Diabetes UK is the UK’s largest and most prominent diabetes charity. Have a look here and you will se the charity proudly proclaiming that: We stand up for the interests of people with diabetes by campaigning for better standards of care. However, I’m doubtful that Diabetes UK is fulfilling its brief in this respect, seeing as it continues to suggest that diabetics should include starchy carbohydrates which every meal (see herefor more on this). You’ll see that Diabetes UK’s advice on this matter starts like this: At each meal include starchy carbohydrate foods such as bread, pasta, chapattis, potatoes, yam, noodles, rice and cereals.
Yet, these starchy staples break down into sugar, and some of them can release their sugar quite quickly into the bloodstream too. And if we eat them in quantity, like we often do, that only adds to their disruptive effects. Now, what rationale is there for diabetics to include at each meal foods that are disruptive to blood sugar? Here’s at least some of Diabetes UK’s ‘logic’ on this: The amount of carbohydrate you eat is important to control your blood glucose levels.
This is perhaps the vaguest and woolliest sentences I have ever read. What does it mean? I suppose what Diabetes UK would like people to take it to mean is Diabetics need to eat starchy carbohydrates with every meal.
However, I reckon there’s another, far more relevant way of interpreting this sentence which goes something like: The more starchy carbohydrate you eat, the more out-of-control your blood sugar level will be, the more ‘diabetic’ you will be, and the more likely you are to start to take medication for this or to need to increase your medication regime. Remember the advice to eat generally sugar-disruptive starchy carbs with each meal comes from the UK’s largest diabetes charity which, it says, campaigns for better standards of care for diabetics.
What sort of care is it referring to, do you think? Because on the face of it, it doesn’t look like nutritional care is part of its remit. And if that’s the case, maybe what’s being referred to here is medical care including medication.
Now, that would help to explain why Diabetes UK recently had a bit of a PR push on the idea that many diabetics are not taking their medication as prescribed. See this story from the Guardian in the UK for a typical instance of how this story was reported. The story, details the hundreds of thousands that are not taking their prescribed diabetes medication, and warns of the ills that may befall them as a result.
If Diabetes UK were so very concerned with the health of diabetics perhaps they could start by giving some decent nutritional advice for a change. How about starting by telling diabetics that the more starchy carbohydrate they eat, the more likely they are to require medication, and the more of the medication they are likely to need over time.
Elsewhere, I read that Douglas Smallwood, chief executive of Diabetes UK, has said it is a “tragedy” that many diabetics do not take their prescribed medication. My opinion is that the real tragedy here is the fact that Diabetes UK gives advice which makes it more likely to need that medication in the first place.
Those of you who clicked on the link to the Guardian newspaper and read it may have noticed that the Diabetes UK research was, in fact, partnered by the Association of the British Pharmaceutical Industry. Now, looking on the Diabetes UK website I can find no mention of where the charity derives its funding. Under ‘Corporate Partners’ Diabetes UK states this (and only this):
UK funds research for a future without diabetes while teaching children and adults to live with diabetes today. Our corporate partners provide vital and valued support to our work.
In the UK, 2.3 million people have been diagnosed with diabetes and more than half a million people have the condition but don’t know it yet. These figures are set to double by 2010.
Diabetes can develop at any age and those with friends or relatives with diabetes will understand how difficult it can be to learn to live with the condition.
People are being diagnosed with diabetes at an alarming rate; each year 100,000 people are diagnosed with Type 2. One in 20 of your employees, colleagues, friends and family will develop diabetes in their lifetime.
Lots of scary stuff about diabetes there, but no detail at all about who the corporate sponsors are and to what extent they fund Diabetes UK.
I was, however, able to find a letter in BMJ from 2003 that draws our attention to the need for charities and patient advocacy groups to declare their funding [1]. In this letter, the author states: Diabetes UK received around £1m from 11 pharmaceutical companies manufacturing diabetes drugs but this is not mentioned in the annual report.
I don’t want to come across unduly cynical, but is it right that a diabetes charity should have a less-than-transparent financial relationship with the drug industry. And is it right that this charity should be giving nutritional advice that, at the end of the day, looks likely to benefit the pharmaceutical industry. And after all of this, should it then go on to partner with that pharmaceutical industry in ‘research’ highlighting the need for people to take their diabetes medication. Or did I miss something?
References:
1. Hirst J. Charities and patient groups should declare interests. Letter BMJ 2003;326:1211
A hundred years ago, or just a few decades ago, carb restriction was the obvious answer for diabetics. And they stayed quite healthy on their fatty low carb diets!
Now there’s insulin so diabetics can have all the carbs they want (and they’re encouraged to eat them!) so there’s no need for anyone to feel deprived. And they get all kinds of complications from high blood sugar/high insulin that are considered inevitable, a natural progression…
So what if you, as a diabetic, would end up blind and amputated in a wheelchair because you didn’t want to “restrict” yourself to a low carb diet? Would you think back and fondly remember all the delicious breads, cakes and pasta you enjoyed, and think it was all worth it?
Maybe not.
>>end up blind and amputated in a wheelchair
Tsk Tsk!
You forgot the kidney transplant and heart bypass.
Cathy also states:
“So not many of these studies met their own < 30% stipulation and appeared to be typical SAD diets. ”
There wasn’t a <30% stipulation. The stipulation was to compare high carbohydrate, high fibre diets with lower carbohydrate, lower fibre diets.
“I’d call that dishonest”
I’d be more circumspect in future if I were you.
“as is including the Hoffman study when it doesn’t support his theory at all.”
That’s not dishonest! That was the whole point of the meta-analysis, to review the evidence, not to only review evidence that supports a particular hypothesis.
Mike Kelly might have been wise enough to follow a low carb approach but what about all the others who take everything the Diabetes Educators take as gospel.
Cathy said:
“And why leave out this low carb study (20%) with 31 subjects, a control group and 44-month follow-up when it fit their criteria?”
Nielsen & Joensson, Low-carbohydrate diet in type 2 diabetes: stable improvement of bodyweight and glycemic control during 44 months follow-up.
http://www.nutritionandmetabolism.com/content/3/1/22
Published: 14 June 2006
You mean leave it out of the Anderson review? Anderson et al. Carbohydrate and Fiber Recommendations for Individuals with Diabetes: A Quantitative Assessment and Meta-Analysis of the Evidence?
http://www.jacn.org/cgi/content/abstract/23/1/5
Published in the Journal of the American College of Nutrition, Vol. 23, No. 1, 5-17 (2004)?
