Low-Carbohydrate Diets and All-Cause Mortality:

[mo1905]

Good initial post which is now a "copy and paste" contest lol ! Too much testosterone !


Sent from the Diabetes Forum App

 

[Yorksman]

Re: How & why scientific reports are published

Should we expect professional peer challenges? When the objectives, methods & conclusions are published & reasonable, what is there to challenge?

Well you interject yourself into my reply to jddukes so perhaps you are at cross purposes but to answer your question peer challenges are made by other peers when they question the validity of the objectives, methods & conclusions to use your terminology. These are usually on the basis of flawed data, invalid models or the use of unrealistic assumptions which the models rely on, but may also be due to an overemphasis of the interpretation of one outcome to the detriment of another, equally possibile outcome.

You wrote "We have already seen that many aspects of the report are questionable" but I see no peers raising those questions. Where are they?

I am not vox pops which is essentially an uninformed public opinion. I am a professional scientist (B.Sc. Hull, 1961, & have worked as a professional scientist all my life. I've been diagnosed diabetic for 12 years, so I have rather more than your 4 months experience.

An emotional appeal to authority followed by a rather babyish comment. You cannot expected to be taken seriously if you behave like that.

Passion and prejudice masquerading as reason and supported by bitterness and acrimony. I'm not going to waste my time.

 

[Robinredbreast]

Good initial post which is now a "copy and paste" contest lol ! Too much testosterone !


Sent from the Diabetes Forum App

Definitely, after reading a few sentences, I start to lose the will to live. :yawn: :lol:

RRB

 

[phoenix]

A pity people haven't stuck to the actual meta analysis
I haven't seen any real discussion of it. It used normal procedures for the selection of studies for inclusion. Contrary to one post on here and a similar one in comments on the paper it is not the study discussed by Denise Minger in 2010. It isn't a study looking at glycemic control and diet, it is what it says on the box looking at cohort studies of normal (non diabetic) people eating their habitual diets and the associations with mortality over time. It is not claiming to look at high fat diets though I think it is probable that most higher fat diets are also higher protein (In the Swedish study quoted by Ian, people in the low carb group were advised to eat double the protein advised in the higher carb group))

They found no RCTs and this lack is stressed in their conclusion. Unfortunately it is the case that any RCT over a long enough time to show up differences would be horrendously expensive. Moreover getting people to stick to a prescribed diet over time is notoriously difficult (as M Daly pointed out to Ian when replying to Ian about the results of his 2 year trial) As far as I can see the Swedish study was not a controlled study at any point beyond the first 6 months )

Since the post has gone on to discuss diets for weight and glycemic control in T2 , there is a fairly recent meta analysis on this.

Effects of low carbohydrate diets on weight and glycemic control among type 2 diabetes individuals: a systemic review of RCT greater than 12 weeks.
The Nielsen trial is included in this .
conclusion

Although low carbohydrate diets may appear to be effective over the short-term on weight loss in non diabetic individuals, our review indicates that differences on weight, A1C and lipid profiles changes over the long-term comparing a low carbohydrate diet with a low fat diet, a usual care diet or a low glycemic index diet were not consistent and conclusive. Therefore, further investigation on the long-term effects over cardiovascular outcomes and safety in subjects with type 2 diabetes is needed.

http://scielo.isciii.es/scielo.php?scri ... so&tlng=en
This analysis has the added interest of an independent peer review performed by the Centre for Reviews and Dissemination at York University .
(part of the National Institute for Health Research)
http://www.crd.york.ac.uk/crdweb/ShowRe ... 2012031605

Lastly a review paper on diets and diabetes that reviews several diets and has a title that perhaps says it all
Is there an optimal diet for patients with type 2 diabetes?
Yes, the one that works for them!
http://www.bjdvd.com/content/13/2/60.full.pdf+html

 

[Daibell]

