This article has not yet been rated on Wikipedia's content assessment scale. It is of interest to the following WikiProjects:

Please add the quality rating to the {{WikiProject banner shell}} template instead of this project banner. See WP:PIQA for details. Medicine: Neurology FA‑class Mid‑importance This article is within the scope of WikiProject Medicine, which recommends that medicine-related articles follow the Manual of Style for medicine-related articles and that biomedical information in any article use high-quality medical sources. Please visit the project page for details or ask questions at Wikipedia talk:WikiProject Medicine.MedicineWikipedia:WikiProject MedicineTemplate:WikiProject Medicinemedicine FA This article has been rated as FA-class on Wikipedia's content assessment scale. Mid This article has been rated as Mid-importance on the project's importance scale. This article is supported by the Neurology task force (assessed as Mid-importance). Please add the quality rating to the {{WikiProject banner shell}} template instead of this project banner. See WP:PIQA for details. Food and drink FA‑class Low‑importance This article is within the scope of WikiProject Food and drink, a collaborative effort to improve the coverage of food and drink related articles on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.Food and drinkWikipedia:WikiProject Food and drinkTemplate:WikiProject Food and drinkFood and drink FA This article has been rated as FA-class on Wikipedia's content assessment scale. Low This article has been rated as Low-importance on the project's importance scale. Food and Drink task list: To edit this page, select here Here are some tasks you can do for WikiProject Food and drink: Help bring these Top Importance articles currently B Status or below up to GA status: Agaricus bisporus (i.e. mushroom), Beef, Bread, Curry, Drink, Food, Ham and eggs, Soy sauce, Sushi, Yogurt Bring these Top Importance articles currently at GA status up to FA status: Beer, French cuisine, Sugar Bring these High Importance articles currently at GA status up to FA status: Apple, Burger King, Fish Participate in project-related deletion discussions. Delete unrelated trivia sections found in articles. Please review WP:Trivia and WP:Handling trivia to learn how to do this. Add the {{WikiProject Food and drink}} project banner to food and drink related articles and content to help bring them to the attention of members. For a complete list of banners for WikiProject Food and drink and its child projects, select here. Provide photographs and images for Category:Wikipedia requested images of food Consider joining this project's Assessment task force. List any project ideas in this section Note: These lists are transcluded from the project's tasks pages.

	Ketogenic diet is a featured article; it (or a previous version of it) has been identified as one of the best articles produced by the Wikipedia community. Even so, if you can update or improve it, please do so.
	This article appeared on Wikipedia's Main Page as Today's featured article on January 17, 2010.
Article milestones Date Process Result July 9, 2008 Good article nominee Listed November 12, 2009 Peer review Reviewed December 16, 2009 Featured article candidate Promoted
Current status: Featured article

A recent edit replaced most instances of "calorie" with "food energy" or "energy" with the comment "this isn't unit-dependent, don't be using names of units as substitute for quantity being measured, and those critters are hard to hold down to get a "number" count". Of the changes, some are neutral, some are grammatically wrong and most are non-idiomatic. Nobody talks about "energy-free drinks". "low-calorie" is the idiomatic term for food; "low-energy" is the idiomatic term for lightbulbs. Wikipedia is written for the general reader, not the pedant.

I've looked at my journal sources, which one would expect to use acceptable terminology. The neurologist papers use calories everywhere. One set of papers by a US dietitian uses calorie everywhere. Another set of papers, by a UK dietitian uses "energy" fairly consistently in one paper (but never "food energy" and once "dietary energy") but is completely mixed in another paper.

The other change was the addition of kJ as a unit. I'm not aware that any country uses kJ as their unit of food energy. It appears as an additional unit on food labels in the EU, but nobody reads it. My dietitian papers all use calories ("kcal" when they are being precise). Only one paper out of over 200 offers kJ as an alternative to kcal. I wonder if there are any readers who benefit from this additional unit, which clutters the sentences. Thoughts? Colin°^Talk 17:30, 15 January 2010 (UTC)[reply]

I didn't find the changes useful; people understand calories. SandyGeorgia (Talk) 18:04, 15 January 2010 (UTC)[reply]

I just looked in my kitchen cupboard; all of my packets of rice, dried pasta, flour etc are labelled with both kJ and kcal. None are labelled "food energy". The articles that I have checked on PubMed uses "calories". Most people understand "calories" in this context."Food energy" is not idiomatic. I suggest the changes be reverted—the expert reviewer did not criticise the use of the term. Graham Colm ^Talk 18:16, 15 January 2010 (UTC)[reply]

I agree - "calorie" is an accessible term to the lay reader, while "food energy" and "energy" is simply confusing. Considering that the sources use the word "calorie" as well, we should stick with that. If the sources switch to a different terminology, we can revisit the issue then. Awadewit (talk) 19:38, 16 January 2010 (UTC)[reply]

There really are people who measure their caloric intake in the modern SI units, kilojoules, rather than archaic—and thousand-fold ambiguous, plus numerous variants in precise definitions at each size level, compounding the problem—non-SI metric units not coherent (in the metrology-jargon sense) with any system of units. That's part of the reason why this article already included some conversions to kilojoules, before I added more. See any European nutrition label, for example. Consider common usage in Australia. This diet isn't unit-dependent. And in any case, using the name of a unit of measure as a substitute for the quantity being measured is in general bad form. Doing so does not "exemplif[y] our very best work". Wikipedia:Featured article criteria.

Another aspect of the globalization is that there are a whole lot of people who aren't going to understand why the units are written out as "calories", yet they use the symbol "kcal" with "k" for the "kilo-" prefix. That might be common in various geographical locations, or among specialized jargon within certain journals, but it is not by any stretch of the imagination a universal practice. I never see "kcal" used in any general audience publications in the United States, for example.

Consider also usage elsewhere on Wikipedia such as in Template:Nutritional value which only accepts energy in kilojoules as a parameter, and which, unlike common practice in U.S. nutrition labels, does not use "calories" not as a unit of measurement but rather as a substitute for the quantity being measured. Its output field identifies the quantity as "Energy", given in both kilojoules and calories in the infobox. There may be other usages in other templates in Category:Food infobox templates, but from what I've seen lately on Wikipedia, I would be surprised any of them seeing significant use would use "calories" as a field name, when what is meant is "energy". Gene Nygaard (talk) 18:50, 15 January 2010 (UTC)[reply]

See, e.g., this paper dealing with guidelines for New Zealand found on a .au (Australia) website, "Food and Nutrition Guidelines for Healthy Adults: A Background Paper" http://www.broadmeadows.grassroots.org.au/library/community_resources/items/2004/01/71818-upload-00001.pdf

1. It uses kJ (and kJ/g, etc.) standing alone much of the time; in Table 3 there are conversions to kcal, but the text is always "kilojoules" or "kJ" (calorie is never spelled out, near as I can tell).
2. The other thing is constant is that the quantity being measured is always "energy", not "kilojoules" and not of course "kilocalories" (as I said, calories are never spelled out in this paper). Gene Nygaard (talk) 19:17, 15 January 2010 (UTC)[reply]

Example 2: then, of course, we have SportsDieticians Australia talking specifically about ketogenic diets, saying things like:

"Low carbohydrate diets achieve weight loss because they are low in kilojoules, not because carbohydrate is fattening" and "therefore reducing the number of kilojoules you eat" and "will help you keep a lid on your kilojoule intake". This does use units of measure to mean the quantity being measured, the same problem this Wikipedia article has, just with different units.

Nary a "calorie" to be seen on that page, however. Gene Nygaard (talk) 19:50, 15 January 2010 (UTC)[reply]

I have restored the kJ measurements. Australia/New Zealand do seem to be trying to use those as an official measurement. However, Googling their websites (health, Weight Watchers, etc) show that every day people use the term "calories" just as we do. In Europe, both measurements appear on food labels but nobody in the UK talks about kilojoules when discussing diet. Yes we know that pedantically it is "energy" but nobody talks like that. I've removed the Globalize tag unless someone can show that some English speaking counties no longer understand the idiom "calories" wrt food energy intake. Colin°^Talk 19:57, 15 January 2010 (UTC) I've made a few other tweaks which leave calorie/calories only for idiomatic (i.e., "energy") usage and use "unit of energy" or "kcal" in other cases.[reply]

Well! What can I say? You've managed to leave me dumbfounded. But I'll try to work my way through it.

Let's see if I've got this right. You are now convinced that you have everything in this article fixed so that "kcal" always means "calories", and so that "calories" never means calories but always means "energy"? Wow! That's profound.

And—the mind-boggling part—you think that now fixes everything? So much so, that this brilliant move on your part means that the issue is resolved, so you can unilaterally decide the discussion has to be over and on your own whim remove the link to this discussion from the article page? Before you even get a response from anybody about your strange solutions?

I guess the good news is that now, if we do them in the right order, it will only take two simple search-and-replace operations to get the article so that it actually says what we intend it to say.

Note that the globalization issues do not have anything to do with "English-speaking countries" per se. But I'm probably at least partly at fault there, by not more clearly distinguishing the globalization issues from the bad-writing and clarity-of-expression issues. Usage in the rest of the world, in whatever language, is also relevant. Nutrition labels throughout the EU use kilojoules, as to those in other places such as Norway and Japan; about the only place where they are rarely found is the United States.

You claim nobody in the British public ever uses the kilojoules which do appear on all their nutrition labels. Not real sure I believe that. But be that as it may, there is a more important factor about usage which you seem to be unaware of. Nobody in the U.S. general public ever uses "kcal" for calories, even though some researchers do use that symbol in some publications. And, unlike the situation in the UK, "kcal" never appears on any of our U.S. nutrition labels.

In other words, an unexplained jump from "calories" to "kcal" can indeed cause confusion. But you've come up with a really bizarre solution to that: let's just never use "calories" as a unit of measure, even though that's what a lot of people are familiar with, especially in the UK and the U.S.

I can come up with only one possibility that could possibly explain some of your recent actions. Are you laboring (sure, you'd probably be labouring, but unlike the Americans reading "kcal", you aren't going to have any difficulty understanding what I meant) under a misapprehension that the units really are and always have been "kilocalories", and that it is only through laziness in general usage that this has come to be abbreviated and shortened up to "calories"? If so, you are flat-out wrong. Furthermore, calling them "kilocalories" is actually what is new, a development that has taken place in my lifetime, and it remains a minority usage.