What a lamentable reflection of modern medical scientific culture is Mike Kelly’s remark that his personal experience isn’t evidence. How very confused we have become.
Ross you completely missed my point. I wasn’t “hiding” the fact they were maintenance diets, just making the observation that it seems pointless to use such diets to recommend very high carb eating when so much of what D-UK recommends is based around losing weight. I don’t know what your Diabetes clinics look like but mine are full of very large people. How are they supposed to follow the dictum “lose weight” when they are being told to eat the high carb diet that DOES NOT promote weight loss?
Think about it Ross ” if these diets promoted weight loss don’t you think the study design would have factored this as an endpoint? It’s akin to taking a snowplough into the desert to prove it doesn’t work. My inner sceptic tells me they deliberately designed the diet as “weight maintenance” to obfuscate the fact that these diets cannot promote weight loss. So using this perverse circular logic how can you claim Anderson’s review supports the notion that high carb diet is good for diabetes when 2 of 3 type 2 diabetics are being told to also lose weight. I have repeatedly heard dieticians telling patients that weight loss is especially important for diabetics because fatness impacts their health more than non-diabetics. What dietary advice should they give to the majority of type 2 diabetics Ross given their fatness is so deadly?
Also, the inference that diabetics are too thick to understand anything more complicated than the plate example is grossly insulting ” is that how you want to be treated by a health professional? Patient = dumb?
Mike Kelly I am pleased you can control your diabetes on diet and exercise alone and I agree that the dieticians and other health professionals are not generally “motivated” (not consciouslessly anyway) by anything other than trying to help people. I have worked in diabetes for years and while some dieticians can be arrogant cows (fair comment because I never met a male one) I’d say most were genuinely trying to help. The problem occurs because the drug industry is so intimately involved in the day-to-day activities of these health professionals ” not just wining and dining but in CME too. These drug company reps were not monsters; they were very personable people and I can say with my hand on my heart that (apart from a couple of exceptions) I genuinely liked them. The system is what I have a problem with ” if most “education” comes from drug reps and pharmaceutical monographs, especially when you work in a resource-limited setting and are “grateful” for these “gifts”, then a large amount of bias seeps into the equation. A bias shown by Anderson’s review ” meta-analyses are supposed to include all studies that fit the criteria, favourable or not, but Anderson et al. stuffed their credibility by cherry-picking ” which is unacceptable in such a large review with such far-reaching consequences.
I have also worked in medical publishing and have seen first-hand how institutions that are supposed to be independent are also a bit too beholden to the drug industry, and how medical and statistical data can be manipulated to look perfectly reasonable and acceptable to health professionals.
Lastly, at the risk of hogging the blog here, I want to recount some very real experience of working in diabetes.
1) Adolescent type 1 diabetics that repeatedly and deliberately under-used insulin because they figured out fairly early on that insulin caused weight gain ” this put them at serious risk of complications due to hyperglycaemia as well as death from DKA.
2) Type 1 diabetes patients that had figured out the carbohydrate-insulin issue by their own endeavours and were controlling BG quite nicely with insulin in single-digit IU ” compared to other patients I had needing > 100 IU twice per day!
3) Both type 1 and 2 diabetics being advised sorbitol-sweetened “diabetic foods” are acceptable when sorbitol is deadly to diabetics (if you want more info on this just ask).
4) Advising both type 1 and 2 diabetics to substitute diet cola for HFCS-sweetened cola as a less-fattening option, when pre-formed caramel is extremely high in advanced glycation end-products, which are also highly problematic for diabetics (ditto above).
So Mike, do you think these adolescents deserve to know that insulin-induced weight gain doesn’t have to be inevitable and they don’t have to risk their own precious lives, limbs, sight (the worst retinopathy I saw was in a 16-year-old doing the sub-optimal insulin thing), kidneys and hearts to avoid it? Or that the big picture is not restricted to macronutrient ratios, or should we just all sit back, put our feet up and let D-UK take this mostly erroneous albeit “well-meaning” stance?
“Ross you completely missed my point.”
Cathy, I think that you have completely missed the point of the Anderson review. You seem to want it to be something it isn’t and then criticise the authors of the review and the individual studies because it doesn’t provide the results you want.
I think it would be more valid to bemoan the lack of studies and evidence for the low carb approach to diabetes management rather than criticise a review of the good studies that have been done examining other approaches.
The Anderson review looks fine to me and I can’t find any examples of the authors engaging in ‘statistical shenanigans’ or being ‘dishonest’. Do you want to retract those words?
I also wouldn’t expect to find the kind of massive error that you accuse the Anderson review of (incl. 13 studies that don’t fit their own reviwe criteria) in an open access peer reviewed journal, certainly I would have expected it to have been picked up after 2 years. If you still think you are correct on this point are you going to write a letter to JACN pointing out this massive howler?
“I wasn’t “hiding” the fact they were maintenance diets”
Well, you said that “Not only that but in all the above studies bodyweight did not change. How does this support weight loss in fat diabetics?”. Not only was this statement incorrect (study 34) but I believe that this was misleading when none of the studies had weight loss as objectives, many didn’t have weight loss as measured outcomes and several required maintenance diets to measure the outcomes that were being studied.
Hence your statement “if these diets promoted weight loss don’t you think the study design would have factored this as an endpoint” doesn’t really make sense when the studies were looking at very specific outcomes or were even looking at the effects of a maintenance diet (31). For example:
27) The effects of high-carbohydrate, high plant fiber (HCF) diets on glucose and lipid metabolism
31)The short-term effects of a weight-maintenance diet high in fiber and carbohydrate (HFHC) was studied in seven very obese individuals with type II diabetes mellitus
31)In comparison to a traditional low carbohydrate diet (LC), the effect of an isocaloric high carbohydrate, high fibre diet (HC) upon the insulin binding to mononuclear blood cells of seven non-insulin-dependent diabetics was examined.
33) The response of blood glucose and serum lipids and lipoproteins to a high-carbohydrate, high-fiber, low-fat diet was assessed in 10 insulin-dependent diabetic subjects.
