Jump to content

Wikipedia:Peer review/Low-carbohydrate diet/archive1

From Wikipedia, the free encyclopedia

This peer review discussion has been closed.
I've listed this article for peer review because it has gone through some significant rewrites and has not had a formal review in some time (although some editors have offered miscellaneous input; much thanks).

Specific issues I'd like reviewed:

  • General article quality.
  • Physiological descriptions: No medical/nutritional experts have reviewed this in some time. I would request, though, that any feedback or changes please distinguish between widely held opinions in the medical community and proven facts. As an example some experts argue that ketosis and ketoacidosis differ only in degree whereas others argue that they are related but different processes (or rather one involves the other but involves other factors as well). Treating one position or the other as established fact, even if the majority agrees with it, would be misleading.
  • Bias check: The topic is severely controversial. I have tried to balance making clear the stance of the proponents of the concept while making clear the stance of the majority of the medical/nutritional community. The most difficult aspect has been trying to objectively portray the research and the science because there is currently a great lack of consensus on these topics in the medical community.
  • Organization of the research information: In an effort to check bias the section on research links to a separate article which presents a cross-section of the research. Although that is a little unusual it seems the best approach since there is such extreme controversy in even interpreting the results of the studies.

Any feedback is appreciated.

Thanks, Mcorazao (talk) 19:08, 5 August 2008 (UTC)[reply]


Review by Una Smith

[edit]

The first thing that strikes me about this article is the narrow scope. It concerns only the low-carb diet popular now in the United States. The article would be much improved by a review of other cultures in which a low-carb diet is the norm. These are principally nomadic cultures that practice little or no agriculture. If the scope is correct, then the correct title of the article would appear to be something like Low carb craze in the United States.

The writing is awkward in places; look out especially for clauses misplaced within sentences. The "no consensus" sentences at the ends of several sections are weasel words; I would delete them.

The section on practices and theory should compare and contrast low carbohydrate with low calorie. At this time, the section in some places conflates these two issues. A diet may be low-carb yet provide excess calories. The question then is, does what happens to these excess calories depend at all on the fraction of calories consumed that are carbohydrates? The article also should explain, or at least refer the reader to an article that explains, nutritive vs non-nutritive carbohydrates. The article does finally explain it, in passing, near the end, under Vegetable and Fruits.

I have often read in popular articles that ketosis involves loss of muscle mass, but this makes no sense, taken in isolation. Ketosis mobilizes fat reserves, not proteins. The key issue is the relationship between ketosis and gluconeogenesis. This issue needs better explanation in this article, or a reference to an article that explains it better.

The article also conflates the proportion of fat in the diet with the proportion of fat as a source of calories expended. These proportions are not necessarily the same, especially when caloric intake does not match caloric expenditure.

I think many of the organizations that discourage low-carb diets do so out of a concern that the average consumer does not understand the difference between low-carb and low-calorie, and that it is all too easy for a low-carb diet to amount to a high-calorie diet. This article really needs a frank explanation of carbs vs calories.

Several of the problems above have to do with the structure of the article more than lack of any content. Key principles are explained not in the Principles and Theories section, but in the final Criticism and controversies section. That is too late.

--Una Smith (talk) 22:29, 5 August 2008 (UTC)[reply]