In fact, the large calories, the ones based on the kilogram rather than the gram, are the older definition of calories. The small calories, or gram-calories, came along later. And all of that, of course, happened before people figured out the advantages of using "coherent" (a metrology jargon term, meaning each derived unit in the system is a unitary combination of base units) systems of units, such as the SI and its joules for energy (1 J = 1 N·m = 1 kg·m²/s² in terms of the base units in this coherent--because of those "ones"--system). You are likely too young to remember when physicists and chemists actually used gram-calories. Those units are pretty much obsolete now—they survive only in the strange notion that there is still a need to distinguish the only calories still in use, the large calories used in a nutrition context, by using the inaccurate symbol "kcal" for them. We can only get away with that because there never was any significant use of the prefix "kilo-" for the large calories used for nutrition, so we can use the symbol for 1000 small calories to mean 1 large calorie. So when we are talking about nutrition, "calories" and "kilocalories" and "cal" and "kcal" and even the grotesque "Cal" and even more strange "kCal" all mean the same thing—none of them ever differs from any of the others by a factor of 1000, despite the appearance of the prefix kilo- or its symbol k in the usage of some writers and its absence in the usage of others (and my spell-checker flags kilocalories and kcal, too). Maybe not exactly the same thing; seems there are some questions as to the precise definition, whether the International Steam table calories (4.1868 kJ) or the thermochemical calories (4.184 kJ) or one of the other variants should be used in this context, but it really doesn't matter because none of our measurements are precise enough to distinguish between any of those definitions. Gene Nygaard (talk) 06:53, 16 January 2010 (UTC)[reply]

The units on UK nutrition labels are kcal and kJ. Look at File:Nutrition-label.jpg for what I typically see on a packet. I reviewed all my papers on the KD (well over 200) which are written by dietitians and neurologists from over the world but with a tendency (both author and publication) towards the US. When precise energy units are required (e.g. 1500 kcal) then "kcal" is the form used and "kJ" is often given as an alternative. As noted above, papers by neurologists and dietitians use the term "calories" for the energy component of food just as the general reader does. These are papers are published in some of our most prestigious journals, which have high editorial standards. If they are happy with this usage of the word calorie then so am I. Wrt British usage of "kilojoule", try Googling a newspaper such as The Guardian. The term just isn't used. Colin°^Talk 11:00, 16 January 2010 (UTC)[reply]

Here's some typical American usage, from Seattle:

The Regulation to Require Nutrition Labeling in King County Chain Restaurants

(Chapter 5.10 of the King County Board of Health Code)

Chain restaurants are required to provide total amounts of the following for each standard menu item (acceptable abbreviations):

▶ Calories (cal)	▶ Grams of saturated fat (sat fat)
▶ Grams of carbohydrate (carb)	▶ Milligrams of sodium (sodium)

<end of quote>Note that "cal" is used many times throughout this guidebook to help the restaurants comply with the law, never "kcal", which are not an acceptable abbreviation under this law.

But what about the real main problem here? Why do you insist on pretending that this diet is somehow unit-dependent? Local journals don't make the deliberate efforts to globalize their usage, as we do here. This diet does not only work if the units you use measure energy are calories, not if you use joules. The usage is particularly strange in this article as it stands now, given that you have so carefully eradicated any use of "calories" as a unit of measure. Gene Nygaard (talk) 14:59, 16 January 2010 (UTC)[reply]

Gene Nygaard, I'm unaware if you have ever had an article on the mainpage, or realize what a stressful time it can be. Lowering the sarcasm in your responses, and raising the AGF collaborative factor, would be appreciated and would help reduce the TLDR factor. SandyGeorgia (Talk) 15:25, 16 January 2010 (UTC)[reply]

I'm not trying to "pretend that this diet is unit-dependent". You know very well that the use of "calorie" has entered the language in a way that no longer ties it to one measurement system or another. Even our Calorie article documents this usage. Wikipedia is replete with articles that are full of calories. Calorie restriction, Very low calorie diet, Dieting, Obesity, Diet and obesity. Go change all of them to talk about "low energy diets" and see how far you get. Wrt "kcal", I'm following the practice of our best sources, written by dietitians, including those published in the US -- a country you claim doesn't understand the term.

I strongly suggest that we just agree to disagree on this point. I had no intention of wasting today reading about the history of units or the laws of some US county. I'll have more important things to worry about tomorrow. If you continue to feel strongly about Wikipedia's abuse of these terms then I suggest you debate this and establish consensus in a bigger venue elsewhere, not in an article with a handful of watchers on an obscure diet eight hours before it hits the main page. Colin°^Talk 16:00, 16 January 2010 (UTC)[reply]

Unlike the situation with the units of the International System of Units and of the few other units acceptable for use with it, where the CGPM has prescribed uniform, consistent symbols for use throughout the world, these calories are just like old software, legacy applications no longer supported and updated. There are no generally accepted standards for the use of them. THUS, if we are going to use a symbol (kcal) which is prima facie (on its face) incongruous with the generally used spelled-out name of that unit (calories), and one that is also different from the symbol many readers are familiar with, then some explanation is in order. And it would cost little to make this article more readable by putting in an explanation. Gene Nygaard (talk) 16:17, 16 January 2010 (UTC)[reply]

The note (which is now in the Notes section along with usage of the term "fasting", etc) is a reasonable compromise. Let's move on. Colin°^Talk 17:24, 16 January 2010 (UTC)[reply]

This issue is not about "globalisation" at all. It's about one editor who does not like the fact that one useful word—used a few times in the article—was once an esoteric scientific term but is now well-established in everyday language. Calories, in this context is used on both sides of the equator and both sides of the Atlantic, even in Russia. Gene Nygaard, I am struggling to assume good faith here. It appears to me that you are not even trying to achieve a consensus, but are trying to win. There is no support for the changes you suggest. And yes, Colin is right, why choose on this article all of a sudden? There are many others. That ridiculous tag should be removed forthwith. Graham Colm ^Talk 16:33, 16 January 2010 (UTC)[reply]

I agree, and have removed the tag. I tried to fix the note, but may have bungled it. SandyGeorgia (Talk) 16:37, 16 January 2010 (UTC)[reply]

Thanks for formatting the note for me. And suggestions on improving the wording would be welcome. Gene Nygaard (talk) 17:49, 16 January 2010 (UTC)[reply]

Well, GrahamColm, if you had taken the time to address the issues here, rather than attacking me, that would help too. Gene Nygaard (talk) 17:31, 16 January 2010 (UTC)[reply]

Gene, if you continue to tag this article, over a trivial issue, this close to Main page day, I shall be requesting Administrator intervention on your editing abilities. Is that clear? Colin°^Talk 17:24, 16 January 2010 (UTC)[reply]

What do you mean, trying to accuse me of some wrongdoing. You were the one who removed a properly added tag, in an incomplete discussion, a removal hidden away in an edit that didn't even mention that you were doing so. I put it back, saying I was doing so.

Seems to me that we have some ownership problems here, too. Gene Nygaard (talk) 17:35, 16 January 2010 (UTC)[reply]

I'm guessing that you are too close to the issue here, Colin, and can't see the forest for the trees. You've been at this long enough to understand full well that there are a whole lot of sources that use the word "calories" when spelling out the units, and use the symbol "kcal" when abbreviating it. Including, of course, "our best sources, written by dietitians, including those published in the US". Nobody needs to explain that to you. That doesn't mean that the average Wikipedia reader is going to understand this incongruous usage without explanation. Gene Nygaard (talk) 17:44, 16 January 2010 (UTC)[reply]

I removed it in the edit that fixed the globalisation issue (lack of kJ as an alternative to kcal). In good faith, I assumed this was the only globalisation issue. It seems there is no pleasing you. The globalisation tag is a bullying tactic to win your argument. You appear to have no concept of good faith discussion and the idea that people may genuinely disagree with your position. Rather than insulting me with claims of ownership why don't you please go gain Wiki consensus to ban the use of "calories" for food energy and ban the use of "kcal" for units of energy. Start with a bigger article or a Wikiproject talk page. If you have Wikipedia's interest at heart (rather than a selfish desire to win this argument) then argue your case somewhere there is less heat and more eyes. Colin°^Talk 17:55, 16 January 2010 (UTC)[reply]

Choosing to use one of the units of measure as a synonym for the quantity being measured, rather than choosing to use the other unit of measure for the same purpose, is part and parcel of the very same globalization problem.

I'll bet that if this article were instead written in the style of my Example 2 from Australia above, and it consistently used the word "kilojoules" when energy was meant as that Australian article did, you and some of the others backing your position here would be first in line whining about the need to globalise the usage here. Maybe we should just try that for a while, to see what happens. Wouldn't be hard to replace every appearance of "calories" with "kilojoules".

This is the article being held out as an example of the best we can do now. It is the part of Wikipedia most in need of fixing right now; bad usage elsewhere isn't an excuse. This article is quite reasonably, and should be, held to a higher standard.Gene Nygaard (talk) 02:13, 17 January 2010 (UTC)[reply]

Sandy asked me to comment on this. Looking over the discussion here, and particular this edit - I don't have a problem with giving the kcals and providing a parenthetical alternate measurement in kilojoules; however, relacing all instances of calorie with "food energy" ends up with clunky writing. Case and point from that edit I cited: The diet provides just enough protein for body growth and repair, and sufficient calories to maintain the correct weight for age and height. -> The diet provides just enough protein for body growth and repair, and sufficient food energys to maintain the correct weight for age and height. The former sounds natural, the latter does not. Raul654 (talk) 18:05, 16 January 2010 (UTC)[reply]

Additionally, this is a misuse of the {{globalize}} template. The issue at hand is an issue of WP:ENGVAR, not WP:NPOV. It's not like some countries have a different perspective on the how much energy is being represented here: they just have different units. Whether we say "calorie" or "Calorie" or "kcal" or "kJ" is exactly as unimportant as whether we use British or American spelling. I'd as soon fuss over whether Colin tells me his height in centimeters or inches.

The globalization tag is meant for situations in which different perspectives exist -- as in, "Those bally Orangemen might think that, but a true Green would never agree!" or "American and Japanese bankers aren't the only people in the world that have an opinion on Japan's financial situation." WhatamIdoing (talk) 20:09, 16 January 2010 (UTC)[reply]

Colin has admitted that a globalization issue existed. Maybe you need to go look at the article's history page, and get out your dictionary.

1. Find the edit summary "for the benefit of our Antipodean friends"
2. Read the actual discussion in this thread
3. Try to figure out what that edit summary meant.

Gene Nygaard (talk) 01:20, 17 January 2010 (UTC)[reply]

No, you've misunderstood -- and if Colin actually thinks that the choice of units is an issue of systemic bias (which I sincerely doubt: I think he was using your chosen handle as a shortcut), then he's simply wrong.

The first thing I'd like you to understand is that "this isn't how we use that particular template" does not mean "please remove kJ units from this article." Personally I support the inclusion of such a conversion in all relevant articles. {{Globalize}} exists to tag articles that have inappropriate geographic restrictions -- like only representing what's been published in one country about the diet (assuming that there were actually different views in different countries). An article can't be considered either complete or neutrally written if it excludes major viewpoints.

However, whether you use kJ or kcal or both is an issue of style, not content. Converting it to kJ doesn't indicate that Australians believe something different about the role of nutritional energy in this diet; it's the same thing, just "spelled" differently. {{Globalize}} is not meant to identify differences in style -- even if this or that style is less accessible to readers in this or that country.

Wikipedia has a settled system for this specific issue, BTW. You'll find the ENGVAR-style overall information here, and a requirement that kcal be used as the primary unit here. The use of kJ is neither required nor even mentioned.

And, I add, I didn't make these rules, and I can't change them. If you want to continue this discussion, then I suggest going to WT:MOSNUM and telling them that you think their long-standing preference for kcal is stupid. Talking to me, or to anyone here, isn't going to change a single character at MOSNUM. WhatamIdoing (talk) 02:59, 17 January 2010 (UTC)[reply]

The guidelines at WP:MOSNUM are not prescriptive, so they admit some exceptions, if there's a good reason. However, in this article, I can't see any problem: each mention of kcal (the unit) is followed by a conversion to kJ. Just about any reader will be familiar with one or the other of those in this context, particularly as you've taken the time to make a note distinguishing between the use of calorie and kcal in this article. It works for me and meets the guidance at WP:MOSNUM#Units of measurement:

• When used in a nutrition related article, use the kilocalorie as the primary unit.
• Where English-speaking countries use different units for the same measurement, follow the "primary" unit with a conversion in parentheses.

What more could anyone want? --RexxS (talk) 03:31, 17 January 2010 (UTC)[reply]

User:WhatamIdoing, I've dealt with a number of national-varieties-of-English issues. This is not one of them. There is little or nothing to do with national varieties of English in this article.

And sure, I probably could have added a {{copyedit}} tag and probably a whole bunch more of the various specialized maintence tags as well. But what would have been the point of doing so? If I had put a bunch of them there, would that have fostered a better discussion? I doubt it.