“My inner sceptic tells me they deliberately designed the diet as “weight maintenance” to obfuscate the fact that these diets cannot promote weight loss.”
OK. So when a study is designed to look at ‘the short-term effects of a weight-maintenance diet high in fiber and carbohydrate (HFHC)…in seven very obese individuals with type II diabetes mellitus” or another states that “diets were designed to be weight-maintaining and there were no significant alterations in body weight” and both are in the peer reviewed literature, we should disregard this because you have a suspicion that the authors were up to no good? Forgive me for asking, but do you have any evidence to support this hunch of yours?
Do you still want to insist that the Anderson review should have included a study published 2 years in the future?
“A bias shown by Anderson’s review ” meta-analyses are supposed to include all studies that fit the criteria, favourable or not, but Anderson et al. stuffed their credibility by cherry-picking ” which is unacceptable in such a large review with such far-reaching consequences.”
Cathy, please explain your reasons for making this claim. I’d be very interested in your assessment of the biases, particularly when you call Anderson ‘dishonest’ for “including the Hoffman study when it doesn’t support his theory at all.”. Not a great example of bias is it?
cathy,
D-UK recommend weight loss, because, again, there is solid evidence that a BMI (as imperfect a metric as that may be) > 25 is regarded as a Bad Thing for diabetics.
It is not insulting to suggest the ‘plate’ approach, because, again, there is evidence that it is useful tool in encouraging diabetics to think in terms of proportions (whatever the correct ones may be!) when planning meals. It is not insulting the intelligence of diabetics, but a recognition that for some groups thinking in percentages and grams is not easy – or practical in some social situations.
Obviously for well motivated and attentive diabetics other approaches may work better.
I’m not sure that Anderson et al *do* ignore data that fit the selection criteria. Have you any evidence for this? (as ross points out, they can’t really be blamed for not including data not available at the time). Nor is it a valid criticism to say they include data which doesn’t support their position (that’s part of metaanalysis).
w.r.t your point about ‘diabetic foods’ D-UK are very clear that there is no evidence to recommend them and do not do so. That is their stated position. You at least share some common ground there!
The problems of diabetic misusing insulin for weight loss are not really relevant to the D-UKs positions statements. any diabetic misuing their illness for weight loss needs urgent help.
The relationship between drug companies, health care profs, charities and patients is always important to consider. Dr briffa, in his blog post makes the implication that the reason D-UK reccommend ther carb approach may be clouded by drug company patronage.
Other people have pointed out there is solid, peer-reviewed evidence for their approach.
Dr briffa, of course, has a relationship with a pill-maker and was involved in the formulation. It seems only fair that one considers his relationships when reading his blog post.
John Stone “What a lamentable reflection of modern medical scientific culture is Mike Kelly’s remark that his personal experience isn’t evidence. How very confused we have become.”
Personal experience is a very poor guide to determine truth. The easiest person to fool is yourself. My personal experience could form a data point of a larger controlled study but without that control it tells you only that I think it’s what’s helping me. That is why we use the scientific method.
Cathy
The examples you give are tragic but I’m not clear that they got that way by following the D-UK advice (maybe I’m mis-reading it). The point I’m trying to make here is related to John Stone’s mis-understanding of science. You could be 100% right but you need to do the study to find out. Biochemical obviousness, common sense, personal (even clinical) experience may indicate that there’s something worth investigating but it doesn’t give you the answer.
Mike Kelly
My preference (and Cathy’s, I think) is for low-carb/carb-restricted diets for diabetes (though this is clearly not a view held by Diabetes UK). If it’s science you’re looking for, then you might care to click on the links in comment 107. The conclusions here do seem to be in line with clinical experience and common sense.
Personally, I think we have the answer, but whether anyone listens is obviously another matter.
er, dr briffa, I offered some discussion of the kirk et al metaanlysis in post 112.
Essentially (and the authors suggest this) it seems more work is needed to asses the long term benefits and risks of low-carb approach. How, for example, will the increased protein affect diabetics with renal complicactions? Perhaps low-carb is appropriate for certain sections of diabetics, but not for others? exciting work to be done, i think.
So, do you think there is a sufficient body of evidence to justify D-UK changing its advice?
A body of evidence so large, it can overturn the established body of evidence that exists supporting the current D-UK, american diabetic association, canadian, australian, japanes, indian and european diabetes groups’ positions re: carbohydrate intake and long term nutritional managment of diabetes in type I and II diabetics (how many man-hours of common sense, science and clinical experiene formulated those, i wonder?)
If so, the only mechanism I can see for you to address these concerns is to submit something for peer-review in one of the diabetes journals. A paper with such an exciting set of conclusions should waltz into any high-impact journal, i would have thought?
A real blockbuster of a paper could get into Nature if pitched correctly.
Mike Kelly
No, if a substance makes you ill which is shown to be safe and effective in a study with other people, it still is not illusion that it has made you ill. It is a real effect, like eating rotten fish.
John Stone
Your example does not then allow you to predict what will happen the next time I eat fish. Or what will happen when someone else eats fish.
If a study has shown a treatment to be safe and effective it will have put error bars around both of these parameters allowing you to assess the risk/benefit of the treatment.
A collection of individual anecdotes does not allow you to do this.
I have noticed that intelligent and well informed people can become victims of fraudulent treatments because they equate personal experience (their own or others) with evidence. Please try to accept that the purpose and methodology of the scientific method is to overcome this most damaging of blind spots.
>> So, do you think there is a sufficient body of evidence to justify D-UK changing its advice?
Yes.
The American Diabetes Association has (grudgingly) changed its advice and advocated low-carb diets under certain circumstances.
http://care.diabetesjournals.org/cgi/content/full/31/Supplement_1/S61
My guess would be that more changes will follow.
“In an interesting twist this week, in an updated WedMD article, American Diabetes Association spokesman Nathaniel G. Clark, MD acknowledged in an interview that carbohydrate restricted diets help people with type II diabetes control blood sugar.” http://weightoftheevidence.blogspot.com/2006/06/ada-acknowledges-low-carb-diets-help.html
Best,
Michael
Carbohydrate restriction is the way to treat diabetics!