Thank you so much for the feedback!
May I ask for some clarifications?
  • Low carb craze in the United States - Two questions.
  • First, the article was intended to focus on low-carb diets as a deliberate means to promote health (i.e. changing one's diet to be lower carb), not so much as a broader discussion of traditional diets in various cultures. Granted the title could be interpreted to include the broader scope but, in keeping with Wikipedia policy, the title refers to the most common way this term is used which is to refer to the "dieting" scheme, not the broader definition. Do you disagree with that approach?
  • It is true that this article is biased toward the United States although at least some discussion of other countries is included. The U.S. bias is mostly a function of the fact that the majority of the readily available information is biased toward the U.S. Do you have any particular suggestions in terms of information or references regarding other countries?
  • No consensus - I agree that these statements are a little awkward and was trying to figure out how better to put this. I hesitate to simply remove these sentences since, for the sake of NPOV, it seems to me extremely important to make this point clear in some fashion in all the places that one could interpret this wrongly. Any suggestions on how to clarify this in a less awkward fashion?
  • compare and ... with low calorie - This aspect could be discussed more explicitly as could many others. I have tried to keep the length controlled rather than let it get out of control discussing every issue. Are you thinking of a detailed discussion of this particular issue or just a brief mention?
  • relationship between ketosis and gluconeogenesis - There is a lot of the science that can be discussed but, again, the issue is the length of the article. Regarding references, do you have any recommendations of good references to use? Most of the references that I find that explain these sorts of details in a way that would be appropriate are from non-medical, pro-low-carb propaganda sources and, therefore, are questionable to call authoritative. Most of the more authoritative resources do not have succinct discussions of these issues so all I can reference is a whole text book or a series of articles (I have been dinged before when I tried use an authoritative article which barely touched on a point I was discussing; doing so too much can be rightly considered "Original Research").
  • many of the organizations ... do so out of a concern ... the difference between low-carb and low-calorie - I do not think this is really true. Most of the literature from the major organizations refers to saturated fats and coronary problems, ketoacidosis, bone loss, etc. In other words, most of what I read says that the number of calories is not the major area of concern. The concern is that higher levels of fat and protein are suspected of being dangerous. Is there something more specific that you are referring to?
  • Key principles are explained ... too late - Too some extent I agree but the issue driving this is maintaining NPOV. I was trying to keep the "Principles and Theories" section very neutral and save some of the less neutral discussion for the later "Criticisms" section. It is really difficult to discuss most of these issues in a way that all of the experts would consider neutral. If I bring more topics into Principles and Theories the section quickly bloats because I have to tread so carefully in terms of discussing the different viewpoints and lack of agreement. Did you have some specific thoughts on how to rewrite the Principles and Theories? Maybe an outline of what you have in mind?
So I guess the recurring theme here is that the low-carbohydrate trend and associated research has opened up a big can of worms in the medical community highlighting a lack of firm understanding of certain aspects of human physiology (although, not everybody would necessarily even agree with that statement). So the topic in general is very hard to nail down without falling into the POV trap. Any further thoughts on that are welcomed.
Thanks.
--Mcorazao (talk) 15:49, 6 August 2008 (UTC)[reply]

Review by Colin

[edit]

I'll try to have a proper look later. But two quick points:

  • I agree with you about the name. I'm not aware of any major culture that adopts a low carb diet, but if there was then it would be significant for this article as one would expect epidemiological studies to have examined the effect of such a diet in a large population.
  • The statement "low-carbohydrate diets are often discussed as treatments for some other conditions, most notably diabetes[5][6][7] and epilepsy,[8][9][10] although these treatments still remain controversial and lack widespread support" is misleading. Firstly, to say something is "controversial" means you need to find opposing views. A new experimental therapy doesn't become controversial because it isn't proven yet, it becomes controversial if, for example, it is widely used despite lack of evidence or even in spite of evidence showing it is ineffectual or harmful. I don't know much about the diabetes aspect (your sources could do with some work here: see WP:MEDRS). Wrt the epilepsy aspect, it depends what you are talking about:
  1. Ketogenic diet for children. No controversy and it isn't "discussed as [a] treatment" but is in clinical use worldwide. Approved by medical guidelines and evidence-based-medicine (see recent randomised controlled trial published in the Lancet)
  2. Ketogenic diet for adolescents and adults. UK medical guidelines currently discourage this due to lack of evidence. However, this is an area undergoing research and that position may change. I'm not aware of it being controversial as any usage here that I've read about is within the context of clinical trials, rather than mavericks.
  3. Atkins diet for children or adults (a variant of the induction phase). This is a less ketogenic and less restrictive diet than the classic ketogenic diet. It is still being researched and there are no clinical guidelines approving its usage that I'm aware of. Again, I don't think it is controversial.