Back to ENGVAR. Let's just assume that the first major contributor to this article had indeed uniformly and consistently used "kilojoules" to mean food energy, as some people do, something I have already shown to be the case up above. Then, would you be here arguing that in accordance with our national-varieties-of-English rules, the practices of the first major contributor need to be restored? I doubt very much that you would.

But since I really can't speak for what you might do, let's just make it clear that I most certainly would not be doing so. It isn't a national-varieties-of-English issue. I'd be opposing that argument vigorously; I'd still be insisting that this needs to be presented in a neutral point of view, by using the word "energy" when we mean "energy", rather than using the name of one of the two units in widespread use to measure that energy as a synonym for that energy. Gene Nygaard (talk) 11:39, 17 January 2010 (UTC)[reply]

"The diet is effective in half of the patients who try it, and very effective in a third."

Does this mean that the diet is ineffective in one-half or one-sixth? Are we looking at ineffective + effective + very effective being measured separately (17% + 50% + 33% = 100%), or are the "very effective" third part of the "mediumly effective" half?

If the numbers run 17% + 50% + 33% = 100% (ineffective through very), then the second clause should be re-phrased as "...very effective in another third". If the numbers instead run 50% + (17%+33%) = 100%, then the first clause should be re-phrased as "The diet is either effective or very effective in half...". WhatamIdoing (talk) 19:57, 16 January 2010 (UTC)[reply]

It is the second case. Half the patients find it ineffective (reduces seizure frequency by less than 50%). Later we put it another way: "The ketogenic diet reduces seizure frequency by more than 50% in half of the patients who try it and by more than 90% in a third of patients." I was hoping that "effective" without any qualifiers (such as "moderately effective") would be assumed to include the "very effective" set of patients. For that reason "effective or very effective" reads oddly to me. I agree that it is good to be clear on this: something my sources don't always manage. Perhaps there's another way of phrasing it that is clear but simple? Do other people find the existing text ambiguous? Thoughts? Colin°^Talk 20:44, 16 January 2010 (UTC)[reply]

Maybe just reverse it so it's clear the first is included in the second?

"The diet is very effective in a third of the patients who try it, and effective in half."

or:

"The diet reduces seizures considerably in a third of the patients who try it, but reduces them to some degree in half of all patients."

or something like that. SandyGeorgia (Talk) 21:00, 16 January 2010 (UTC)[reply]

The data would line up with the inelegant phrase, wouldn't it? Here are some other options:

• ""The diet is ineffective in half of the patients who try it, helpful to one-sixth, and very effective for the remaining third."
• "The diet is effective in half of the patients who try it. Although half of patients receive no benefit, it is very effective in a third."
• "The diet is effective in half of the patients who try it."

WhatamIdoing (talk) 20:56, 16 January 2010 (UTC)[reply]

I guess I misunderstand then; I thought Colin was saying the opposite. SandyGeorgia (Talk) 21:00, 16 January 2010 (UTC)[reply]

I'm not sure I'm following the "misunderstanding" Sandy mentions. The half (50%) who find it reduces their seizure frequency by less than 50% may disagree that it was "no benefit" or that it wasn't reduced "to some degree". It is just that the 50% reduction is a typical benchmark for anticonvulsants being regarded as "effective". Colin°^Talk 21:36, 16 January 2010 (UTC)[reply]

Here's some examples from the sources: "the ketogenic diet reduced seizures by >90% in a third of the patients and by >50% in half" and "approximately half of children having at least a 50% reduction in seizures after 6 months. Approximately one third will attain a 90% reduction in their seizures." Colin°^Talk 21:54, 16 January 2010 (UTC)[reply]

Two thoughts. First, how about just mentioning one (either the half or the third) and not mentioning the other? It is the lead, after all, and it doesn't need to mention every statistic. Second (and here I am to some extent contradicting the first point, but never mind), how about briefly saying what "effective" means rather than "effective", a word that is vague to the average reader? In short, I propose replacing the sentence with "The diet reduces seizures by at least 90% in a third of the patients who try it." By the way, congratulations (or should it be, commiserations?) on making the main page; I'll be watching, but will not be on-Wiki as much as I'd like. Eubulides (talk) 22:12, 16 January 2010 (UTC)[reply]

That would work for me. I like the simplicity and the focus on the major outcome (i.e., it may not work for you, but if it does, it really works). If wanted, it could be extended to say "...and reduces seizures to a lesser extent in some other people." WhatamIdoing (talk) 02:16, 17 January 2010 (UTC)[reply]

The graph, http://en.wikipedia.org/wiki/File:Ketogenic_diets_pie.svg, seems to imply that all dietary fats are long chain, except for the medium-chain triglycerides included in that variant of the diet. However, I believe there should be appreciable amounts of short and medium chain fats in all of the diets. For instance, cream is cited as a major component of the standard ketogenic diet, and milk fat contains about 8% medium-chain fatty acids and about 3% short-chain fatty acids. ^[1] Perhaps the graph should be clarified in some way. Bluethegrappler (talk) 03:16, 17 January 2010 (UTC)[reply]

Thanks. I'm a bit busy this morning but I'll give a fuller response this afternoon, with a suggested change to the chart. Colin°^Talk 10:04, 17 January 2010 (UTC)[reply]

From what I've been able to tell, vegetable oils tend to contain very little short or medium chain fatty acids. The exceptions are coconut oil and perhaps to some extent palm oil. The classic diet doesn't factor the MCT portion of dietary fat as part of its calculations. The fat does typically come from dairy but doesn't have to. Therefore it would probably be quite hard to get a figure for the portion of MCT in the classic diet. In addition, this chart shows percentage of calorific contribution to the diet, not the proportion by weight. The charts would look quite different if based on weight as fat has over twice the kcal/g of carbs. To complicate things MCT has 8.3 kcal/g compared to 9.0 for LCT.

I think the best thing to do would be to modify the key. Replace "Fact LCT" with "Dietary fat (mostly LCT)" and replace "Fat MCT" with "MCT oil". That should make it more honest even if we can't (without conducting some original research) make the charts more detailed. Colin°^Talk 15:30, 17 January 2010 (UTC)[reply]

Hmm, will work on this. Perhaps instead of making MCT a different color, we should make it the same color as "LCT" and shade it, to imply that MCT is a subset of dietary fat. Fvasconcellos (t·c) 15:40, 17 January 2010 (UTC)[reply]

The problem is that the dietary fat portions in our charts do contain some MCT also. We don't know how much but probably enough to be visible. So really the MCT portion should be bigger and appear in the other charts too. Which is why being specific that it is MCT oil rather than MCT fatty acids we remove the problem. Colin°^Talk 15:52, 17 January 2010 (UTC)[reply]

How's this? Fvasconcellos (t·c) 15:54, 17 January 2010 (UTC)[reply]

With the key changed, of course. Fvasconcellos (t·c) 15:55, 17 January 2010 (UTC)[reply]

Ah. Are you proposing a change to the shading + key change. That's fine. Colin°^Talk 16:00, 17 January 2010 (UTC)[reply]

Yes, there'd be shading + key saying "carbohydrates, protein, dietary fat, MCT oil". Fvasconcellos (t·c) 16:06, 17 January 2010 (UTC)[reply]

Excellent! Since there seems to be concensus, I went ahead and inserted the new graph. Also, I changed the wording of one sentence to say "Normal dietary fat contains mostly long-chain triglycerides", to clarify that even normal dietary fat is not composed solely of long-chain triglycerides. Bluethegrappler (talk) 00:29, 18 January 2010 (UTC)[reply]

It's intriguing that caloric restriction and anticonvulsants both increase lifespan in model organisms and both have some effects in common with the ketogenic diet. I didn't see any clear indications in the literature in a superficial search, but I wonder if a deeper look would turn up something. Were lifespan effects (or telomere shortening) discussed during development of the article? Wnt (talk) 04:52, 17 January 2010 (UTC)[reply]

Thanks. I'm a bit busy this morning but I'll look at my sources on this in the afternoon. Calorie restriction is mentioned in some sources. There's very little data on lifespan issues wrt the ketogenic diet as very few people have been on it for a long period of time (decades say), but there might be some data on mice and the KD. Colin°^Talk 10:04, 17 January 2010 (UTC)[reply]

I haven't found anything regarding lifespan or telomere shortening wrt the ketogenic diet. In fact, PMID 3104076 (which is fruit fly research admittedly) might indicate such a diet may not be the best choice for a long life. The ketogenic diet, deficient in most fruits, vegetables and pulses, is not recommended by the major heart and diabetes organisations as a long-term option for weight control. Over the long-term, a standard low-calorie diet has the same efficacy as these low-carb diets but allows plenty healthy foodstuffs that we know help avoid the diseases of ageing.

The ketogenic diet is calorie controlled. Although the calories are initially set slightly below the RDA, this is due to fat taking less energy to metabolise. The children are monitored to make sure they maintain a healthy weight and growth pattern and the diet adjusted accordingly. Although overweight children may initially aim for a reduction in their weight, being underweight is not the aim. I'm aware some folk use calorie restriction to be permanently underweight and hungry in the hope this extends their life as it does with lab rats. Colin°^Talk 15:49, 17 January 2010 (UTC)[reply]

You've made some excellent points there, Colin. So much so, that I have to wonder if it's possible to include some of them in the article - eveen if it's saying "There is no evidence that ..."? (This happened to me when my FA was on the main page. Answering questions on the talk page made me consider issues not discussed in the main article.) You are going to be the best judge of WP:WEIGHT, so there's no problem if you think it's not appropriate. --RexxS (talk) 19:11, 17 January 2010 (UTC)[reply]

I agree that talk page queries are a good way of finding new material to add. So please fire away with your questions. I covered several points above, a few of which are in the article. Which ones in particular do think could be included? I did consider adding some text about very-low-carb diets for obesity or diabetes -- my notes are in the latest archive. It is a tricky one as we do have Low carbohydrate diet and diabetic diet articles, which are probably more focussed towards such issues. Many such diets aren't ketogenic or are only so for a short initial period. The problem with the telomere/life-extension issue is that it is one thing for me to say "I've found no evidence" but another thing for me to find a source that says this. Life extension isn't a concern of most of the researchers: these sick children have more immediate problems than worrying about whether they will live to 90 rather than 80. Colin°^Talk 19:50, 17 January 2010 (UTC)[reply]

Actually, the responses to the questions are the best place to find new material, because it forces the person who already knows the material well to go into more detail when making an explanation! Please forgive my lack of familiarity with the article, but for example does it cover:

1. "Although the calories are initially set slightly below the RDA, this is due to fat taking less energy to metabolise." – (Implementation ?)
2. "Although overweight children may initially aim for a reduction in their weight, being underweight is not the aim." - ( Indications ?? not sure)
3. "The ketogenic diet, deficient in most fruits, vegetables and pulses, is not recommended by the major heart and diabetes organisations as a long-term option for weight control." – ( Indications or Other applications ?)

I can see that #2 may be difficult to source well, and that #3 is stating a 'non-application', but those factoids are interesting. You're still the best judge of whether they belong in the article. Apologies again if they are covered and I missed them. Take heart that the worst part of today's ordeal is almost over :) --RexxS (talk) 21:13, 17 January 2010 (UTC)[reply]

I can source all three, though the "being underweight is not the aim" is probably just referred to as aiming for a healthy weight. Point one is covered in the Variants-Classic section -- perhaps it should move elsewhere. The slight problem with moving it is that the calculations for the classic diet are well documented; the ones for the MCT diet (modern proportions) are not. We do mention that overweight children go on a 3:1 ratio but we don't say (as far as I can see) that they aim for normal weight first -- that should certainly be added. The last point is mentioned in those obesity/diabetes thoughts that are in the archive and I guess we should include some of that in the "Other applications" section.