From Taubes’ book:
“1986 FDA exonerated sugar of any nutritional harm saying “no conclusive evidence demonstrates a hazard” yet the 200-page report had hudreds of articles re sugar causing higher risk of heart disease and diabetes. The FDA interpreted the evidence as inconclusive. Sugar – innocent until proven guilty, fat assumed guilty until proved innocent – so the existence of ambiguous evidence was considered sufficient reason to condemn fat in the diet, particularly saturated fat, while the existence of ambiguous evidence was simultaneously considered reason enough to exonerate sugar”.
There is something a bit weird going on with the way in which these posts have appeared.
I apologise unreservedly for missing the dates between Anderson’s review and the Neilson study.
Isn’t there also the point to be made that by quoting Anderson’s 2004 review as the definitive evidence that high carb/high fibre is beneficial you are ignoring all the evidence since then?
The discussion re high vs lower carb diet was then followed by the statement “High-carbohydrate, high-fiber (HCHF) diets provide even greater benefits on glucose and lipid values for diabetic individuals when compared to low or moderate carbohydrate, low or moderate fiber diets.”
Please note they specifically state “low” when the studies they included did not support the statement at all so no, I don’t retract my statement on dishonesty. While I admittedly made a mistake on dates I am not writing a (several months’ work) review that has long-reaching implications for a large section of the community.
The only thing this study proves IMO is that if you are going to eat a moderate or high carbohydrate diet, high fibre will improve glycaemic parameters over low fibre (but possibly leave you rather flatulent). Confounding factors include the production of large amounts of SCFA (as I mentioned before and nobody has addressed) that Anderson acknowledged as a “major” factor in a previous paper but omitted from the review, and also that the high fibre diet likely included fresh whole foods where the low fibre diet was probably more like SAD ” full of processed junk foods.
I completely agree that long-term studies are needed for low carbohydrate diet, but I don’t see that happening any time soon as there isn’t an economic benefit to do such a study.
Superburger you are repeating a common misunderstanding regarding Paleo diets ” they are not overly loaded with protein so your question re diabetic nephropathy is redundant. And I don’t think the “body of evidence” on low carb eating needs to be larger than that which exists in the prevailing paradigm ” it only needs to contradict the hypothesis. If something were “absolutely true” there would be no contradiction.
I agree that while shopping for food it is difficult for people to think in terms of grams/% of carb/fat/protein unless they are food scientists with vast databases tucked away in their brains, but what most mainstream advice ignores is 1) the science is not “unanimous” in what constitutes healthy macronutrient ratios and 2) that it isn’t just about macronutrient ratios anyway.
And you say “D-UK recommend weight loss, because, again, there is solid evidence that a BMI (as imperfect a metric as that may be) > 25 is regarded as a Bad Thing for diabetics.” ” That is my point; how does a diet that doesn’t promote weight loss get to be the Holy Grail of diabetes diets?
What is wrong with D-UK giving this information (in addition to the “other stuff”) to patients? ” “There is some limited evidence that dietary control of type 2 diabetes with low carbohydrate diet reduces or eliminates the need for oral hypoglycaemic agents, and assists in weight loss, but long-term studies have not been conducted.” That at least would be more honest and balanced.
People have been brainwashed by media hysteria into believing weight is the be-all and end-all when it comes to health, so much so that they resort to the use of dangerous weight loss medications (phen/fen anyone?) ” so it seems long-term effects of synthetic chemicals need only provide short-term and minimal benefit, but low carb diets should be regarded with extreme caution ” talk about a double standard!
Mike the examples I gave were indeed tragic and while the advice wasn’t dispensed by D-UK as such, in this country the D-UK recommendations are pretty much followed. While GI and GL have been incorporated into dietary advice here the 60-65% energy as carbs remains unchanged.
Dr Briffa ” yes I favour low carb eating; not just in a theoretical kind of way but because I had GD through both my pregnancies despite being skinny and gaining very little weight. It wasn’t until I was writing about diabetes about 10 years ago that I realised how high risk I was for type 2 diabetes (along with family history of type 2 diabetes) as I approached my forties. I can only thank my natural tendency to have always eaten fairly high fat/low carb diet ” especially after the birth of my children when my BMI plummeted to 15.5kg/m2 that 1) GD disappeared post pregnancy and 2) diabetes has never reappeared thanks to my “natural” way of eating. I am fairly convinced if I followed “standard” advice I’d be a sulphonylurea-munching patient now.
Mike, I am a non-diabetic, follwoing this from the side lines, but I was struck by your line:
“A collection of individual anecdotes does not allow you to do this.”
I agree with this statement to a point, but the the thing about diabetics is that their condition may be so serious that there are immediate consequences to poor nutritional choice. They also have access to specialised and precise medical equipment to quantify and determine the state of their BG and the effects of vasrious foods upon it.
Thus, I would apply a great deal more credibility to the ‘anecdotal evidence’ of this group than, for example that of a group of dieters who can underestimate and ‘cheat’ on a diet with little immediate consequence, and who have no precise technical instrumentation (scales are crude), to quantify the effect of their diet.
What I am trying to say is that a diabetic has a more ‘scientific’ way of coming to his/her conclusion on nutritional than would a dieter.
Given the number of diabetics who have come to the conclusion that a ‘paleo diet’ is optimal for them, and given the objectives of DUK, don’t you think that DUK should be looking to resource some form of research in to this area or at least encouraging it?
“Isn’t there also the point to be made that by quoting Anderson’s 2004 review as the definitive evidence that high carb/high fibre is beneficial you are ignoring all the evidence since then?”
The point was to demonstrate that there is an evidence base for DUK’s approach, Which there is. Nobody stated that the evidence is definitive.
“The discussion re high vs lower carb diet was then followed by the statement “High-carbohydrate, high-fiber (HCHF) diets provide even greater benefits on glucose and lipid values for diabetic individuals when compared to low or moderate carbohydrate, low or moderate fiber diets. Please note they specifically state “low” when the studies they included did not support the statement at all so no, I don’t retract my statement on dishonesty.”
Your original point was “the Anderson review beautifully demonstrates the statistical shenanigans the low fat brigade resort to when trying to shore up the hypothesis. They claimed that the meta-analysis included 12 studies with a “low carb” group defined as < 30% carb calories”
The review doesn’t have a <30% stipulation. The stipulation for the 13 studies you refer to was to compare high carbohydrate, high fibre diets with lower carbohydrate, lower fibre diets.