Thanks for last link; very interesting read. The thing that I note is that when the recommendations changed during the 70s/80s from low-carb diet to high-carb diet the stress was on high-fibre carbs and high-fibre foods. Who would disagree with high fibre foods. The problem is that Establishment started using the word 'Starchy carbs' which I have never had a satisfactory definition of in this context. The term 'starchy carbs' is often followed by the usual list of suspects such as bread, rice, pasta and so on without any clarification of them needing be low-GI ones. To me this is the problem as uncooked starchy foods such as whole grains can be good and high fibre, low-GI but when processed and cooked can become high-GI but the Establishment still refers to these foods as 'Starchy'. Fortunately NICE now uses the term low-GI etc in it's Pathways document but I still come across diabetes documents from respected companies and organisations that use this misleading word e.g. DUK website:

"For good health, most of the carbohydrate you eat should be from starchy carbohydrate, fruits and some dairy foods. Carbohydrate from added sugar or table sugar should be limited." It then lists examples:

"Starchy carbohydrates include foods like bread, pasta, chapattis, potatoes, yam, noodles, rice and cereals."

I rest my case......

 

[Yorksman]

Lastly a review paper on diets and diabetes that reviews several diets and has a title that perhaps says it all
Is there an optimal diet for patients with type 2 diabetes?
Yes, the one that works for them!

The dietary requirements of an individual have a parallel with an individual's response to different drug therapies:

We do not have one drug fits all therapies and an individual's response to foodstuffs varies according to his or her genetics.

 

[IanD]

Good initial post which is now a "copy and paste" contest lol ! Too much testosterone !

Sorry Mo, but the initial post made a scary point that cause some low-carbers to question the long-term safety of their diet.

A new study recently out suggests that long term low carb diets are far from safe and are actually shown to increase all causes of mortality.

That study was not concerned with diabetics. My posts were specifically concerned with diabetics, including quotations from major & recent studies by Diabetes UK. While they still question the long term safety of low carb, it is ONLY because they think there might be problems, even though NONE have been reported in the reports they cited.

Despite the OP claiming long term low carb causes increased risk of mortality, that increased risk was marginal, non-specific & not related to diabetes.

What is not in doubt is that T2D is reckoned by the NHS to be progressive, & that following their "healthy eating" plan may help but the disease will still progress.

We are all looking for a solution that will stabilise our condition. Reduced carb is the best option.

Phoenix:
Lastly a review paper on diets and diabetes that reviews several diets and has a title that perhaps says it all
Is there an optimal diet for patients with type 2 diabetes?
Yes, the one that works for them!

Obviously true, but equally obviously unhelpful. How long do we have to follow the NHS/DUK starchy carb diet before realising that the disease has progressed, & the diet did not work? For me, 7 1/2 years to become critical, with debilitation leg muscle pain, but some effects were apparent years before.

How long before an alternative diet shows evidence of working? For me, a few days, as fasting BG dropped from about 7 to about 6; & 3 months to be completely out of pain. Most problems experienced during the first 7 1/2 years have been improved: chronic tiredness; early stage retinopathy; peripheral neuropathy; as well as reduced fasting BG. Reduced kidney function (eGFR) is stable at above 60.

How long before problems with long term low carb show the damaging effects we are warned against? For me, 5 years & no sign of problems.

 

[jddukes]

If there were any professional peer challenges, they'd be listed. Where are these challenges? All I have seen are some vox pops published on web forum.

Strip down medicine to its basic elements and you can usually safely conclude that if enough people report a similar finding than this could be considered the "norm." Many people here, and elsewhere have been low-carbing and doing it for years with only improvements to their health. This seems far from random noise, far from being at the tail-end of the bell-shaped curve, but actually the norm.

You write about the public experience of low carbing. The vox pops I refer to are criticisms of peer published papers by anonymous people. They are completely different things.

Did you not read the majority of my post? You cannot take one part of what I say out of context and say something completely different - that just is not fair. I think I was fairly clear in my post that I was commenting a lot on criticisms of peer-published papers by the general public to be a completely appropriate thing to do. That is the premise of science - not to the elite, but to anyone intelligable to be able to analyse data which (should be but isn't because of greedy publishers not wanting open access) is in the public domain. That is why we have sites like pubmed, google scholar, why people push for open access. The general public are not made up of dumb individuals. Many individuals are smart and able to dissect a peer-reviewed journal and point out the faults. Just because someone is not at PhD level in that chosen field does not mean their input is unvaluable and should be dismissed - it should be considered. Many non-scientists make very good points about a peer-reviewed article's faults; conversely many emminent scientists make very bad judgements or even peer-reviewers do as well. Science is not exclusive to the "elite." I'm quite surprised that you work in science yet give the impression that because something is peer-reviewed it must be correct/or without flaws. That is one of the first things you learn as an undergrad – critiquing any publication, regardless of journal impact factor, author, etc.