Today has been very busy in real life so I'm very relieved that it hasn't been an "ordeal" online at all. In fact, my wikifriends and others have beaten me to whack the vandals every time so I've yet to edit the article today. I have the pleasant task of responding to talk page questions like this. I may not get around to adding this new material tonight. I'm off to relax for an hour with tonight's episode of "Being Human". Colin°^Talk 21:40, 17 January 2010 (UTC)[reply]

I can see now that #1 is covered. That was just me looking in the wrong place, so please don't worry about moving it. I'm really glad today hasn't caused you wiki-stress and I'm sure you'll be able to review all that will come out of it as time allows. There's still no deadline! --RexxS (talk) 22:01, 17 January 2010 (UTC)[reply]

I would like references to anticonvulsants extending lifespan in model organisms. A very promising possibility is that low carb dieting in general (which includes ketogenic dieting and is sort of related to low-glycemic-index dieting) activates life-extending mechanisms (via the insulin/IGF1 pathway) which were originally selected for because they helped organisms cope with famines. See Cynthia Kenyon "The genetics of ageing" http://www.nature.com/nature/journal/v464/n7288/abs/nature08980.html . Dr. Kenyon is I *think* responsible for discovering the genetic basis of aging -- before her work I think the dominant paradigm was that aging was an entropic mechanism sort of like your car wearing out. — Preceding unsigned comment added by Zooko (talk • contribs) 22:30, 17 March 2011 (UTC)[reply]

This discussion is probably more suited to the low-carbohydrate diet talk page. Colin°^Talk 20:17, 18 March 2011 (UTC)[reply]

This edit added a citation needed tag. The tagged sentence is supported by the reference cited at the end of the next sentence. The source is free online so you can check yourself if you doubt the facts. I would prefer not to have to repeat the number at the end of every sentence. This article is fully cited but the number may appear after several sentences or even the end of the paragraph. I have asked Lofnazareth to reconsider the tag. Colin°^Talk 15:20, 17 January 2010 (UTC)[reply]

I see the citation has now been repeated. Oh well. If it is contentious then best to play safe. Colin°^Talk 15:22, 17 January 2010 (UTC)[reply]

Yep. Fvasconcellos (t·c) 15:35, 17 January 2010 (UTC)[reply]

Gluconeogenesis: it is very common to get glucose from protein; you do not have to burn fat if carbohydrate is low. --94.70.98.95 (talk) 20:42, 17 January 2010 (UTC)[reply]

Yes, you can burn protein instead, but the KD has only adequate protein for other purposes than energy requirements so fat is burned instead. On a low-carb high-protein diet like the Atkins diet, you would burn protein too. Colin°^Talk 20:59, 17 January 2010 (UTC)[reply]

Aye, I was aware this may be technically covered in the article itself, hence I said 'misleading' and 'main page'. It may be misleading if read only on the main page in a quick read to think "nice, if I cut bread I'll only burn fat", which would be disastrous if they guy eats tons of meat. --94.70.98.95 (talk) 21:11, 17 January 2010 (UTC)[reply]

Going on the ketogenic diet without medical supervision is asking to end up in the casualty department. I hope both the main page and this article make it clear this is a medical therapy, not a diet-book-fad idea. Colin°^Talk 21:45, 17 January 2010 (UTC)[reply]

Just curious: how could going on a ketogenic diet without medical supervision kill you? Well, let's assume you don't do the initial fast, restricted calorie, restricted water version but instead a more liberal version without those restrictions (which I think has become the de facto standard, per Kossoff 2008 "Optimal clinical management of children receiving the ketogenic diet: Recommendations of the International Ketogenic Diet Study Group"). As far as I know there is no way doing that could kill you. Zooko (talk) 05:34, 28 March 2011 (UTC)[reply]

The comments on starvation are, IMHO, not NPOV and, regardless, are not sourced. It is true that success of the ketogenic diet in the past has been commonly attributed to mimicking starvation, but many scientists today have rejected this viewpoint as misguided. That is to say that it is well understood that many societies, and indeed early mankind, have consumed high-fat, low-carb diets as their predominant diets; so saying that the body is confused by what has been historically a common diet is not consistent with modern research. I would say that it is better to remove this kind of material as it more smacks of pushing a viewpoint than providing educational benefit.

--Mcorazao (talk) 20:25, 18 January 2010 (UTC)[reply]

P.S. Before anyone complains, I am not specifically saying that early societies have specifically eaten this diet. But the implication here is that raising fat and lowering carb intake makes the body think it is starving which is dubious. --Mcorazao (talk) 20:28, 18 January 2010 (UTC)[reply]

The comments on starvation are sourced -- the first and ninth reference are both used to support statements about starvation and both are freely available online so you can check yourself. I would be interested to know of papers published by these scientists who reject the ketogenic aspects of starvation and the ketogenic aspects of a very low carb + high fat diet. I would doubt very much that primitive man could adopt this ketogenic diet, for it needs considerable dairy fat and vegetable oils not to mention the vitamin pills that compensate for the lack of fruit, vegetables and pulses. That ketosis occurs during starvation is well known and is due to burning body fat. A ketogenic diet aims to also produce ketosis but supplies fat so that weight is not lost and supplies protein so you don't waste away. The carb content has to be really low to achieve this. It isn't really that the body "thinks it is starving" but that a process that occurs during starving also occurs on this diet. Somehow that process is beneficial in epilepsy. Colin°^Talk 22:17, 18 January 2010 (UTC)[reply]

If you read the history of the diet you also find that it was specifically designed to mimic the body chemistry of an individual starving, with out actually starving them. The idea that this diet is or has ever been common is also nearly ridiculous. 4 grams of fat per every gram of combined carbohydrates and protein is not even possible in a primitive society. the only culture in history that has ever even gotten close is the inuit I believe, and even they had more protein than fat.Nitack (talk) 20:26, 28 September 2010 (UTC)[reply]

Do you have a source for the Inuit getting more protein than fat? Vilhjalmur Stefansson said they got about 80% of their calories from fat, about 20% from protein, and practically none from carbohydrate. If it was true that they were getting more calories from protein than from fat then even if they were sneaking in 10% carbs without Stefansson noticing they would have to get at least 45% of their calories from protein, but that seems very unlikely. See this blog post by Stephan Guyenet, especially the first two references about maximum protein intake in humans: http://wholehealthsource.blogspot.com/2008/12/myth-of-high-protein-diet.html . Almost certainly all Inuit were in deep ketosis for their entire lives until Westernization arrived. — Preceding unsigned comment added by Zooko (talk • contribs) 06:44, 18 March 2011

I suspect the "more protein than fat" statement is wrong and would probably not be achievable. However, when compared to a normal western diet, it is a high protein diet, as are most lifestyle low-carb diets. For the ketogenic diet to be effective in epilepsy, you need to burn the fat rather than protein or carbs, so protein is limited to just enough for childhood growth and body repair. Colin°^Talk 20:15, 18 March 2011 (UTC)[reply]

I thought that Modified Atkins Diet was just as successful (so far) as KD? Wouldn't that mean that it doesn't matter if you have higher protein than the minimum needed? Zooko (talk) 05:36, 28 March 2011 (UTC)[reply]

Not sure what to make of this recent review: it is speculative, but on the other hand the current article says very little about the mechanism of the ketogenic diet, saying only that it's a mystery (listing some disproven hypotheses), and it surely wouldn't hurt to list at least one active hypothesis.

• Masino SA, Kawamura M, Wasser CA, Pomeroy LT, Ruskin DN. Adenosine, ketogenic diet and epilepsy: the emerging therapeutic relationship between metabolism and brain activity. Curr Neuropharmacol. 2009;7(3):257–68. doi:10.2174/157015909789152164. PMID 20190967.

Eubulides (talk) 20:08, 4 March 2010 (UTC)[reply]

Thanks. I'll read it in detail later. I'm struggling to find time for Wikipedia research at the moment due to real-life pressures. Plus, much of the mechanism stuff just goes way over my head and I realise this is a weakness in the article. If there's someone reading this who can read and understand these papers fully then please contact me on my talk page. The following article provides quite a good summary of various hypotheses and the one after that is a letter by the same author as the above paper.

Our article doesn't just list disproven hypotheses; the short paragraph "The ketone bodies are possibly anticonvulsant in themselves...and may confer a neuroprotective effect." lists a couple of ideas. Perhaps this could be expanded. I'm unsure how to cover the adenosine hypothesis wrt WP:WEIGHT without a review that compares several. Clearly the above paper is written to advance one hypothesis that the authors favour.

• Rho JM, Sankar R. The ketogenic diet in a pill: is this possible?. Epilepsia. 2008;49 Suppl 8:127–33. doi:10.1111/j.1528-1167.2008.01857.x. PMID 19049610. PMC 2692867.
• Masino SA, Geiger JD. The ketogenic diet and epilepsy: is adenosine the missing link?. Epilepsia. 2009;50(2):332–3. doi:10.1111/j.1528-1167.2008.01771.x. PMID 19215286.

Colin°^Talk 21:41, 4 March 2010 (UTC)[reply]

This is just to acknowledge that I have read this discussion. I'll read the papers and if I can help, I will. Graham Colm (talk) 21:59, 4 March 2010 (UTC)[reply]

Thanks Graham. The basic science behind the KD is a big topic where there's lots of research. If we applied WP:WEIGHT to the Mechanism section based on the contents of PubMed, it would be at least half the article, but we'd lose most of our readers. Literature reviews on the KD aimed at clinicians tend to have a short Mechanism section but bigger than ours. They can start the discussion with certain assumptions about reader knowledge. I think the topic deserves more coverage here than it does at present so it would be great if we could expand it. More papers:

• Nylen K, Likhodii S, Burnham WM. The ketogenic diet: proposed mechanisms of action. Neurotherapeutics. 2009;6(2):402–5. doi:10.1016/j.nurt.2009.01.021. PMID 19332336. -- I have this paper as a PDF.
• Freeman J, Veggiotti P, Lanzi G, Tagliabue A, Perucca E. The ketogenic diet: from molecular mechanisms to clinical effects. Epilepsy Res.. 2006;68(2):145–80. doi:10.1016/j.eplepsyres.2005.10.003. PMID 16523530.
• Epilepsia Special Issue: Ketogenic Diet and Related Dietary Treatments Volume 49 Issue s8 , Pages 1 - 133 (November 2008) -- I think all the papers in this issue are freely available and the "In a Pill" paper is part of that issue.