But OK, let’s look at the statement in ‘Comments’ section of the review:
“High-carbohydrate, high-fiber (HCHF) diets provide even greater benefits on glucose and lipid values for diabetic individuals when compared to low or moderate carbohydrate, low or moderate fiber diets. HCHF diets are accompanied by significant decreases in all aspects of glycemic control including HbA1c. Of course, these benefits result from increases in both carbohydrate and fiber intakes. Earlier studies have documented the glycemic and lipidemic benefits of increases in carbohydrate and decreases in fat intake [38,40,41]”
So if we look at the abstract of (38) it says:
“The metabolic effects of high-carbohydrate (70%), high-fiber (70 g) (HCHF) and low-carbohydrate (39%), low-fiber (10 g) (LCLF) diets were examined for 10 subjects with insulin-dependent diabetes mellitus (IDDM).”
The abstract of (41) says:
“The findings support the hypothesis that high-fat, low-carbohydrate diets are associated with the onset of non-insulin-dependent diabetes mellitus in humans. Am J Epidemiol 1991 ;134:590-603.”
You said “Please note they specifically state “low” when the studies they included did not support the statement at all”.
It seems pretty clear that they specifically state ‘low’ when the studies they are referring to do support the statement.
You seem to have confused the review methodology. The data were classified as follows:
“The levels of carbohydrate intake were classified as follows: high carbohydrate, 60% of energy; moderate carbohydrate, 30% to 59.9% of energy; low carbohydrate, <30% of energy.”
The reviews were stated as:
“Twenty-four studies met the inclusion criteria and summary characteristics are presented in Table 2. Eleven studies compared moderate carbohydrate, high fiber (MCHF) to moderate carbohydrate, low fiber diets…Thirteen studies (Table 2) examined high carbohydrate, high fiber diets (HCHF) with lower carbohydrate, lower fiber diets”
So, once again, the Anderson review looks fine to me and I can’t find any examples of the authors engaging in ‘statistical shenanigans’ or being ‘dishonest’. Do you want to retract those words?
If you still think the review is dishonest or wrong (including 13 studies that don’t fit their own review criteria) then a) as it’s been available in an open access peer reviewed journal for 2 years why do you think that nobody has picked up on this before? b) are you going to write a letter to JACN pointing out that the review authors are dishonest or wrong and c) if not, why not?
er mike, the ADA position statement recognises that low-carb or low-fat diets are useful for *weight loss* in short term (25 is a risk factor, even if BMI is a crude indicator.) Any eating plan in which calories in < calories burnt will result in weight loss.
Exercise is the other key factor, as not only aids weight loss be important for cardiac disease.
skeptictactoe – of course more diabetes research is needed -on every aspect of the disease – and D-UK have a role to play in that alongside all the other interested parties.
But to those (including Dr Briffa?) who think D-UKs advice is in need of urgent change then why not write it up and submit it for peer-review. Far bigger audience than this blog…..
Cathy “The only thing this study proves IMO is that if you are going to eat a moderate or high carbohydrate diet, high fibre will improve glycaemic parameters over low fibre (but possibly leave you rather flatulent).”
Cathy, the study certainly says:
“For diabetic subjects MCHF diets compared to MCLF diets are associated with significantly lower values for: postprandial plasma glucose”
But then it also gives significantly lower values for “total and low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, and triglycerides.”
But wouldn’t you also agree with the other findings?
“HCHF diets compared to MCLF diets are associated with lower values for: fasting, postprandial and average plasma glucose; hemoglobin A1c; total, LDL-cholesterol, HDL-cholesterol and triglycerides.”
Mike Kelly
I mentione rotten fish – not fish.
There is the reverse problem that evidence which hold good for populations may be useless for individuals, and medicine has to treat individuals. It is nonsense to suggest that the diet which suits you is not evidence, at least in regard to you.
Superburger,
“”er mike, the ADA position statement recognises that low-carb or low-fat diets are useful for *weight loss* in short term (25 is a risk factor, even if BMI is a crude indicator.) Any eating plan in which calories in < calories burnt will result in weight loss.”
Duh.
There is no study in the world that would make them reverse course overnight.
The only legally and politically safe course for them is to very slowly shift position “as current science emerges.”
“Short Term” and “for weight loss only” are both meaningless qualifiers that will slowly be dropped.
Cigarette companies were in the same position – admitting to actively and knowingly killing people is not a way to keep your job.
michael, so what your saying is that if the weight of evidence becomes greater (assuming it does) then the ADA, D-UK, etc, etc, will change their positions to reflect this?
Because that sounds like they will genuinley take on board new evidence, consider very carefully new science and generally do the right thing.
Couldn’t agree more – sounds very sensible. Good science always ‘shifts its position’ as and when evidence emerges.
(although, sometimes, a single piece of work can change the course of science – the british doctor’s study on smoking, watson and crick on DNA, jenner on smallpox, for example)
You cigarette smoking analogy seems a little unfair – there is plenty of honest peer-reviewed evidence for the D-UK position.
Sadly, dr briffa seems to prefer to imply that D-UK’s position is clouded by their relationship with drug companies and that there is already sufficient evidence for D-UK to change their position.
Cathy, I’d just like to ask some questions and would appreciate honest answers to them:
Do you think your analysis of the Anderson review was accurate?
Are you still accusing the Anderson authors of dishonesty?
Are you still accusing the Anderson authors of statisitcal shenanigans?
Are you still accusing the authors of (38) and(41) of deliberately and fraudulently designing their studies?
Given the fact that the Anderson review provides an evidence base for DUK’s advice (albeit advice you don’t agree with) do you still think theyare ‘merely self-serving do-gooders’ that ‘actually do more harm than good’?
Are you still suggesting Ms Spears has blood on her hands?
>>You cigarette smoking analogy seems a little unfair – there is plenty of honest peer-reviewed evidence for the D-UK position.
Once you stop buying in to the “fear of fat” hypothesis, there are many studies that do not support their own conclusions.
This is fairly common in science, or any human endeavor.
There are some “well known truths” that color all thinking.
Social scientists call this an “Information Cascade.”