As for particular genes well sure I agree with you but that does not prove from the obvious – despite our genetic variation generally biological effects in humans fall under a bell-shaped curve. This means there WILL be something that works for most people and there WILL be a fraction of people at either extreme. That is the point here – it is not binary (and for diabetics it "seems" the norm/most consistent for best glycaemic control is low carb diets).

A pity people haven't stuck to the actual meta analysis
I haven't seen any real discussion of it.

*FACEPALM* did you not read my first post in this thread? I spent an extensive time going through the methodology and appropriateness of this meta-analysis (to this forum) yet you are saying no one has discussed it? Well I am clearly wasting my time.

It isn't a study looking at glycemic control and diet, it is what it says on the box looking at cohort studies of normal (non diabetic) people eating their habitual diets and the associations with mortality over time.

This statement is misleading – it is specifically looking at “low-carb diets” and it does not matter if a study “does what it says on the tin” as if the design is flawed, or erroneous, how can useful conclusions be made?
Flaw #1 – definition of low carb inconsistency from trial to trial
Flaw #2 – not looking at healthy people – looking at a range of people which they have “risk-corrected for” (see – someone is now analysing the actual study as you wished...but I wrote about this a few pages back). Many of the groups have not been corrected for diabetes, for example, and statistically this is incorrect given the cohort size.
Flaw #3 – too large a range of time for inclusivity of studies: 2 of the studies are in the 80s where detection of disease or risk factors would be of a different stringency. Dietary advice and definitions would have altered. Grouping studies that were done 20 years apart may not be wise in this area for analysis.
There are more flaws, but again the authors do a good job of pointing out the limitations of the study.

This does not mean the study is useless, it is just that there are large limitations on its usefulness given many assumptions, inconsistencies and definitions.

This is not a “general” forum for discussion of peer-reviewed papers. It is GOOD that this type of discussion occurs, however people saying we are taking this too far into the diabetes slant when the study did not ever intend to address that have missed the point completely. We are in a diabetes discussion forum! People on here low carb to control their diabetes. Someone posts a review about low carb with potentially serious implications (suggesting it increases our chance of death!), we have a duty to discuss it in its context HERE – i.e. what does this mean for the diabetic. 75%+ of my criticisms of this paper are with that in mind. If someone wishes to discuss what this paper was meant to address – “healthy” individuals without this slant then it should be posted in a general forum, not a diabetes discussion. We are entirely right then to critique how this paper applies to diabetic low-carbers – and that is justified to be debated by “non-peer-reviewers” i.e. general public who are trying to discuss the implications of such research for their chronic diseased state.

J

 

[phoenix]

I apologise for missing your original post, I read the other two and not that one. My response on methodology was to the person who had suggested the choice of studies for inclusion was biased/cherry picked.
I agree with you that the definition of low carb varies from study to study. I also note that some of them show a
I stand by my statement that the studies included in the analysis were of normal peoples diets and that for the most part these people were healthy at outset hence no need to adjust for diabetes or CVD.

However, most of the target populations were free of chronic disease at baseline and it is less likely that the dietary habits had been modulated according to their previous health status. A dose- response of relative risk was confirmed in few studies, which might make the results less plausible

I didn't read all the studies but this was certainly the case in the Greek Study.

(ah want to say more but have to stop here)

 

[Yorksman]

Did you not read the majority of my post?

I am not under any obligation to read anything you write so please do not project blame onto me if I don't read beyond the first few sentances.

I think I was fairly clear in my post that I was commenting a lot on criticisms of peer-published papers by the general public to be a completely appropriate thing to do.

That rather depends on how familiar they are the the corpus of literature, hence my post re. fig1 and table 2. Contrary to what you write about me taking it out of context, I did grasp your point, I just happen to disagree with it.