Colin°^Talk 19:11, 5 March 2010 (UTC)[reply]

Today turned out to be an unexpectedly busy one. I'm at work all this weekend, but I have next week free, (my Spring break) to devote to this. It will be an interesting challenge. Graham Colm (talk) 21:45, 5 March 2010 (UTC)[reply]

No problem, Graham. I'll try to do some research myself and collate what I've found here. Colin°^Talk 09:36, 7 March 2010 (UTC)[reply]

I had to memorise the Krebs cycle when I was at university but this has been of little use to me until now. I have read the papers and after much contemplation, I think this article gets the balance just right wrt weight. Colin is right in saying that any Mechanism section that tries to include all the hypotheses and experimental results would be huge, and I add, of little value to readers. The current thinking that the mechanisms are "multiple, parallel and synergistic", first written in Bough and Rho (2007) and repeated in the "pill" article, is probably the most accurate description of what is known. On reading these papers, it is difficult to resist the temptation to form one's own theories. In my view any attempt to expand this section, at this time, would be difficult, and the adenosine hypothesis seems just as speculative as some of the others. I think we should be patient and wait for a proven mechanism, or at least an hypothesis that has very strong experimental data to support it. Graham. Graham Colm (talk) 21:06, 13 March 2010 (UTC)[reply]

OK; thanks for looking into it and summarizing. Eubulides (talk) 01:29, 14 March 2010 (UTC)[reply]

Thanks Graham. I'm sure the section could be improved. But it probably isn't worth trying to teach the reader about some obscure metabolic or neurological process for the sole reason of helping them understand some speculation that may turn out to be duff. Colin°^Talk 20:48, 14 March 2010 (UTC)[reply]

Whilst I don't doubt that treatment of epilepsy in children is important, it seems wrong that this article focuses so heavily on this use of a ketogenic diet. The majority of people following a ketogenic diet worldwide will be doing so for health or weight loss reasons (e.g. induction phase of atkins/low carb diets) and this should be reflected in the article.--Hontogaichiban (talk) 03:12, 24 February 2011 (UTC)[reply]

It is 100% weighted towards the treatment for epilepsy aspect. There's a short section at the end mentioning its medical use for other serious conditions, but they are just at the research stage. If you look at the hatnote (the italic sentence at the top), it points readers wanting to learn about other uses of a ketogenic-low-carb diet to another article. It really isn't possible to discuss the two radically different usages in one article. Plus many low-carb diets are only ketogenic in certain phases or are only barely ketogenic. Most of those are also high-protein. Colin°^Talk 08:32, 24 February 2011 (UTC)[reply]

In this change,

• I've reverted the change from "The ketogenic diet is not a benign, holistic or natural treatment for epilepsy" to "The ketogenic diet is not necessarily a benign or natural treatment for epilepsy" I think you were implying the above other uses but it ends up changing the meaning to say it is "necessarily a benign or natural".
• I've added a source for the bone fractures and retarded growth. The source came from the following sentence. Generally, a footnote in this article is the source for all the text preceding it.
• I've restored the "calories" comment on the Modified Atkins Diet. The sentence is comparing it with the ketogenic diet, which is calorie-restricted, not the standard Atkins diet.

Colin°^Talk 08:39, 24 February 2011 (UTC)[reply]

This edit changed the figure from 10% to 20% in the sentence "About 20% of children on the ketogenic diet achieve freedom from seizures, and many are able to reduce the use of anticonvulsant drugs or eliminate them altogether" citing PMID 19535814. The original text cited (later in the paragraph) PMID 17241227 (available online here). That source says "About 10% of individuals become completely seizure-free and often then discontinue use of anticonvulsant drugs while on the diet". The 20% figure seems to come from this sentence in the first source "In approximately 20% of patients, anticonvulsant medications can be successfully discontinued without recrudescence of seizures." That source itself cites this to PMID 16499771 and PMID 15256185. I need to read some more and get hold of this last paper. I can't see where the 20% comes from or exactly what population group it is referring to. Is 20% of those who already have seizure freedom on the diet who then discontinue? If so, that's not the same as saying "About 20% of children on the ketogenic diet achieve freedom from seizures" which sounds far too high. Colin°^Talk 22:43, 2 March 2011 (UTC)[reply]

Hi, I'm a newbie at Wikipedia editing, so please bear with me. Recently I removed the sentence "The ketogenic diet is not a benign, holistic or natural treatment for epilepsy", and I see now that my change was reverted.

I think that sentence sounds frightening--it makes it sound like there is something weird, dangerous, or unnatural about the KD--but it is actually not that meaningful, since "benign", "holistic", and "natural" are vague words that are subject to interpretation. The rest of the section which relates actual facts about adverse effects seems sufficient to inform the reader without needlessly scaring them with words like "unnatural" or needlessly reassuring them with words like "natural". It is not NPOV.

As a kind of light-hearted consideration, imagine if you had to write "The KD is X" instead of "The KD is not X". I guess you would have to write "The KD is malign, atomistic, or unnatural.". ;-) Zooko (talk) 22:12, 17 March 2011 (UTC)[reply]

Okay, I changed it to: "Complications associated with the ketogenic diet are generally less severe and less frequent than with anticonvulsant medication or surgery.". I hope this is wiki-correct behavior on my part.

http://en.wikipedia.org/w/index.php?title=Ketogenic_diet&action=historysubmit&diff=419369426&oldid=418921178

Zooko (talk) 22:17, 17 March 2011 (UTC)[reply]

KD isn't (necessarily) benign, natural, or holistic. It carries risks, has complications, is distinctly odd/unnatural, is (very) disruptive, etc. I believe the purpose of this sentence is to prevent some very common misconceptions. It is not automatically "safer" to use food to manipulate the body's physiology instead of using a pill to do the same thing. WhatamIdoing (talk) 02:55, 18 March 2011 (UTC)[reply]

So, risks and complications I understand. The referenced articles go into detail (some of the worst adverse effects were caused by the initial fast and/or limitation on water intake, which if I understand correctly is no longer used in the modern KD protocol). So something like the following would be accurate: "The ketogenic diet has risks and complications, although they are less severe and less frequent than those from anticonvulsant medications or surgery". But "Odd/unnatural" sounds either wrong or misleading to me. A lot of people, especially people who haven't studied humans scientifically think that things humans do can be separated into "natural" and "unnatural" and the former are healthy and/or moral and the latter are unhealthy and/or immoral, but that doesn't really mean much scientifically. Traditional Inuit, for example, would presumably think that our modern habit of eating all sorts of fruits, vegetables, and grains was distinctly odd/unnatural, and I guess they would be just as correct! (They certainly enjoyed better health on average.) So in that sense it is more of a cultural judgment than a scientific one. If you mean that it is rare or uncommon, then that's certainly true but it isn't really relevant to a section on adverse effects. If you mean that it may have adverse effects then that is also certainly true, but we should just say that instead of alluding to some scary, ill-defined, culturally-determined property. — Preceding unsigned comment added by Zooko (talk • contribs) 06:44, 18 March 2011 (UTC)[reply]

Let's deal with the three words. It is not "benign" because it can cause harm. This can range from something requiring emergency medical treatment or longer term like stunted growth or kidney stones. The side effects are a significant risk no matter what the initial protocol is -- nobody suggests trying this diet without close expert medical supervision. It isn't "natural" because our bodies were designed to eat carbohydrate as our main source of fuel and following this diet requires vitamin and mineral supplements. It is not "holistic" because it only deals with physical health rather than mental, emotional, social and spiritual aspects. Don't confuse this diet with the Inuit diet or other ketogenic low-carbohydrate diets. A discussion and comparison of modern lifestyle diets with various primitive diets or weight-loss diets really belongs on the talk pages of low-carbohydrate diet.

I have many review papers on the KD that begin discussion of side effects with a statement countering a belief held by some parents and health practitioners that this diet is "holistic", "all-natural" or "alternative". For example, in PMID 9823821, an Editorial in Archives of Neurology Nov 98, ES Roach says "Many parents are initially drawn to the ketogenic diet because they assume that a diet must somehow be more natural and less risky than a medication. They soon find that there is nothing at all normal or natural about the ketogenic diet!" In "The Ketogenic Diet" book (ISBN 1-932603-18-2), which is aimed mainly at parents, the "Myths and misunderstandings" section begins with 'Contrary to the belief of some parents, the ketogenic diet is not "all natural", "holistic", "organic" or "pure."' So I think that per WP:WEIGHT there is sufficient evidence that this is a widespread misconception and therefore should be mentioned by the article. Colin°^Talk 20:03, 18 March 2011 (UTC)[reply]

Also, KD isn't just 'eat lots of fat and protein, with almost no carbs'. In the more restrictive forms, KD means 'eat exactly this amount of food, in precisely this balance, on a fairly rigid schedule, with zero extra if you're hungry, and zero less if you're not'. There are zero cultures that have required children to eat only precise amounts of food at set times. WhatamIdoing (talk) 18:18, 22 March 2011 (UTC)[reply]

The page says "Case reports of two children indicate that it may be a possible treatment for astrocytomas, a type of brain tumour.". Please, if anyone has a source for this I urgently would like to read it. Thanks. (Also, if nobody does then perhaps we should remove that sentence. :-)) Zooko (talk) 22:51, 17 March 2011 (UTC)[reply]

Click here to see half a dozen. WhatamIdoing (talk) 02:51, 18 March 2011 (UTC)[reply]

This article is 100% sourced. If you want to find the source for a statement, just look along the line or paragraph for the next superscript number hyperlink and it will take you to the reference. In this case, I used a review, which is freely available online here. The review cites the two case reports, if you want to look them up. Colin°^Talk 20:08, 18 March 2011 (UTC)[reply]

People use "ketogenic diet" to refer to other things than these treatments for epilepsy. Body-builders and weight-loss practitioners use the phrase "ketogenic diet" to refer to any diet which promotes a continuous state of ketosis, e.g. Lyle McDonald "The Ketogenic Diet": http://www.amazon.com/Ketogenic-Diet-Complete-Dieter-Practitioner/dp/0967145600 . Oncologists have used a ketogenic diet as a treatment for cancer, e.g. Nebeling 1992 "Effects of a Ketogenic Diet on Tumor Metabolism and Nutritional Status in Pediatric Oncology Patients: Two Case Reports". The American Heritage Stedman's Medical Dictionary defines "ketogenic diet" simply as "A high-fat, low-carbohydrate diet that includes normal amounts of protein.".

Wikipedia needs a place to describe ketogenic diets in general -- any diet which promotes ketosis has a lot in common with any other diet which promotes ketosis (as evidenced by the fact that the Modified Atkins Diet treatment for epilepsy has similar good results as the traditional or modern Ketogenic Diets for epilepsy). The current paradigm of reserving this page for epilepsy treatments and directing other ketogenic diets to the low-carb page is suboptimal, since most low-carb diets are not ketogenic and facts about ketogenic dieting would be less relevant on a low-carb page and more relevant to the medical treatment of epilepsy, cancer, etc. — Preceding unsigned comment added by Zooko (talk • contribs) 16:05, 25 March 2011 (UTC)[reply]

The term "Ketogenic diet" was invented by the epilepsy researchers who came up with the diet described in this article. Since "ketogenic" is just an adjective, other diets have arisen that can be described as "ketogenic" or are "ketogenic" for a limited period (such as the induction phase of the Atkins diet).

Does Wikipedia need a "place to describe ketogenic diets in general" any more than it needs a place to describe "green things" or "expensive things"? Do people talk about ketogenic diets in general? To a very limited degree, and I reckon mostly at the level of speculation. Pretty much all the academic research on the ketogenic diet for the last 90 years has focused on very ill children, children who are fed meals prepared by their parents or get fed through a tube, children only using the diet for at most a few years, and on whether their epilepsy gets better. It is a diet so complex that it is custom prepared and adjusted by a dietitian and supervised by neurologists and specialist nurses. Can that research be extended to say anything about adults, well people, people changing their diet for life, people wanting to lose weight or avoid heart disease? Not really, other than some educated guesses. Conversely, can anything written about body-building adults be at all relevant to sick children with epilepsy. No.

Medical use of the ketogenic diet for cancer is at the research stage. It is way too early to start calling at "treatment for cancer". I'm not aware that the use of a ketogenic diet for body building is anything more than a fringe practice. Most very low carb diets for weight loss or designed as long-term diets for treating diabetes or preventing heart disease are only barely ketogenic or are only ketogenic initially for the simple reason that it is unpalatable and if you expect people to voluntarily stick at a diet, then that's a big put-off.