See Thomas Kuhn’s work on Paradigm Shifts:
http://en.wikipedia.org/wiki/Paradigm_shift
We saw it not that long ago with Ulcers – it was “well known” that the problem was acid, stress, and spicy food.
These “well known facts” were in fact false.
Helicobacter Pylori is the culprit, and the “acid, stress, spice” hypothesis was false.
Everything that doctors had learned about ulcers in school and practice was false.
No evil intent or conspiracy.
All the studies about acid reduction.
All the studies comparing alternate surgery techniques.
All the psychotherapy for stress.
All the bland cooking techniques.
ALL FALSE.
We may well see the implosion of the low-fat theory of heart disease. It is not getting good results in practice. Once that theory is no longer supported, the justification for high carb intake for diabetics is gone.
Nature Publishing Group’s “International Journal of Obesity” just published a chinese study comparing vegetable intake.
They found a correlation higher carbs = higher BMI.
Amazingly, they concluded it was the fat.
Here is a great discussion about how the facts were tortured to fit the current low-fat paradim:
http://www.proteinpower.com/drmike/obesity/another-china-study/
I will find the references to some other studies that found bad things about saturated fat in a high carb diet, and concluded it was the fat. The perfectly reasonable alternative, that refined carbs might be the problem was just not explored at all – it did not fit the paradigm.
>>Sadly, dr briffa seems to prefer to imply that D-UK’s position is clouded by their relationship with drug companies and that there is already sufficient evidence for D-UK to change their position.
Well, he did show my old theory on this mess to be false.
I was thinking it was a conspiracy by general practitioners, kidney transplant surgeons, heart surgeons, and laser eye surgery doctors to keep their waiting rooms full and extract the maximum revenue per diabetic client.
It is just so sad how the high-carb push coupled with lots of medication manages to keep diabetics well enough to work and have health insurance, but sick enough to consume lots and lots of medical services.
Perhaps it is not an evil conspiracy after all.
But the result is the same.
You are claiming that D-UK has its recommendations based on solid science and “the weight of the evidence”. Why are the studies with piffling (~13) numbers of subjects with short follow-up considered “weighty” enough to generalise to large populations? Why is it that small, short-term studies on the “benefits” of high carb are deemed weighty enough but studies such as Samah’s [1] must be interpreted with caution? ” Double standards IMO.
Was Anderson dishonest? ” Yes, especially for omitting the issue of high fibre and SCFAs (which I note you have still not addressed) out of his discussion when he knew it was a significant factor. A very high fibre diet is essentially a high fat diet so it cannot be used to extrapolate the advice to “include starch at every meal”. And he was also dishonest in the comment that his review included “low-carb diet” because clearly it didn’t.
Soft, surrogate endpoints do not convince me of very much anyway. Consider that the ACCORD study found that more intensive BG control was correlated with higher mortality. Soft endpoints do not suffice to prove overall benefit ” I want to know that if I take D-UK’s advice, it will reduce mortality and morbidity. This isn’t theoretical for me; I had gestational diabetes 20 years ago and don’t particularly want to encourage the development of type 2 diabetes now I am in my forties ” so, as for others on this blog, it’s my health and welfare at stake here.
A prospective, randomised, long-term (life-long) human lab study would be impossible (not to mention unethical), so the next best thing is long-term observational studies of large populations. The Harvard Nurses’ Study and the Women’s Health Initiative Study (with some hard endpoints) certainly do not convince me high-carb, low-fat is the optimum diet for human beings.
Nor do I think the long-term effects of such high (70-80g) levels of dietary fibre are known. All fibre is not created equal; some types may be beneficial while other types may actually be harmful. Advising people to eat 70-80g without addressing or specifying which is which is irresponsible and hardly scientific.
I have never accused these health professionals of being motivated by anything other than wanting to “help” the patient ” that doesn’t preclude them having blood on their hands if their recommendations do in fact harm people. Nor does it absolve them of their ignorance of the growing body of evidence on vitamins and supplements in the treatment of disease. Dieticians advise that vitamins and supplements are worthless when there is plenty of evidence for benefits [2, 3], including potentially preventing diabetes in the first place according to experimental studies [4].
In the Anderson review they had a decrease in glycosylated haemoglobin but what happened to their fructosylated haemoglobin and other AGEs or ALEs? Large intakes of vegetables and fruit are known to increase AGEs [5] so excuse me if I don’t wet my pants over the FPG, PPG or lipid reductions.
“If so, the only mechanism I can see for you to address these concerns is to submit something for peer-review in one of the diabetes journals. A paper with such an exciting set of conclusions should waltz into any high-impact journal, i would have thought?”
This statement is either very naïve or disingenuous ” anyone who has been in the system or read anything on the politics of science would know that the peer-reviewed literature does not readily publish dissenting views.
1. Samaha, F.F., et al., A low-carbohydrate as compared with a low-fat diet in severe obesity. N Engl J Med, 2003. 348(21): p. 2074-81.
2. Afkhami-Ardekani, M. and A. Shojaoddiny-Ardekani, Effect of vitamin C on blood glucose, serum lipids & serum insulin in type 2 diabetes patients. Indian J Med Res, 2007. 126(5): p. 471-4.
3. Rodrigo, R., et al., Decrease in oxidative stress through supplementation of vitamins C and E is associated with a reduction in blood pressure in patients with essential hypertension. Clin Sci (Lond), 2008. 114(10): p. 625-34.
4. Lautt, W.W., et al., HISS-dependent insulin resistance (HDIR) in aged rats is associated with adiposity, progresses to syndrome X, and is attenuated by a unique antioxidant cocktail. Exp Gerontol, 2008.
5. Krajcovicova-Kudlackova, M., et al., Advanced glycation end products and nutrition. Physiol Res, 2002. 51(3): p. 313-6.
cathy,
“A very high fibre diet is essentially a high fat diet”
how so? Do no high fibre, low fat foods exist?
The smith paper is for obese people – not all were diabetic for a start – and the authors themselves are extremely cautious about reccomending the diet for long term use. And weight loss was the primary measured outcome. Do you think it fitted Anderson’s inclusion criteria?
But as i think I’ve repeatedly said, new work is interesting and important – and if and when a substantial new body of evidence is created maybe D-UK (and every other western diabetes association) will change their advice. That’s science.