Check out the sources I used on ketogenic diet. A couple of the review papers I cited could be cited here too. Colin°Talk 08:35, 8 August 2008 (UTC)[reply]

Thank you!!!
  • There actually have been some studies on the Inuit but it seems nobody agrees on the interpretation of the results. The more detailed and controlled studies have involved "industrialized" populations so this is what is focused on in the research discussion.
  • There are opposing views on these treatments although they were not cited here (frankly I figured nobody would disagree that they were controversial so I didn't bother to include references). There is the general opposition that says low-carb is inherently dangerous and is to be avoided under all circumstances and then there is specific opposition to the use for these specific conditions (until this year even the ADA opposed use of low-carb at all; they now approve it for short-term weight loss but they still do not support it as an actual treatment for diabetes). I'll find some good citations to use for the opposition.
  1. The keto diet for children is certainly much less controversial and is actually a recognized treatment method. But there are still some who say that, despite its effectiveness, it is still dangerous. Still if we were to restrict the discussion to this specific subtopic it might be reasonable to omit referring to this as controversial.
  2. The keto diet for adults is less controversial than other treatments but I think you will find a significant contingent that will strongly oppose it on the basis that, in their opinion, there is no evidence of efficacy and low-carbohydrate diets are known to be dangerous.
  3. There are certainly variants that are more or less ketogenic than others. But the ketogenic aspects are not the only controversy. Nutritional deficiency, saturated fats, and other issues are frequently cited as dangers of the diet. I think if you look at what the AHA has to say they will tell you outright that they oppose low-carb diets for anything and everything. Granted you can argue that to some extent what they paint as a "low carb diet" is somewhat of a caricature but the point is that this is just one well-respected medical organization that would call anybody supporting any variant of "low carb" crazy. The point is that, for the sake of keeping things neutral, you have to acknowledge that mainstream still doesn't accept these things even if some of the leading researchers in those specific areas often do (c.f. if Stephen Hawking says that blackholes are twice as dense as anybody else, you can argue that Hawking would know better than anybody else but it is still unfair to imply that Hawking speaks for the entire scientific community).
The point is I don't mind looking at other phrasing but trying to imply that these things are not controversial seems to me misleading. Mind you, the time may be coming very soon when these things will not be so controversial but I do not think that you can say we are there yet.
Do you have specific suggestions for rephrasing?
Thanks again.
--Mcorazao (talk) 18:47, 8 August 2008 (UTC)[reply]
I'll try to follow up your leads. Perhaps I can clarify that I accept any use of a very low-carb diet as a medical therapy, undertaken without strict medical supervision by a specialist (neurologist, etc) and dietitian, is widely regarded as as dangerous. Perhaps the controversy here is the self-medication concern that folk with diabetes might go off and eliminate most major foodstuffs from their diet and end up in hospital. There also seems to be a concern about low-carb high-protein diets -- the ketogenic diet isn't a high-protein diet. I would very much doubt that the AHA have an opinion on its use in serious refractory epilepsy, and are more concerned with long-term use as a lifestyle choice or weight-loss technique. It sounds like we need to be more specific in pinpointing where the controversy lies, and find/cite some sources that actually criticise their medical usage. Have you got any evidence that the AHA, for example, have criticised these specific medical uses? Colin°Talk 23:36, 8 August 2008 (UTC)[reply]
Yes, specificity is an issue. Unfortunately specificity is also the problem. If you look at the AHA response to the 2008 study showing the benefits of the Atkins and Mediterranean diets over the AHA diet you see that the AHA actually sidesteps the what the study actually says and shows to bolster its position. In essence they are ignoring a lot of one the study says and implying things it does not say. One problem in this whole arena, of course, is the lack of a definition of low-carb diets (some are high-protein, some are high-fat, etc., etc.). But even when restricting the discussion to, say, Atkins groups like the AHA frequently focus on and exaggerate one aspect completely neglecting all the other aspects. So the question is when trying to be more specific whose specific opinion do you use? Or more to the point, when expert A says he is talking about X but you think he/she is talking about something completely different, is it ok to neglect their opinion or does that constitute original research? I would argue it is original research even if I think A is a jackass.
As far as whether the AHA has criticized the diets for medical treatments ... well, I think that is pigeon-holing a bit. For example, if a doctor says arsenic is dangerous but doesn't say arsenic is dangerous for runners then can I conclude he/she approves arsenic as being good for runners? I will do some digging to see what the AHA and others have to say about some of these things. Maybe there is something that can be said objectively that removes some of the "controversy" statements.
--Mcorazao (talk) 21:01, 12 August 2008 (UTC)[reply]
I've only skimmed the AHA report but I think they are concerned the study has merely shown efficacy wrt a surrogate endpoint -- that is, although it helped with weight loss, and weight loss is usually associated with greater health, it doesn't necessarily follow that those on that diet lived longer or were healthier. This is an important point and one where secondary sources help when trying to make sense of primary research. I agree with you that cherry picking the arguments or synthesising a case from multiple sources would be OR.
I think your arsenic example is a little extreme as nobody is suggesting that is beneficial for anything. Our lifestyle choices and medical choices involve balancing the risk of good and bad aspects. If you spend every evening down the gym, you may end up healthier overall but suffer more minor injuries and see less of your family, which might lead to divorce and depression... Serious chronic medical conditions like epilepsy and diabetes should idealy be treated by therapies that don't in themselves shorten life, but if left untreated (or ineffectually treated) then those conditions may shorten life. All anticonvulsants have risks and side-effects. For example, sodium valproate often causes weight gain and should be avoided in pregnancy, but is a very useful and effective drug. The balance of risk and inconvenience means the ketogenic diet is not a first choice therapy since if epilepsy is controlled by the first drug or two, that is more likely to be better for the patient.
You may be interested in this paper. Although great efforts are made to ensure children on the ketogenic diet are given adequate protein and calories, there is a problem with growth. Something to think about if children are using very-low-carb diets for weight control (I'm not advocating you use that paper as a source for anything in this article, as it studied a very specific group of patients who might statistically have problems with growth even without the diet). BTW: the diet is generally only used for a couple of years; it isn't advocated as a life-long therapy, though there may be a case for that in some patients. What about general low-carb diets -- are they generally suggested as life-long diet plans, or merely for short-term weight control?
Colin°Talk 08:09, 13 August 2008 (UTC)[reply]
surrogate endpoint - Well, the AHA certainly makes the surrogate-endpoint argument but in every statement they make they say that Atkins and the like are dangerous and to be avoided. I have never seen anything from them that even hinted that it might be appropriate for some things. I have not seen anything specifically about the Ketogenic diet for epileptics from them, though.
arsenic example is a little extreme - Obviously it is extreme. I was trying to illustrate a point. But wrt the AHA, as I say, I have never seen anything from them that suggested that the low-carb is anything but dangerous. You are presumably right that if somebody is going to die and low-carb is the only thing left to try they presumably would not object. But the point I was making on epilepsy (and other things) is not simply that they are used in extreme cases but some experts are saying these may actually be better therapies even in the non-extreme cases.
are they generally suggested as life-long diet plans - Well that depends on who you ask. Obviously there are lots of people who say that a proper low-carb diet (i.e. not being stupid and just eating anything that is low in carbs) is the healthiest way to eat. But opinions on that run the full spectrum so, to repeat myself, there is nothing resembling a consensus, even among the advocates of low-carb.
--Mcorazao (talk) 17:04, 13 August 2008 (UTC)[reply]