Just because someone is not at PhD level in that chosen field does not mean their input is unvaluable and should be dismissed - it should be considered. Many non-scientists make very good points about a peer-reviewed article's faults; conversely many emminent scientists make very bad judgements or even peer-reviewers do as well. Science is not exclusive to the "elite." I'm quite surprised that you work in science yet give the impression that because something is peer-reviewed it must be correct/or without flaws.

You shouldn't assume that I work in science. I graduated in mathematics. My interest in genetic anthropology is personal and limited to a narrow field of study of the Y chromosome. Studying the peer literature has enabled me to advise in some small capacity, professional historians and archaeologists, another field of study that interests me. Correspondence with some leading researchers in both fields takes place at a meaningful level only because I have a grasp of the basics.

It is not my experience that the general public know more than the researchers especially when it comes to statistical inferences based on stochastic models. More to the point, non specialists invariably miss the significance of the caveats or constraints of the models used.

The problem with an ad hoc mining of the peer literature is that it often leads to spurious conclusions because context is lost. Important caveats for example carefully explained by Wilson et al in 2000 and reinforced by Weale et al in 2002 and Capelli in 2003 get ignored by even semi specialists such as Oppenheimer. In 2005 Santos Alonso warns that a circular argument has entered the literature as a result and published data which shows one popular interpretation to be false, yet the error was still propagated. Not until 2012 was it shown to be wrong and even then, not to everyone's satisfaction. The point is unless you are familiar with all of it, simply dipping into this or that study out of context will likely result in a false inference.

What we witness in this thread however is a rather more base example of bruised egos and the fact that someone may hold an alternative view, or even suggest that a view is worth considering, is percieved as a slight. I have heard 'holier than thou' arguments before but never 'more diabetic than thee'. None of this is science. It's rant and dogma.

 

[SamJB]

Some very interesting and articulate responses. Great to hear from people working in the pharma, like jddukes.

For me, the bottom line is that regardless of how appropriate or well-selected the data is, the risks presented in this study are smaller than the risks of having high sugar levels. If you control your diabetes better by going low-carb then you are better off overall.

I think we'd all like more research into the safety of low-carbing in the context of diabetes. In particular though, I'd like to read more about its efficacy in large-scale studies for diabetics. In my personal experience - an experience shared by many diabetics - its efficacy is in no doubt whatsoever.

 

[Osidge]

One of the main findings of the work that Sid brought to the attention of the Forum is that more research is needed. I would hope that we could all agree that such research would increase our knowledge so that we can make the best decisions about managing our diabetes.

Regards

Doug

 

[Yorksman]

One of the main findings of the work that Sid brought to the attention of the Forum is that more research is needed. I would hope that we could all agree that such research would increase our knowledge so that we can make the best decisions about managing our diabetes.

It'll probably be the case when the costs of the genetics is further reduced. Great progress is being made. Just over a decade ago the cost of squencing an entire human genome was just over $1,000,000. today it is around the $10,000 mark. It is expected to fall as low as $100 in the next ten years.

The Wellcome Trust Case Control Consortium is comprises 50 research groups accross the UK and aims were to exploit progress in understanding of patterns of human genome sequence variation along with advances in high-throughput genotyping technologies, and to explore the utility, design and analyses of genome-wide association (GWA) studies. It has substantially increased the number of genes known to play a role in the development of some of our most common diseases and has to date identified approximately 90 new variants across all of the diseases analysed. As well as confirming many of the known associations, some 28 in total, the WTCCC has also identified many novel variants that affect susceptibility to disease.

Studies concerned with diabetes are headed by:

Type 1 Diabetes
David Clayton, JDRF/WT Diabetes and Inflammation Laboratory, Cambridge Institute for Medical Research
John Todd, University of Cambridge

Type 2 Diabetes
Andrew Hattersley, Diabetes and Vascular Medicine, Peninsula Medical School
Mark McCarthy, Oxford Centre for Diabetes, Endocrinology and Metabolism (OCDEM)

Some publications I have found in the list include:

Interrogating type 2 diabetes genome-wide association data using a biological pathway-based approach. Perry et al, 2009

Meta-analysis of genome-wide association data and large-scale replication identifies additional susceptibility loci for type 2 diabetes. Zeggini et al 2008

Localization of type 1 diabetes susceptibility to the MHC class I genes HLA-B and HLA-A. Nejentsev et al, 2007

Robust associations of four new chromosome regions from genome-wide analyses of type 1 diabetes. Todd et al, 2007

Replication of genome-wide association signals in UK samples reveals risk loci for type 2 diabetes. Zeggini et al, 2007

"What has been achieved in this research is the analysis of half a million genetic variants in each of seventeen thousand individuals, with the discovery of more than ten genes that predispose to common diseases."