Per WP:COMMONNAME this article discusses the typical usage of the term. The hatnote points readers at low-carbohydrate diet for "ketogenic low-carbohydrate diets as a lifestyle choice or for weight loss". That article could be improved greatly (I'm not interested in lifestyle diets and most of the sources on that subject are poor so it isn't attractive to me). That article could contain a section on ketogenic diets. If that section expands to the point where a dedicated article is needed, then we can consider what to call such a thing. Colin°^Talk 19:48, 25 March 2011 (UTC)[reply]

Interesting! I just learned that Dr. Wilder coined the term "ketogenic diet" in 1921. While I do like etymology for its own sake, I guess it is common usage rather than etymology that we want to follow here. :-) I suspect the use of the phrase "ketogenic diet" to mean things other than epilepsy treatment is more widespread than you realize. For example, google search for "ketogenic diet" yields this wikipedia page, five pages about KD for epilepsy, three about KD for weight loss and body-building, and one about Modified Atkins for epilepsy. It also mentions that there are two books on the topic, one of which is about epilepsy (Freeman 2006) and the other of which is about weight loss/body-building (McDonald 1998). If I understand correctly, your suggestion is that information about ketogenic dieting for purposes other than epilepsy treatment could be included as a section on the low-carbohydrate diet page. That sounds pretty reasonable, considering the hatnote. Frankly, the low-carbohydrate diet page looked like such a train-wreck of bias and slop, when I looked at it, that I was discouraged from trying to contribute to it. Zooko (talk) 05:24, 28 March 2011 (UTC)[reply]

One of these days, I'll write an essay about the misuse of Google search results in an argument :-) McDonald's book is self-published and his income derives from people finding his personal website(s) and buying his books. Naturally, he has an interest in ensuring his site is found in the first page of Google results. Like any minority group, those who follow such a diet regime will blog about it and sing its praises. See WP:WEIGHT (which isn't just about opinions but about balance of coverage). We judge this by what is written about the subject in reliable publications. What do reliable publications on body building diet have to say about ketogenic diets? What do reliable subjects on weight loss diet, or diets for diabetes have to say about ketogenic diets? We need to make sure WP covers this in a neutral way, rather than repeat the evangelistic praises of a diet "guru" and his disciples.

I would be interested in the essay on the problems of using Google search results as a measure of usage. :-) It wasn't that I did the google search and that convinced me that "ketogenic diet" was widely used for other things, it was that I've already heard it used that way for many years. However, I can't point you to all those conversations I've seen, but I can point you to google search results as an indicator that they must be widespread. Using a google search is kind of like looking for your lost keys under the streetlight instead of where you lost them. :-) Anyway, having just re-read WP:WEIGHT, I'm not sure how it applies to this. The people who use "ketogenic diet" to mean the way they eat (for example when body-building) aren't expressing a difference of *opinion* contrasted with the epilepsy researchers—they're just using the same phrase to mean a closely related but slightly different practice. "What do reliable publications on body building diet have to say about ketogenic diets?" I don't know. That's a good question. I guess the way WP:WEIGHT applies is that the use of the phrase to mean anything other than a treatment for epilepsy could be used only by a tiny minority of people? That would, I agree, be a reason to omit it from the wikipedia page. I guess I'm confused about the difference between theories or opinions, which is what WP:WEIGHT seems to be mostly about, and English usage—I currently have the idea that "ketogenic diet" is used by a non-tiny set of people to mean ketosis-producing eating in general.

Hm, so if I understand correctly (it is interesting learning how wikipedia is made), then "What do reliable publications on body building diet have to say about ketogenic diets?"—if they don't mention it, then we shouldn't mention it here, if they mention it then we should mention it here, although what we say about it may be influenced by how they describe it. Is that right? Zooko (talk) 05:41, 2 April 2011 (UTC)[reply]

Yes, but you missed the first step. The very first step is "decide what the page is about". If you decide—as was done here—that the subject of this page is the ketosis-triggering diet that is used as a formal medical therapy, then what body-builders or weight-loss gurus happen to think about some tangentially related, plainly non-therapeutic diet is completely irrelevant ("undue", in wikispeak). WhatamIdoing (talk) 06:18, 2 April 2011 (UTC)[reply]

Okay, so this page is about *that* ketogenic diet, and there is a note at the top saying "If you were looking for another thing called 'ketogenic diet', then look over there.". Got it. Thanks. Zooko (talk) 03:58, 3 April 2011 (UTC)[reply]

Yes, the low-carbohydrate diet page is very poor. It probably attracts folk who want either to promote or disparage such diets. If you approach the topic neutrally, and invest time in finding out what experts have to say, then you could certainly improve it. Colin°^Talk 07:33, 28 March 2011 (UTC)[reply]

Ah, well I guess I'm too biased to take on that task. I have a hard time finding bad things to say about low-carbohydrate diets. At the moment I don't understand how edits I made to that page, even if my edits were themselves small incremental improvements, would really make it more likely that the page would ever become good. Zooko (talk) 05:41, 2 April 2011 (UTC)[reply]

The article discusses several ketogenic diets with quite different compositions, including traditional, modified Atkins, and MCT. Although the term "the ketogenic diet" has been widely used (more so in the older literature) it has become increasingly misleading. There's no need to use it here.

Accordingly, I've edited the article to replace "the ketogenic diet" with phrases that include "ketogenic diets", "a ketogenic diet", "therapeutic ketogenic diets", and so forth. This (and the judicious insertion of the qualifier "therapeutic" in places) also avoids some of the controversy reflected in the above discussion of the scope of the article; without in any way changing the scope or the content, it avoids the incorrect implication that the term "ketogenic diet" denotes a unique medical therapy.

Although I've tried to edit the surrounding text for consistency with these changes, the article could no doubt benefit from a proof-reading pass by someone with a fresh eye. — Preceding unsigned comment added by 66.127.52.100 (talk) 04:27, 6 June 2011 (UTC)[reply]

Although there are variations, all these therapeutic diets are ketogenic and are used to treat epilepsy. They vary in their ketogenic ratio and whether special oils are used or how strict the diet is. The literature uses the term "classic ketogenic diet" to refer to the original form, which is still the most popular in the US, and the one for which most studies have been performed. The MCT ketogenic diet has been shown to be equivalent to that by an RCT -- I'm pretty sure the literature is happy to include that under "the ketogenic diet" banner. Things are less clear with the "modified Atkins diet" and the LGIT. I shall review the literature again but I think that equivalence has not yet been proven for those two and, for example, a medical guideline, or insurance company, recommending the KD would not include those in its scope. If we wish to write "a ketogenic diet" or "ketogenic diets" then we need to be very careful that the literature is referring to any/all of these too.

Although the term "ketogenic" is used as an adjective for some other diets, these are typically [with exceptions for those being scientifically researched] fad lifestyle or bodybuilding diets and are unrelated to this diet, are not prescribed by dieticians or physicians, and are never discussed together in the scientific literature. The patients on this diet are not trying to lose weight (calories are controlled to maintain weight/growth) and since they consume a normal amount of protein, they are not building extra muscle. This article, per the literature, discusses only the therapeutic use of certain ketogenic diets for the treatment of epilepsy. There are other articles on WP that discuss other diets, some of which are ketogenic, or a ketogenic for a period. Therefore, this article does not need to qualify the term "ketogenic" with "therapeutic".

The literature does use the phrase "the ketogenic diet" when discussing the epilepsy diets in this article. It tends to use "a ketogenic diet" when discussing some animal experiments or the research use of a diet that is ketogenic for treating other illness. It doesn't use the phrases "standard ketogenic diet" or "strongly ketogenic diets", for example.

I appreciate the edits must have taken some time and have been done with care. But they have introduced numerous errors or expanded the scope of discussion beyond what the sources support (i.e., they may be technically correct, but we're not aloud to make that leap). I'm going to revert it, I'm afraid. However, I shall review the literature again. This has prompted me to see how our sources distinguish the diet variants, how they discuss the use of a diet that is ketogenic for other research treatments and in animal experiments. If I find I'm wrong about some aspect of what I wrote above, I'll update you and consider how to reword things. This may take me a wee while, as my time is limited. Colin°^Talk 16:00, 6 June 2011 (UTC)[reply]

• Super High-Fat Diet Can Work Miracles for Seizures in Kids. ABC News with Diane Sawyer.
• How Can a High-Fat Diet Treat Epilepsy? Dr. Besser Reports. ABC News with Diane Sawyer.

• I have tried to include a section about the usual Inuit/Eskimo diet (as before contact with Europeans) being ketogenic, but always another user deleted it, e.g. the edit at 12:02, 12 July 2011 "undo Inuit diet. This is not an article on all diets that are ketogenic. This is a medical therapy and wasn't based on or generally gets compared to the inuit diet.) This Google search contains plenty references to the Eskimo diet being ketogenic. Please, is this topic relevant here or not? Anthony Appleyard (talk) 20:43, 31 July 2011 (UTC)[reply]
• "Not" is my opinion. There are a lot of diets that are ketogenic. In fact, plain old fasting is ketogenic. That doesn't make every single diet that happens to produce ketones be the same subject. Wikipedia has half a dozen separate articles on diets that produce ketones. If you're interested in the Inuit diet, then you should help improve Inuit diet. WhatamIdoing (talk) 17:42, 3 August 2011 (UTC)[reply]
• Agree: I agree that the ancient North American diet, as well as most ancient diets, must have been ketogenic. Most ancient societies (prior to agriculture) probably did not have access to large quantities of carbohydrates. Humans, as well as most wild animals today, ate mostly meats or fish, milk, eggs, animal fat, and vegetables having low amounts of digestible carbohydrates. I agree that this would be important to include in the article, since the studies of ketosis as a diet to treat metabolic disorder (common obesity) are few and contradictory. At Wikipedia, articles are crafted by those who edit, so valuable content does not survive unless there is general agreement among editors to retain it. David Spector (user/talk) 20:39, 3 May 2012 (UTC)[reply]

• "Should Include" but for different reasons. It is often stated that the human liver and/or pancreas cannot support a fat based diet, (hence one reason for advocating MCT oil, and even then that the liver/pancreas cannot support MCT oil at very high levels, hence one reason for advocating ketone-esters etc.). But the traditional Inuit diet proves it is possible, at least for that ethnic group. "The reports of these early scientists imply that the Inuit people were physically unhampered despite consuming a diet that was essentially free of identifiable carbohydrate." Ketogenic diets and physical performance,Stephen D Phinney Nutrition & Metabolism 2004, 1:2. Regarding the diet, when asked to reproduce it under supervision It is interesting to note from the careful observations published from the Bellevue study that Stafansson ate relatively modestly of protein, deriving between 80–85% of his dietary energy from fat and only 15–20% from protein. This was, and still remains, at odds with the popular conception that the Inuit ate a high protein diet, whereas in reality it appears to have been a high fat diet with a moderate intake of protein. In his writings, Stefansson notes that the Inuit were careful to limit their intake of lean meat, giving excess lean meat to their dogs and reserving the higher fat portions for human consumption (ibid)

.--Antifesto (talk) 20:09, 4 December 2012 (UTC)[reply]

• The text in question is:-

  ===Inuit===

  For thousands of years the traditional diet of the Inuit or Eskimos (meat, fish, blubber, little or no starch of plant origin, no access to sugar or honey) has been naturally ketogenic, and there are many documented cases of modern non-Eskimo humans living in these societies for extended periods of time. On the other hand, it is speculated that the Inuit may have a genetic predisposition allowing them to healthfully eat a ketogenic diet.

