the symptoms of diabetes will return in no time.
Incidental reduction in carbs with associated rapid weight loss. Starvation induced ketosis. Personally I’d rather low carb, skip the starving part, and get similar results.
BMD, you can keep saying it, but I’ll keep disagreeing. I absolutely do not find this paper “sloppy” or “lax” - it is in my view entirely consistent with the numerous papers that get published every week. In any case making the revisions you think are so important would not change the results of the paper one bit. And it is the results of the paper, not the way it was written, that are striking.
Also, it isn’t just this paper that has caused a media over-reaction - it seems like every scientific study that is published related to diabetes is discussed in the media as an impending cure - I have a friend who helpfully tells me at least once a year about a cure he’s read about in the paper.
Jag1, I referee four or five manuscripts per quarter for a scholarly journal. A paper submitted to me for review in the form of this one would not get a pass by me. Lack of detail with regard to important aspects of the methodology, overreaching conclusions, and a the less-than-honest title would all have prompted me to send the manuscript back to the authors for revision.
What is at stake here is the credibility of the journal, and in my view, the journal Diabetologia dropped the ball. They were given an opportunity to publish a very interesting article; however, sloppy editing and lax publishing standards did the journal, the authors, and the public all a disservice. Although I find the preliminary research interesting, I find the article itself quite disappointing. As a reviewer, I would have held the authors to a higher standard.
Jag1, we will just have to agree to disagree on this issue. Take care.
The aim of the study was to replicate the results of weight loss surgery using other methods, so this is what the folks doing the study claim.
BMD, due to the anonymity of the internet I have no idea whether you have ever reviewed a manuscript. Maybe you have, maybe you haven’t - I just dont know - but I would be a fool to automatically accept someone’s claimed credentials on an anonymous forum just because they claimed them.
But if you have, then I am bothered by your incendiary language in describing this paper. The points you’ve raised are, in my opinion, at best fairly minor. They are a far cry from outright fraud or dishonesty. And yet you are using words to describe it as if it was blatently deceitful - words such as “less-than-honest”, “overreaching”, “sloppy”, “lax”, “disappointing”, “a disservice”. Really? If your standards are that high, then you must have a hard time reading the vast majority of papers published in any journal.
Yes, I will agree to disagree on this issue. Hopefully we’ll see more eye-to-eye on the next topic we both comment on.