"Amongst the most significant new findings are four chromosome regions containing genes that can predispose to type 1 diabetes and three new genes for Crohn's disease (a type of inflammatory bowel disease). For the first time, the researchers have found a gene linking these two autoimmune diseases, known as PTPN2."

"The pathways that lead to Crohn's disease are increasingly well understood and we hope that progress in treating Crohn's disease may give us clues on how to treat type 1 diabetes in the future"


Research from the Consortium had already played a major part in identifying the clearest genetic link yet to obesity and three new genes linked to type 2 diabetes, published in April in advance of the main study. It has found independently a major gene region on chromosome 9 identified by independent studies on coronary heart disease.

I think this last bit refers to the KLF14 gene discussed here by McCarthy:

Genetic ‘master switch’ identified in obesity and diabetes
http://www.ox.ac.uk/media/news_stories/2011/111605.html
The paper: http://www.ncbi.nlm.nih.gov/pmc/article ... -35276.pdf

 

[IanD]

D

One of the main findings of the work that Sid brought to the attention of the Forum is that more research is needed. I would hope that we could all agree that such research would increase our knowledge so that we can make the best decisions about managing our diabetes.
Regards
Doug

No. The main finding that Sid brought to the attention of the Forum was that:

long term low carb diets are far from safe and are actually shown to increase all causes of mortality.

oug, Osidge,
I wish we had introduced ourselves at the research group meeting we attended in March. The only report I have seen since was my own, & your comments on my email. I agree that diet research, particularly into low carb, will be of wide benefit to all diabetics, particularly T2.

Looking at the first page of the thread, a number of newly diagnosed low carbers were concerned at the long term safety of that diet. They were obviously worried by a report presented as fact, with a serious over-emphasis of the statistics for all-cause mortality, not specifically related to diabetes nor diet.

Your point is valid, but we did not need THAT report to make it. DUK have been trying for over 10 years to establish the long term safety of low carb diets, but for all their concern expressed have not substantiated any specific problems. Now one of their researchers (cited in DUK's "Position Statement" & a contributor to the "Evidence-based nutrition guidelines
for the prevention and management of diabetes" (Joy Worth) expresses concern that the necessary research will not be forthcoming because of the DUK attitude.

However, to date there has been no randomized controlled trial in type 2 DM patients and health care professionals remain wary of their use, particularly as standard dietary advice from Diabetes UK does not support this approach.7

Ref 7 is a 2003 report by the "Nutrition Subcommittee of the Diabetes Care Advisory Committee of Diabetes UK"

Abstract:
These consensus-based recommendations emphasize the practical implementation of nutritional advice for people with diabetes, and describe the provision of services required to provide the information. Important changes from previous recommendations include greater flexibility in the proportions of energy derived from carbohydrate and monounsaturated fat, further liberalization in the consumption of sucrose, more active promotion of foods with a low glycaemic index, and greater emphasis on the provision of nutritional advice in the context of wider lifestyle changes, particularly physical activity. Monounsaturated fats are now promoted as the main source of dietary fat because of their lower susceptibility to lipid peroxidation and consequent lower atherogenic potential. Consumption of sucrose for patients who are not overweight can be increased up to 10% of daily energy derived from carbohydrate provided that this is eaten in the context of a healthy diet and distributed throughout the day. Evidence is presented for the effectiveness of advice provided by trained dieticians. The increasing evidence for the importance of good metabolic control and the growing requirement for measures to prevent Type 2 diabetes in an increasingly obese population will require major expansion of dietetic services if the standards in National Service Frameworks are to be successfully implemented.

A 2003 report has established an attitude by DUK that inhibits the necessary research we agree is necessary.

 

[IanD]

Yorksman - your genetic postings are off-topic. I have suggested to the mods that you start a new thread.

 

[jddukes]

I am not under any obligation to read anything you write so please do not project blame onto me if I don't read beyond the first few sentances.