  Anthony Appleyard (talk) 05:05, 5 May 2012 (UTC)[reply]

David's speculation about primitive diets being ketogenic isn't supported by sources. Have a look at the sources in Paleolithic diet. For example, this one gives hunter-gatherer diets at "19–35% for dietary protein, 22–40% for carbohydrate, and 28–58% for fat" compared to US diet of "protein 15.5%, carbohydrate 49.0%, fat 34.0%, and alcohol 3.1% of total energy intake". These US figures correspond with the pie charts on this KD article. Now look at the classic KD for epilepsy. It has 8.4% protein, 1.8% carbs and a whopping 90% from fat. These primitive diets were not ketogenic. They had a lot of meat but it was lean meat. They didn't eat loads of cream and butter. They didn't eat our fatty domesticated animals. They, unlike the inuit, didn't have blubber. So the inuit diet is rather unusual among primitive diets. We have a separate article (Inuit diet) on that diet. WP:WEIGHT requires us to give weight to aspects of a topic according to their weight in high quality sources on the topic. None of the literature on the medical ketogenic diet makes reference to the inuit diet. It isn't based on it. Colin°^Talk 09:58, 5 May 2012 (UTC)[reply]

Thank you for this interesting and relevant information. Do we have an independent corroboration that the ancient Inuit diet was not ketogenic? We need expert opinion, not original research either way. It is an important topic since the transition of the Inuit diet in modern times due to prolonged outside contact may have increased the incidence of morbid obesity and certain diseases (along with tooth decay) enormously. David Spector (user/talk) 17:45, 18 July 2012 (UTC)[reply]

This is off-topic. The subject of this article is a medical nutrition therapy. We have other articles that discuss the Inuit diet. Colin°^Talk 18:08, 18 July 2012 (UTC)[reply]

You are incorrect. The subject of this article is any diet which creates ketosis, a form of fat-based metabolism that generates ketone bodies in the urine. If early human diets created ketosis through lowered carbohydrate and raised fat and oil dietary components, they are precisely relevant to this article. There is no a priori well-known evidence as to whether the original Inuit diet was ketogenic or not. That is why I suggested that expert input would be relevant, rather than relying on what may be obvious to an editor (which would be original research). It is true that the current article is skewed toward medical diets used to cure very specific conditions, you are right there, but that is probably a historic accident, reflecting the interests of an editor. The article title is clear, and the limitation to the medical diet is incorrect. Thank you for your sincere efforts to improve WP. In this case, your specific comments are incorrect. You might consider moving this article to a new title reflecting coverage only of diets used to cure specific diseases. A new article stub for Ketogenic diet would then be created to cover the range and effects of all ketogenic diets, and would include Atkins (induction phase) and brief details of the biochemical and physiological characteristics of such diets (with details in main article ketosis). I would support such a reorganization and would do the work myself if I didn't fear that my amateur work would be reverted by expert and helpful editors such as yourself. David Spector (user/talk) 19:02, 20 July 2012 (UTC)[reply]

No, it's not. The subject of this article is the dietary medical therapy, not "diets that happen to produce ketosis". The article title is not the be-all and end-all of the subject of the page.

Perhaps this will make it clearer: Step #1 in creating this article was to decide that Colin (because it was mostly Colin) wanted to write about the dietary medical therapy that is called "ketogenic diet". Step #2 was to pick an article title for that subject, following the advice found at WP:Article titles, which recommended that Colin follow whatever name the high-quality reliable sources typically use for the subject. Step #3 was to realize that some people use the same term as a description, "ketogenic diet" to mean "any diet that produces ketosis" (i.e., any diet that causes fat metabolism), and therefore to add a WP:Hatnote at the top of the page to warn people that if they're looking for the non-medical diet that sometimes has the same name, then they need to go to the page about the non-medical diet, because this page is all about the medical therapy.

You've got the chicken and egg backwards: the contents of the page control the title; the title does not control the contents of the page. WhatamIdoing (talk) 19:14, 20 July 2012 (UTC)[reply]

The article predates my time on WP and but the original author may well have followed those steps to create an article on the epilepsy therapy, not on all diets that create a state of ketosis. The steps that WhatamIdoing are a necessary consequence of the restriction that we can have only one name for an article, so try to choose the name that is best, but this may be ambiguous. Just as Paris doesn't discuss everything with "Paris" in its name or everywhere called "Paris". Whether or not the Inuit diet was ketogenic is of interest to people studying the diets of the Inuit, not of people reading about an epilepsy therapy. None of the epilepsy literature, for example, mentions the Inuit. And I'm quite sure none of the serious literature on the Inuit mention the epilepsy diet (except perhaps in passing as a "did you know" sort of novelty fact). There's really nothing to be learned from linking the two. Colin°^Talk 22:29, 20 July 2012 (UTC)[reply]

I've removed some text inserted on "Ketogenic Enteral Nutrition". This appears to be a brand name for a product that involves an NG tube combined with a liquid formula, and is currently being advertised as a weight loss treatment for adults. I haven't found any evidence that it is being used by physicians for the treatment of epilepsy. The article already covers similar variations in the "Prescribed formulations" section but those use established powdered formula available on prescription rather than privately from a Harley Street practice. We need information from a reliable source (such as a peer-reviewed journal or professional book) before this can be included. -- Colin°^Talk 13:34, 9 November 2011 (UTC)[reply]

The article says that cholesterol goes up on a ketogenic diet, but the randomized controlled trials of Atkins-like diets show that people do really well on them, probably better than low-fat, and there are long-term studies showing normalizing of cholesterol on the ketogenic diet, e.g., see http://homepage.ntlworld.com/nigel.kinbrum/Long%20term%20effects%20of%20ketogenic%20diet%20in%20obese%20subjects%20with%20high%20cholesterol%20level.pdf

This would seem to require a bit more scientific scrutiny to not be a one-sided misleading wikiperida entry. — Preceding unsigned comment added by 68.2.120.106 (talk) 21:08, 17 August 2012 (UTC)[reply]

Unlike KD, the Atkins diet does not have people getting 90% of their calories from fat. WhatamIdoing (talk) 00:58, 7 November 2012 (UTC)[reply]

I've removed an edit that advanced a hypothesis that the ketogenic diet might cure a "cryptic brain infection". The text is sourced to a blog. I have not found any reliable sources, or research, to support this idea. And it seems most unlikely. The KD is often used in children whose epilepsy is the result of serious brain damage or malformation due to a genetic disease or congenital problem. There's no infection involved. Although epilepsy may be caused by a brain infection such as meningitis and in the third world is very often due to parasitic infection (neurocysticercosis), the epilepsy is due to the acute inflammation or chronic brain damage resulting from such infection. Infections would be treated with drugs such as antibiotics, not by going on a diet. Colin°^Talk 10:22, 22 September 2012 (UTC)[reply]

Qidalar ile cox yaglarin qebulu qanda keton cisimlerinin miqdarinin artmasina sebeb olur ki, bu metodla 1920-ci ildendir epilepsiya xesteliyinin mualicesinde istifade olunur.Kontrol edile bilmeyen epilepsiyalarin kompleks mualicesinde xestelere ketogenik dieta verilir. Qidada olan yaglarin karbohidratlara ve zulallara nisbeti 4:1 nisbetde verilir. Son arashdirmalarda bu usulun sicanlarda beyin shishlerinin mualicesinde yaxshi effekt verdiyi qeyd edilir. — Preceding unsigned comment added by Mamed202 (talk • contribs) 18:53, 6 December 2012 (UTC)[reply]

Google translate (Azerbaijani): "Ketone bodies in the blood of a very fat foods leads to an increase adoption of this method of treatment of epilepsy since 1920 olunurKontrol can not be used in patients with complex treatment epilepsiyalarin ketogenik 4:1 ratio of proteins, carbohydrates and fats in the diet verilirQidada are compared. Last arashdirmalarda yaxshi effect of this method of treatment is shishlerinin the brain in rats." Biosthmors (talk) 02:05, 15 December 2012 (UTC)[reply]

The sentence "As of 2008, research in this area is regarded as having provided insufficient data to produce clear practice parameters for clinical protocols" seems too full of jargon. I'm not sure what a lay reader is supposed to get out of it. And why it is necessary for the lead? Seems like it might need updating. Can it say "best practices for clinicians" instead of "clear practice parameters for clinical protocols"? Biosthmors (talk) 02:01, 15 December 2012 (UTC)[reply]

What do you think about "As of 2008, there has been too little research to clearly identify the best practices"? WhatamIdoing (talk) 05:55, 15 December 2012 (UTC)[reply]

The "identify best practices" and "clear practice parameters for clinical protocols" still sound like text written for physicians. In the end, what counts is that it isn't a standard treatment yet, and the only folk using it are for random case studies and small clinical trials. I've tried to update the lead. The Stafstrom & Rho paper found by Pottinger's cats is pretty good and I'll see if I can used it to expand and update some other stuff. It is exactly the sort of source that bridges the gap between rat research and clinical practice, which WP editors can't do themselves. I guess I should have a surf of PubMed and see if there's anything else new to report... -- Colin°^Talk 08:28, 16 December 2012 (UTC)[reply]

Since you say "This article is only about the medical diet for treating epilepsy." Then call it something like that, and include the word therapy in the title.

1. Morphine is an opiate regardless of why you take it. But it is only a pain therapy if you have pain. A diet is a diet if you eat it. Similarly an article entitled "ketogenic diet", should discuss the diet regardless if the person eating it has epilepsy, whereas "ketogenic therapy" should discuss the therapeutic use.
4. The use of the word "diet" is a misleading way describe this therapy because it implies that the therapy works via metabolism, but I think there is more evidence that the benefit is via celluar metabolism. (The distinction is more important now than in 1921, because then consumption of a ketogenic diet was the only known way to raise ketone levels in blood stream, that has slowly changed, with the introduction of MCT oils, maybe raspberry ketones (I know nothing about them except that tons of people sell them) and most recently, the (experimental) manufacture of ketone esters.)
6. This article isn't only about epilepsy, since it mentions alzheimer's. Some promising recent alzheimer's work is with ketone esters as supplements, hence the patient diet is not necessarily a ketogenic diet, but the therapy is ketone based.

You asked

Does Wikipedia need a "place to describe ketogenic diets in general" any more than it needs a place to describe "green things" or "expensive things"?

Yes, at least because those things generally won't make you sick (or conceivably kill you). (I've seen lots of articles on ketogenic diets suggesting replacing all dietary fat with MCT oil - articles which omit any mention of essential fatty acids.)

Do people talk about ketogenic diets in general?

Yes, and the very trendy "Paleo diet" has many more people researching what a ketogenic diets are.

and I reckon mostly at the level of speculation."

Sadly, since so little is known; but more is known than if you construct something new, yourself. Just one reason for using ketogenic diets is when factors push one so close to a ketogenic diet, that many people chose to move to a known place.