But i still think dr Briffa is wrong to suggest that drug company links were the reason for D-UK to give the advice that they do.
re: submitting for peer-review
“This statement is either very naïve or disingenuous ” anyone who has been in the system or read anything on the politics of science would know that the peer-reviewed literature does not readily publish dissenting views.”
Not my experience, I’m afraid. And even if true, that’s not a reason not to try. Dr Briffa makes some serious allegations about the dietary advice of D-UK, if “lives are at stake” and he has evidence based on “science, common sense and clinical experience” then I don’t see a reason not to challange the status quo by submitting these sources of evidence for peer review.
Cathy – “And he was also dishonest in the comment that his review included “low-carb diet” because clearly it didn’t.”
The review refers, accurately, to “low carb” studies when those studies are described by their own authors as “low carb”. It refers, accurately, to “lower carb” when it states that it will review studies that examine high carbohydrate, high fibre diets with lower carbohydrate, lower fibre diets.
Please explain how this is dishonest. Please point out the statistical shenanigans.
You set up a strawman argument when you say I am “quoting Anderson’s 2004 review as the definitive evidence”. Please point out where somebody has stated that the evidence is definitive.
You set up another strawman argument when you say I am claiming “that D-UK has its recommendations based on solid science and “the weight of the evidence”. ” I haven’t mentioned the ‘weight of evidence’ so why put the phrase in quote marks?
I said “The point was to demonstrate that there is an evidence base for DUK’s approach”. If there isn’t an evidence base for DUK’s approach, please supply the evidence.
You said: “A bias shown by Anderson’s review ” meta-analyses are supposed to include all studies that fit the criteria, favourable or not, but Anderson et al. stuffed their credibility by cherry-picking ” which is unacceptable in such a large review with such far-reaching consequences.”
Cathy, again please explain your reasons for making this claim. I’d be very interested in your assessment of the biases, particularly when you call Anderson ‘dishonest’ for “including the Hoffman study when it doesn’t support his theory at all.”. Not a great example of bias is it?
You cherrypick the results of the review to suit your preconceived notions when you state “The only thing this study proves IMO is that if you are going to eat a moderate or high carbohydrate diet, high fibre will improve glycaemic parameters over low fibre (but possibly leave you rather flatulent).”
You ignore the rest of the findings.
Your statement that “in all the above studies bodyweight did not change. How does this support weight loss in fat diabetics?”. Was incorrect (34) and misleading – none of the studies had weight loss as objectives, many didn’t have weight loss as measured outcomes and several required maintenance diets to measure the outcomes that were being studied.
You display a misunderstanding of the what he objectives of the Anderson review were “especially for omitting the issue of high fibre and SCFAs…out of his discussion when he knew it was a significant factor.” Your reason for bringing this issue into your argument is “If they wanted to compare low vs high carb they should have left other variables like fibre the same because it is a huge confounder.” This is another strawman argument. They didn’t want to compare LC vs HC, they wanted to “review international nutrition recommendations with a special emphasis on carbohydrate and fiber, analyze clinical trial information, and provide an evidence-based recommendation for medical nutrition therapy for individuals with diabetes.” Your criticism seems to hinge on the fact that the review isn’t what you want it to be.
This is backed up by your statement “In the Anderson review they had a decrease in glycosylated haemoglobin but what happened to their fructosylated haemoglobin and other AGEs or ALEs” etc etc. How is the Anderson review responsible for the measurable outcomes of the included studies when all they say they have done is “evaluated all clinical trials that provided glycemic outcome data.”?
Your statement “Dieticians advise that vitamins and supplements are worthless when there is plenty of evidence for benefits [2, 3], including potentially preventing diabetes in the first place according to experimental studies [4].” is also a strawman argument. Dieticians do not advise that vitamins and supplements are worthless.
Instead of defending your view that Ms Spears has blood on her hands you talk instead about ‘these health professionals’ who you accuse of ignorance “of the growing body of evidence on vitamins and supplements in the treatment of disease.” Have the 3 studies you quote changed your clinical practice? Should they have changed the advice DUK give? Wouldn’t you prefer to see “long-term observational studies of large populations” for this kind of supplementation? Or does this only apply to the current advice and evidence base?
In light of the above I’d appreciate it if you could answer these questions, a number of which I note you still have not addressed:
Do you think your analysis of the Anderson review was accurate?
Are you still accusing the Anderson authors of dishonesty?
Are you still accusing the Anderson authors of statistical shenanigans?
Are you still accusing the authors of (38) and(41) of deliberately and fraudulently designing their studies?
Are you still suggesting Ms Spears has blood on her hands?
Research can be interesting and lead to amazing insights, but most of the time they’re fairly useless epidemiological studies that don’t really give any answers…
The reality is that carbohydrates do raise blood sugar. It’s a simple fact. Anyone disagree?
If you’re insulin resistant/diabetic your blood sugar will rise to damaging levels when you eat “normal” amounts of carbs, even if you choose so called slow carbs. The body counteracts this by releasing huge amounts of insulin, or you need to inject insulin.
The problem here is that both high blood sugar and high insulin levels are damaging. I really shouldn’t have to point this out, but it seems like this gets ignored or forgotten a lot of the time, along with basic human biochemistry.
There is no way a high carb diet can be beneficial to a person with a damaged sugar metabolism. It’s a simple fact.
Mike Kelly, you said the following: “Ulf_s. I believe type 2 was only recognised as a separate disease in the 1930s. But you’re quite right, before Banting and Best type 1s were put on low carb diets and mostly died young.”
Diabetes, particularly type 1 of course, was a much more dangerous disease to live with before the discovery of insulin. Insulin has certainly saved a lot of lives and has been a true blessing for countless diabetics!
Before injectable insulin was available, the only treatment was a low carb diet. This works very well for most type 2s, but like you say many type 1s died young. I hope you’re not suggesting that they would have been better off eating a low fat, high carb diet…?
But even though insulin is a real life saver, it shouldn’t be abused. High levels are damaging.
The best way for a diabetic to remain as healthy as possible is to minimize carbs so that the blood sugar stays in the lower range at all times, which minimizes the need for insulin.
Really basic and simple.
Would Ross or somebody here explain to me how eating starchy foods can improve glycemic control?
Seriously.
My meter says that starchy food raises BG rapidly.