Of course you are when you are taking something I say out of context and missing the rest of it. Your point is invalid - you made a sweeping statement that I was saying one thing, which I was not soleyl saying. If you make such a statement, you have a duty or at least a courtesy to at least get your facts correct.

I am not here to argue with anyone and I am now done with this discussion as it is going nowhere - it seems almost everyone is missing everyone's points. I will exit this conversation though by stating my stance - general public have the right to analyse what a publication is saying relevant to them. I am not nor have I ever meant to imply that they can peer-review in the exact context the researchers and those in their field meant to be. But they still can look at the data, and make meaningful and useful conclusions about its relevance to a particular situation.

Fact - this study was posted on a diabetes forum raised as a discussion as there are people that follow "low carb" diets on here

Fact - this study was not intended to determine low-carb effects in diabetics who are already pre-disposed towards greater mortality (which is even more slanted in our relevant scenario as this was corrected for in some of the studies)

Fact - if people read low-carb diets increase chance of death without a healthy debate why this factually is innaccurate/not been shown by this study and they alter their lifestyle as a result of this single study, we are doing them a disservice

Fact - this study does not show that low-carb diets are dangerous or increase mortality to diabetics - it does not apply to them

These are all facts and the point of analysing this study in detail with a diabetic view in mind. I, and most others, are not saying the study is invalid, is useless, is wrong. I (and others) are mostly saying it does not apply to diabetics and therefore we should dismiss its sweeping title as relevant to us. In your posts however, you heavily imply that anyone who makes conclusions about this study is just public opinion which means nothing as it is not a peer-reviewer. Yet we are not judging the scientific validity of the study per se, we are analysing its relevance to diabetics, as this is a diabetic forum and that is the primary concern here.

 

[Daibell]

Hi jddukes. I agree ref relevance to diabetics; well said.

 

[Yorksman]

Yorksman - your genetic postings are off-topic. I have suggested to the mods that you start a new thread.

Shameless. You even changed the title of this thread which was started by Sid Bonkers

from Low-Carbohydrate Diets and All-Cause Mortality

to How & why scientific reports are published
viewtopic.php?f=1&t=40478&start=45#p379138

and you accuse me of going off topic.

Why do you make posts under the title How & why scientific reports are published when the poster's chosen subject is Low-Carbohydrate Diets and All-Cause Mortality? - just so you could stand your soapbox on top of Sid's post.

It is obvious that this thread was started to discuss the perceived wisdom of your preferred type of diet and you have persistently challenged anyone who seeks to discuss it.

The forum moderator raised the hope:

"I would hope that we could all agree that such research would increase our knowledge so that we can make the best decisions about managing our diabetes."

and I have posted details of genetic aspects of this research and links for anyone interested.

You on the otherhand seemingly wish to stop any discourse of which you personally do not approve and which may diverge from your view of things. What next, the burning of heretics?

 

[Yorksman]

I am not here to argue with anyone

That's all you do. You don't seek information nor do you provide any. All you do is make posts written in the tones of a Dickensian schoolmaster.

I will exit this conversation though by stating my stance - general public have the right to analyse what a publication is saying relevant to them.

I advised as much back on page 3 of this thread:

"I don't see how anyone can sensibly comment on this paper without referencing the cited papers in 'Table 2. Methodological assessments of the included studies', and 'Fig 2, Adjusted risk ratios for all-cause mortality associated with low-carbohydrate diets' for the details of what is meant by all cause mortality, low carbohydrate score or low carbohydrate high protein score."

viewtopic.php?f=1&t=40478&start=30#p378465

I believe that cursory glances at studies, mining them for the bits which apparantly support their preferred view whilst neglecting the caveats or points which do not support the reader's prejudice result are worthless.

I haven't read them and have stated that I hold no firm view on the matter.

But that doesn't stop you or another poster from assuming what my view is and arguing with it.

 

[hanadr]

the definitions of things like "Low Carbohydrate" are not clear through the whole of this argument.
I've just picked up this point from the definition of "starchy carbohydrates"

"uncooked starchy foods such as whole grains"

To my knowledge, uncooked grains are indigestible and very likly toxic too, in much the same way as uncooked kidney beans.
Hana