So I do see separate needs for articles on ketogenic diets (which focuses on the metabolic aspects) and ketogenic (or ketone) therapy for epilepsy. — Preceding unsigned comment added by Antifesto (talk • contribs) 08:39, 17 December 2012 (UTC)[reply]

• So I do see separate needs for articles on ketogenic diets ...: And for people who eat a ketogenic diet all the time, e.g. the old-type Eskimo diet before contact with Europeans: see this diff. Anthony Appleyard (talk) 08:57, 17 December 2012 (UTC)[reply]

  The Inuit diet has less than half the amount of fat that the normal KD for epilepsy does. They are really not comparable diets, especially if considered on a year-round basis rather than comparing the year-round KD to the Inuit diet only at the depths of winter. WhatamIdoing (talk) 01:02, 24 December 2012 (UTC)[reply]

The term "ketogenic diet" is used in scholarly literature to refer to this epilepsy treatment in the vast majority of cases. Blogs, self-published books and other crap turned up by Google do not count as reliable sources. Wikipedian's are not free to invent their own names for things or to combine subjects when no scholar would dream of combining them. There is currently no mainstream clinical use for the ketogenic diet, as described on this page, for anything other than epilepsy: everything else is just research. Colin°^Talk 14:18, 17 December 2012 (UTC)[reply]

Several sections of this article are superseded now because of a good quality systematic review of randomised controlled trials of ketogenic diet in children and young people with epilepsy (I've given it the ref name of Levy). I've incorporated it into the article intro and the outcomes section, with minimal changes to existing text and deleting a discussion based on a single primary source. However, I think that the intro on outcomes also needs updating, and there is too much detailed discussion of individual studies that are not randomised controlled designs, which could be deleted. Do others agree? I didn't want to hack away if not, or if there is someone else who knows the rest of the article in more detail who could do this with care. Hildabast (talk) 16:09, 8 June 2013 (UTC)[reply]

Hi, Hildabast. Thanks for updating this. I hope to take a more detailed look over the weekend. I'll get back to you later. Colin°^Talk 17:24, 8 June 2013 (UTC)[reply]

G'day Colin! That's great to hear - thanks! Hildabast (talk) 17:57, 8 June 2013 (UTC)[reply]

I've read the full review now. I must admit I'm extremely disappointed with it:

• The statement "Diets have been used in an attempt to control epileptic seizures throughout the centuries, indeed there is a biblical reference to prayer and fasting in epilepsy (StMark 9: 14-29)" contains two falsehoods. Firstly that diet has been used to treat epilepsy "throughout the centuries". Other than some brief mentions by the ancient Greek and Roman physicians, history is pretty silent on diet until the 20th Century, at least according to Temkin's widely respected work on the subject. Second is the oft repeated assertion that Jesus says something about fasting as a treatment for epilepsy. It makes me groan when I read this. Leaving aside that some Bible manuscripts don't even mention "fasting", the statement by Jesus concerns the disciples' need for earnest prayer rather than the patient's need to abstain from food. This is important because if there was even a hint that Jesus was suggesting people with epilepsy should fast, we'd have 2000 years of such practice, rather than it being discovered by a crackpot at the start of the 20th century.
• They include four randomised trials but only one of them "Neal 2008" has any power to show efficacy as this is the only one to have a non-treatment control group. The other trials looked at induction method, or varying aspects of the diet. While the review does note this, it greatly diminishes the value of those other three trials. The review summary claims the view that there is a "beneficial effect" is "strengthened by the publication of four randomised controlled trials". This over-simplification is highly misleading, something that our article text currently repeats. Three of those trials are no better than the many other prospective trials at showing patients get better on the diet (but not necessarily due to the diet).
• Their comments about reasons for stopping therapy are over-simplified and imo over-emphasised. These are children with refractory epilepsy who have failed to respond to several medications. Their chance of any further treatment being beneficial is small, never mind the idea that they might achieve a 90% reduction or even be seizure-free. Those trying another drug would perhaps be on it for a few months and very few of them would still be on that drug after a year -- for the same reasons that all the previous drugs tried failed. So when discussing this therapy wrt treating this difficult group of patients, it would have been more helpful to explain how the drop-out rate differs from other therapy choices. They just really mention tolerability, side-effects and ineffectiveness as reasons for not continuing long-term. They don't mention that often children are weaned off the diet and retain their seizure control. They don't mention that long-term, there are concerns about growth and bones which encourage physicians to be wary of long-term usage. They don't mention that children do sometimes grow out of their epilepsy. Look at PMID 11581442 (Hemingway 2001) and compare whether "reported follow-up data for three to six years and suggested that only 10% of the participants remained on the diet as a long-term lifestyle change. The reasons for discontinuation were mainly due to ineffectiveness or the restrictive nature of the diet." is a fair report of that study's findings. I don't think so.

I'm tempted to restore the article text to something closer to what we had before. As far as the history of studies into efficacy, the Neal study is a landmark one that deserves documenting imo. I don't think it is optimal that we're using Neal 2008 as a source, per MEDRS, so I'll see if we can pick another review. I'm not sure I want to use the Levy review at all, despite the reputation that Cochrane has: it just doesn't seem to be a thorough job. Colin°^Talk 18:26, 10 June 2013 (UTC)[reply]

Wow, shame on Cochrane! Sometimes even the "gold standard" can have tarnish, I guess. This was out over a year ago, have you seen follow-up commentary on this report? Zad68 19:14, 10 June 2013 (UTC)[reply]

Thanks, Colin - I'll look soon and get back to you. I thought this was one was better than the usual Cochrane review, many of which have far more severe problems than differences in interpretation, or errors in the background section. You definitely can't assume it's good quality just because it's any particular "brand", that's for sure - maybe even especially when it's so heavily marketed as this one is. I'll look more closely at the points you raised and get back to you - as well as checking whether there's a better systematic review. Hildabast (talk) 19:20, 10 June 2013 (UTC)[reply]

I've had a chance to look at all of this properly. I'm going to comment on each point separately. Firstly, the issue of two statements about history/bible in one sentence in the background. The first part (diet used for centuries) is unreferenced, but it's actually so vague that only a tiny number of eccentric people even would need to have avoided some food or other and this would be true. I did a quick search of an ancient tradition involving diet as medicine (Ayurveda) and the first thing I found suggests this goes back a long way in Ayurveda for epilepsy - possibly even thousands of years. So the statement is right.^[1] I checked the biblical reference (now there's something I haven't done in a long while - thanks Colin!!!!), and according to the King James version anyway, the Cochrane review is right: it does mention fasting.^[2] Even if this weren't correct though, I don't think it's material to the systematic review of the intervention, because you don't have to agree with every sentence written in a source when it's cited. I'll work progressively through the comments, but I'd also like to note at the outset that someone changed what I originally wrote to something that was no longer accurate (eg it implied per protocol analysis, not intention-to-treat analysis - wording changes were made which rendered the statements no longer accurate). Hildabast (talk) 02:27, 12 June 2013 (UTC)[reply]

The next point - mixing in trials comparing types of diet or ways of introducing the diet with the one comparing to no diet is a very important point. And that relates more, I think, to how we describe it. The outcomes expressed were only using data from the trial with the comparison group not having the diet, but that wasn't clear at all in the very short wording. The review is actually mostly rather precise, and did not combine the data from these trials. But, the summary of findings table did make that mistake, and there are places in which they could/should have worded things differently. It's not perfect, but there isn't much that is perfect. I don't think this negates the value of the 3 other trials: they are just addressing different questions. The big issue though is not to give the impression that the evidence is stronger than it is, and I totally agree with you on that. That said, the previous version (just using the Neal trial) also had some of the same problems: it didn't mention adverse effects (and the adverse effects section isn't based on trial data) or the trial's weaknesses - such as the great risk of bias caused by so much missing data. I didn't try to work it out, but if a large proportion of those children who went missing actually had serious cardiovascular effects and increased epilepsy, the whole picture would change (not to say that they did,but just to make the point). Summary: I agree this requires a change of wording, but I don't think it means that the systematic review should be discarded. After all, the Wikipedia article is not itself a structured systematic review that can give readers confidence that all studies were found, whereas the search in the Cochrane review seems reasonably ok - I only found one other trial in another review, and that was again comparing diets, was small, low quality and in the 70s - so the oversight is pretty irrelevant. The Cochrane review includes a further 11 studies and I can't see that all of those were in the Wikipedia article. And there isn't a better review that I could find (I'll come back to that). Hildabast (talk) 03:29, 12 June 2013 (UTC)[reply]

The third point was the explanation of why people stopped being over-simplified, and I can see what you mean. The review overall is relatively slim, and doesn't go into much detail - especially of the non-randomized studies. On the other hand, the evidence is so weak, it cannot really bear the weight of a lot of interpretation - and certainly not individual studies. Hemingway has long follow-up, but it is still not high quality evidence. Even if others might do it differently, that they took the RCTs more into consideration (and the toleration issues in the other trials are relevant here) isn't really enough to discredit their work. They have over-summarized - but less so than others have done, and at least it's a reasonable systematic coverage of studies that they're summarizing. Yes, there's much they don't mention. But that's an argument as to why this systematic review doesn't displace everything else - but not an argument to disregard it. The question here would be, is there a better one? I couldn't find one as good or better. There is the 2012 update of the NICE guideline/systematic review.^[3] It has the 1976 comparison of types of diet, but is missing the 3 that Cochrane has, and doesn't review non-randomized studies. I couldn't find a list of excluded studies or quite a few methodological basics I needed to see to assess it as a systematic review. So it's not a stronger review than Cochrane. I would have totally been prepared to reject the Cochrane one, but its main problems are in interpretation and reporting, not in itself as a systematic review gathering the evidence adequately. As systematic reviews go, the Cochrane one is reasonably ok and better than the NICE update. So I'ver changed the wording to reflect these issues: I hope I've improved it. Hildabast (talk) 03:52, 12 June 2013 (UTC)[reply]

Thanks for all the research. I can guarantee you did more research into diet & epilepsy than the authors of that review. It's just a replay of other background sections in other papers. Anyway, I disagree with diet statement and the source you found doesn't confirm it. I will accept that Temkin is weak on eastern medicine. We must be careful not to confuse something one eats as medicine and a dietary change. Also other traditions emphasise dietary change more than we do, so this raises the question of (a) whether its use in epilepsy is notable there at all, (b) whether it was done to specifically treat "epilepsy" or some other imbalance/illness in the person believed to be associated (c) whether it was actually effective. One must consider why anyone would mention the use of diet in epilepsy in history. The statement is in danger of falling into the "Ford cars are green" trap: the statement can be shown to be true in some cases, but not generally and it is not a helpful statement to make. I think it is just puffery put in to imply there's some long tradition being drawn on, when in fact this is not the case.

Wrt Jesus. Modern translations, drawing on better source material, don't include "fasting". But even if they did, Jesus was talking to the disciples and was instructing them on how devoted their prayer needed to be. There's absolutely no hint that he intended the sick boy to fast, and nobody else in 2000 years has suggested this. This meme appeared in some KD background/history paper and has been uncritically copied since. I've looked at biblical scholarship on this story and find nothing suggesting the boy should fast. This to me is not only a stupid mistake to make, but repeating in the paper it shows stupidity, and one wonders if someone is trying to get holy blessing on the therapy, such is the desperation. Look: Jesus says you should fast for epilepsy. It is rubbish. It goes to confirm WP:MEDRS disapproval of using the background section of any paper as a source: this is weak material that is not critically peer or editorially reviewed. There's plenty proper scholarship on the history of epilepsy or on the biblical text that we can draw on here, and none of them mention this.

I'll look at the other issues later. I think the other RCTs can be useful in their areas in the article (induction protocol, ketogenic ratio, Atkin's variant) but they supply no strengthening data wrt efficacy and I strongly feel we shouldn't combine them to say "four randomized controlled trials of the ketogenic diet". The fact that they are RCTs is not a significant defining factor of those trials. They looked at different things and combining them at all is an apples and oranges mistake. Which RCT has the "missing data" problem? Colin°^Talk 08:26, 12 June 2013 (UTC)[reply]

Very convincing argument about the history aspect - if you had time at some point to comment on this on that systematic review, it would be good, as it should be deleted. Yes, agree about food as medicine - I wasn't trying to be thorough, just doing a quick check if food was seen as a contributor to seizures in other medical traditions, and Ayurveda was the most obvious one to check quickly. I've actually added detail about the number of studies: ordinarily I'd agree with you, but the article is actually itself a review speaking about lots of individual studies without a perspective of how many there are, in what category, and why the ones discussed have been selected. It's not to give the review more weight, it's because it provides needed perspective on the body of evidence. They didn't combine them: there was no meta-analysis. Doing a review that covers several questions and study types is legitimate. I can see why you wouldn't want to split that review into multiple reviews - but if I were the editor, I would certainly have wanted more clarity on the separate questions in the report, and doing that might certainly affect the way people interpret the results: it's another important that readers of the review need to know, I agree. Not discussing attrition bias is a worse problem, in my opinion. It was Neal 2008 - and Neal et al don't discuss the bias it brings, just that the rate of attrition was higher than in other studies. 32 missing out of 145 is 22% attrition, and that's noteworthy (see Biester and Dumville). There's even more bias in non-randomized studies, but it does mean that overall, there just isn't strong evidence about this diet, so all the estimates are fuzzy and all it would take is one good really strong trial to tip the balance. Hildabast (talk) 11:06, 12 June 2013 (UTC)[reply]