Any non-tiny portion of starchy food raises BG too high.
I must be missing something…
Ross, stop complicating the issue. carbohydrate restriction is the best treatment for diabetics – that’s it!
Do you think your analysis of the Anderson review was accurate?
Are you still accusing the Anderson authors of dishonesty?
Are you still accusing the Anderson authors of statistical shenanigans?
Are you still accusing the authors of (38) and(41) of deliberately and fraudulently designing their studies?
Are you still suggesting Ms Spears has blood on her hands?
YES
sue, i disagree. I think there is solid evidence to support D-UKs position at the present time.
Think the authors of the anderson metaanalysis are academics trying to look through a vast body of often confusing evidence and trying to come to some evidence based conclusions.
You disagree (which is fine). But i think you need to accept that the scientists who disagree with you are not fools, do not have a particular agenda (and acknowledge conflicts of interest where present), and also have a genuine interest in good evidence for diabetics.
When two competing ideas exist, it doesn’t mean both positions are honestly held, but sometimes one position has more evidence.
Accusing acadmics of dishonesty is a serious allegation. I strongly suggest you think about the allegation you make and admit it is unfair and unjustified.
>>Accusing acadmics of dishonesty is a serious allegation. I strongly suggest you think about the allegation you make and admit it is unfair and unjustified.
Correct.
Simple “confirmation bias” and groupthink is sufficient to produce this sad state of affairs.
Gary Taubes’ book lays out a compelling case.
Go read it.
superburger, the “solid evidence” you’re talking about is anything but solid, since it’s based on epidemiology. As you say the “evidence” is often confusing. Guess what? Epidemiology can’t prove anything! The “evidence” would likely be a lot more consistent if it came from carefully designed, properly conducted studies.
And please remember: whatever epidemiological studies “prove” is probably bogus if it goes against human biology…
superburger, do you agree that carbohydrates raise blood sugar?
If you do agree, would you please explain the basic biological mechanisms that make them healthy to eat for a diabetic? Why would/should a diabetic eat a lot of fruit, bread or pasta? (And I don’t mean references to epi. studies that may or may not include actual diabetics.)
Sorry, make that “mostly based on epidemiology, junk science or reviews of them” on the second line…
John Stone (back at 175, sorry, I have a job so big gaps inevitable)
Look up lutefisk. Not exactly rotten but it made me vomit every meal I’ve had since I was six; I’m pretty sure my last heave had my socks and underpants in it.
So, using your criteria, all Scandinavians should immediately cease eating the rotten gelatinous mess?
You can’t extrapolate from anecdote to treatment.
Personally, I think there are two separate arguments going on here; both highly interesting:
1) Is low-carb the best method of controlling diabetes?
Of huge personal interest to me and I keep a very close eye on the literature. There’s a lot of work yet to be done here especially as the ACCORD study showed just focussing on the surrogate markers (blood glucose, Hb1Ac, blood lipids etc.) may not indicate better outcomes for the real markers (Dying!!, blindness, Look Ma! No Kidneys!111!eleven).
2) Is D-UK deliberately giving bad advice when they know better and…
2a) Are they influenced by ties to Big Pharma.
I think the other commentators (Superburber et al) have done a good job showing that D-UK are advising consistent with the current evidence base and even if the advice they give turns out to be sub-optimal there is no reason to assume bad faith.
As for 2a I’m still confused over Dr. Briffa’s immunity to the influence of the pharmaceutical ties he points out in others.
Bit of a classic Tu Quoque there, but I’m reminded of an old interview with Mary Whitehouse where she was asked how she resisted the depraving influence of the films she reviewed. Apparently having Jesus in your heart protects you, I suspect I’m seriously screwed then.
Ulf_s
“And please remember: whatever epidemiological studies “prove” is probably bogus if it goes against human biology…”
We know everything about human biology?
Some comments about the ACCORD study: http://diabetesupdate.blogspot.com/2008/06/giant-step-backwards-misinterpreting.html
In short: the people in the trial were pretty sick from the beginning, were advised to eat a high carb/low fat diet and were put on just about every diabetes medication that is known to mankind. No wonder they dropped like flies…
#193: I certainly don’t…
But I do know that a diabetic shouldn’t eat large amounts of carbs, even if some studies claim they are terrific and good for you.
Why feed sugars to someone who can’t metabolise them properly…?
Can anyone of you who advocate a high carb diet please explain the thinking behind this? I simply do not get it.
I can measure my own blood sugar and see how carbs make it go up, so how can it be good for me to eat them?
Good stuff Ulf_s
But I note that the people in the study already being pretty sick is the opinion of the blogger and not one I picked up from the study (unless the assumption is that an American study automatically means sicker people?)
I haven’t seen a study with non-surrogate endpoints looking at carb content so I don’t feel I’m in a position to take a position. I’m handling my disease in such a way as to minimise my blood sugar without drugs but I’m not an evangelist for it. I COULD BE WRONG.
>>Why feed sugars to someone who can’t metabolise them properly…?
Can anyone of you who advocate a high carb diet please explain the thinking behind this? I simply do not get it.
The only rational reason I have ever heard is the fear of fat.
As best I can tell, the idea that fat is so very dangerous that we should allow high blood sugar levels to get low fat.
That whole line of reasoning is flawed – the studies “proving” the dangers of fat… are all in the context of a high carb diet.
#196: I don’t think I’ve ever seen a well-designed study which compares a traditional low carb/high fat diet with the new high carb/low fat diet that can actually show that the high carb diet is superior.
And this relatively new high carb diet is still what is recommended today. Isn’t it scary?
I’ll stick with the traditional advice until some other approach is proven to be healthier…
New Israeli low-carb vs Mediterranean vs low-fat study
http://www.dailymail.co.uk/news/article-1035779/Atkins-diet-safe-far-effective-low-fat-says-study.html
“The low-carb diet was best for reducing levels of bad cholesterol, while all three diets had the same beneficial effect on liver and inflammation function, the researchers said. … The researchers concede that the study has some flaws. Around 85 per cent of the volunteers were men – and the effects could be different for women, they say.”
Dr Eades has posted up a link to a paper detailing the advantages of long-term low-carbing (44 months) for obese, type 2 diabetics:
http://www.nutritionandmetabolism.com/content/5/1/14