I think you are making it harder than necessary. The AHA is speaking to a particular (lay) audience about a particular use of the diet (weight loss, improved health). The folk using the ketogenic diet for epilepsy are fully aware of the health issues of a high fat diet and factor those risks into their clinical decisions. However, lets move on...

You may be interested in this review paper which is quite readable yet more authoritative that some of the website sources the article uses.This paper gives a definition of a low-carb diet. You should use such an independent definition rather than making your own: "For the purposes of this discussion, we focus on " which also isn't encyclopaedic language. Speaking of which, the text does sometimes lapse into campaigning journalism rather than a dispassionate encyclopaedia article.

The sources generally need a fair bit of work. Try to find more reviews in peer-reviewed journals (like the one above), rather than citing websites and news sites. For example, the "Glucose availability" cites HowStuffWorks, which is a lay tertiary source. It may pass WP:V but isn't a first-class choice. Plus it only really covers a small aspect of the paragraph, the rest of which is unsourced. You can't say "A common argument in favour of" without a source saying it is a common argument. And then to follow it with "This argument, by itself, omits certain salient details" ends up making this an article about the disputes rather than about the subject itself. Find out what the best sources say on the matter and report them; don't report the false arguments unless they are truly notable. BTW: I think this section is conflating low-carb with high(er)-protein (such as the Atkins). The low-carb/adequate-protein/high-fat ketogenic diet is in fact much more readily metabolised (due to the high fat content) and this means that calorie levels can be set lower than those for a typical diet and still maintain the same weight. I think the issues with taxing the liver are high-protein issues rather than specifically low-carb issues. Colin°Talk 12:56, 14 August 2008 (UTC)[reply]

  • "AHA is speaking ..." - I agree that we're beating that issue too much. I am simply saying that there is a larger issue than whether the experts would agree on these treatments for severely ill patients. At the other end of the spectrum some experts are saying that because low-carb is so effective for treating many epileptics, diabetics, etc. that high-carb may actually be causing their condition. Obviously most of the establishment would not accept that conclusion.
  • "This paper gives a definition" - I think you have a point about basing the article's focus on a recognized authority's definition. But we have to be careful because
  1. Their definition is not universally recognized.
  2. Certainly not all studies on "low carb diets" are necessarily consistent with that definition.
In any event I'll change the article to use this definition. Thanks.
  • "The sources generally need a fair bit of work" - Yeah, I'd like to do better sourcing but I do not have access to a medical library and rely largely on layman's books and online searches. The problem with a lot of these issues is that most authoritative sources that I find do not concisely address the issues at hand. That is, ideally you want to find a paragraph in some medical reference that states exactly what you're trying to say but what I tend to find instead is whole chapters or longer that you would have to read and digest to arrive and what I am writing in one sentence. And, of course, as soon as I do that questions come up as to whether I am paraphrasing or creating original research. So it is often easier to reference less authoritative sources (and sometimes I reference a concise statement in an authoritative source that only addresses a part of what I'm saying; question is, is it better that the source be authoritative or complete?). Any help you can offer in finding better sources is much appreciated.
Thanks.
--Mcorazao (talk) 04:10, 15 August 2008 (UTC)[reply]
There is a sentence in the article similar to your "some experts are saying that because low-carb is so effective for treating many epileptics, diabetics, etc. that high-carb may actually be causing their condition" which made me frown. The actual sentence doesn't specify which ill health, but it comes after the cure for epilepsy/etc sentence. There really is absolutely no link with diet causing epilepsy, which often has a genetic or physical cause. If you know of any source that suggests such a link, I'd be interested. I know less about diabetes but remember there is more than one kind and poor diet is already well known to be a cause of one form.
I'm afraid I can't supply you with non-free sources but there are quite a lot at PubMed that you can read in full for free. I advise against citing a paper after only reading its abstract. If you are aiming for GA or FA, then I think you will need access to this best research. Therefore, I suggest you seek out a writing parter who has access. You could try posting a request at WP:MED or some other wikiproject, or look at what editors write related articles. Condensing a whole book or chapter is a problem. Scientific/medical review papers are pretty condensed to start with and I've found you can sometimes cite many sentences or paragraphs to one review. If you need any help with PubMed, post me a note on my talk page or ask the medical wikiproject for help. There are folk there who are only to glad to help good-faith editors improve medical articles. Colin°Talk 20:18, 15 August 2008 (UTC)[reply]
  • "high-carb may actually be causing their condition" - As far as prevention I was more referring to diabetes than epilepsy. There are some statements I have read which seem to indicate that the author believes that diet may be the cause of some types of epilepsy but I have never seen an expert actually come out and say it so I have not said this either.
  • "poor diet is already well known to be a cause" - I know. But there is debate about what part of a "poor diet" is the primary trigger in causing diabetes or whether more than one thing may be the primary cause. As with everything related to low carb, there is no consensus.
  • "I suggest you seek out a writing parter who has access - I was actually hoping this review process might ferret out such a person.
Again the problem with discussing the theories behind low carb diets is that the medical texts (at least all that I have seen) do not concisely discuss the combinations of issues that are relevant. There are "experts" (i.e. physicians) who do discuss the issues concisely but, because they back low-carb, they are considered fringe elements and, therefore, citing their summaries of the relevant physiology comes off as suspect. Trying to find "neutral" sources that discuss the same topics so concisely is difficult (mostly because the metabolic states associated with low carb are considered by most doctors to be disease processes and, so are discussed only in the context of disease).
Anyway, I'll try wikiproject as you suggest. I can do a little more cleanup according to your suggestions but I don't think I can make drastic improvements in referencing without help.
Thanks.
--Mcorazao (talk) 22:09, 15 August 2008 (UTC)[reply]

Review by Eubulides

[edit]

I have lacked time to do a proper review but here are a few things:

  • The article does not state up front what is the difference between this article and the Ketogenic diet article. Why are there two articles and not one? It's not at all clear from the lead, which is hard to follow in this area.
  • The article says "Other studies have found possible benefits to individuals with ... autism." and cites Evangeliou et al. 2003 (PMID 12693778). I've helped editing Autism therapies so this one caught my attention. The only review citing Evangeliou et al. that Google Scholar reported, namely Erickson et al. 2005 (PMID 16267642), mentions the study and says only "This study lacked a control group and, therefore, it is difficult to interpret the results." Which is being charitable. This is just a spot check of course, but if it's an example of the sort of evidence being used in this article, then there is a real problem with the evidence basis.
  • There are few citations to recent reviews in the medical literature. I did a quick Pubmed search and found 9 recent reviews, of which only 2 are cited. Here is the full list, which I suggest be looked at:
  • Malik VS, Hu FB (2007). "Popular weight-loss diets: from evidence to practice". Nat Clin Pract Cardiovasc Med. 4 (1): 34–41. doi:10.1038/ncpcardio0726. PMID 17180148.
  • Westman EC, Feinman RD, Mavropoulos JC, et al. (2007). "Low-carbohydrate nutrition and metabolism". Am J Clin Nutr. 86 (2): 276–84. PMID 17684196.
  • Williams PG, Grafenauer SJ, O'Shea JE (2008). "Cereal grains, legumes, and weight management: a comprehensive review of the scientific evidence". Nutr Rev. 66 (4): 171–82. doi:10.1111/j.1753-4887.2008.00022.x. PMID 18366531.
  • Last AR, Wilson SA (2006). "Low-carbohydrate diets". Am Fam Physician. 73 (11): 1942–8. PMID 16770923. This one is being used.
  • Freeman JM, Kossoff EH, Hartman AL (2007). "The ketogenic diet: one decade later". Pediatrics. 119 (3): 535–43. doi:10.1542/peds.2006-2447. PMID 17332207. This one is being used.
  • Gibson LJ, Peto J, Warren JM, dos Santos Silva I (2006). "Lack of evidence on diets for obesity for children: a systematic review". Int J Epidemiol. 35 (6): 1544–52. doi:10.1093/ije/dyl208. PMID 16984930.
  • Nordmann AJ, Nordmann A, Briel M, et al. (February 2006). "Effects of low-carbohydrate vs low-fat diets on weight loss and cardiovascular risk factors: a meta-analysis of randomized controlled trials". Arch Intern Med. 166 (3): 285–93. doi:10.1001/archinte.166.3.285. PMID 16476868.
  • Kushner RF, Doerfler B (2008). "Low-carbohydrate, high-protein diets revisited". Curr Opin Gastroenterol. 24 (2): 198–203. doi:10.1097/MOG.0b013e3282f43a87. PMID 18301271.
  • Samaha FF, Foster GD, Makris AP (2007). "Low-carbohydrate diets, obesity, and metabolic risk factors for cardiovascular disease". Curr Atheroscler Rep. 9 (6): 441–7. PMID 18377783.

Get good sources, rewrite the article to match what the sources say (rather than what you hope they say), and good things will happen. Hope this helps. Eubulides (talk) 09:32, 30 August 2008 (UTC)[reply]

Wrt to the separation of this and ketogenic diet, the lead paragraph of both provides most of the important facts but the subsequent parts of LCD cloud the issue and should be revised. "The ketogenic diet is a high fat, adequate protein, low carbohydrate diet" whereas most LCDs are high fat and high protein and aimed largely at weight loss rather than specific disease control. Not all LCDs produce ketosis (which is required on the ketogenic diet, hence the name) and some only produce ketosis during certain stages (Atkins induction). The ketogenic diet is also strictly calorie controlled to the gram (second paragraph of that article) where as many LCDs have no such control. The KD is only used under medical supervision. Probably more differences. Anyway, this article should make its scope clear and regard the KD as a rather unusual subset of an LCD, to be mentioned largely in passing IMO. It should concentrate on the health/weight-loss aspects. Colin°Talk 18:11, 31 August 2008 (UTC)[reply]