Talk:Saturated fat

NPOV dispute - Research showing no link between saturated fat and CV risk is missing from this article

This paper was published September, 2022. It unfortunately does not fit popular opinion despite utilizing scientific rigor. I assume that's the reason for its exclusion. https://academic.oup.com/eurjpc/article-abstract/29/18/2312/6691821

It came to the same conclusion this meta-analysis did in 2014: https://www.acpjournals.org/doi/10.7326/M13-1788?articleid=1846638

Which came to the same conclusion this meta-analysis did in 2010: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2824152/

(The point is to illustrate that this has been an ongoing debate for at least 12 years.)

Apparently the science is not so clear cut on the relationship between saturated fat and CV risk, but that's not the impression someone reading this article would get. Are we going to present all the scientific data, or are we taking sides? Agenda or no? 2600:4040:5012:7500:3000:9EF2:F8B5:125A (talk) 13:29, 26 March 2023 (UTC)

2014 is old, but the more recent paper might be worth mentioning. Or is it just an outlier? We cite WHO stuff which is more recent. Bon courage (talk) 13:56, 26 March 2023 (UTC)

The best sources on this are the World Health Organization (2023) [1], American Heart Association [2] (2017), UK Scientific Advisory Committee on Nutrition (2019), [3] and Cochrane review (2020) [4]. In total they have looked at hundreds of studies. The scientific consensus is clear on this, high saturated fat intake does increase CVD risk. Psychologist Guy (talk) 16:15, 26 March 2023 (UTC)

We have about 60 years research on this, it has been known for a long time since the research of D. Mark Hegsted and colleagues that palmitic acid increases LDL-c and total cholesterol. The debate has continued in recent years about specific saturated fatty acids on CVD risk. Long-chain saturated fatty acids increase CVD risk (palmitic and myristic acid for example) but stearic acid a medium-chain saturated fatty acid is neutral or may lower LDL. It's probably worth mentioning this on the article at some point. The World Health Organization have discussed it. Psychologist Guy (talk) 16:38, 26 March 2023 (UTC)

Yeah, Cochrane is a always a great source. Let’s see what it actually said:

“ We found little or no effect of reducing saturated fat on all‐cause mortality (RR 0.96; 95% CI 0.90 to 1.03; 11 trials, 55,858 participants) or cardiovascular mortality (RR 0.95; 95% CI 0.80 to 1.12, 10 trials, 53,421 participants), both with GRADE moderate‐quality evidence.

There was little or no effect of reducing saturated fats on non‐fatal myocardial infarction (RR 0.97, 95% CI 0.87 to 1.07) or CHD mortality (RR 0.97, 95% CI 0.82 to 1.16, both low‐quality evidence), but effects on total (fatal or non‐fatal) myocardial infarction, stroke and CHD events (fatal or non‐fatal) were all unclear as the evidence was of very low quality. There was little or no effect on cancer mortality, cancer diagnoses, diabetes diagnosis, HDL cholesterol, serum triglycerides or blood pressure, and small reductions in weight, serum total cholesterol, LDL cholesterol and BMI. There was no evidence of harmful effects of reducing saturated fat intakes.” Antisoapbox (talk) 21:41, 9 June 2023 (UTC)

Hopefully you have read that review, rather than just the abstract. Firstly, the trials in most cases were only two years long and they were on "participants at varying levels of risk of cardiovascular disease, men and women, with mean ages from 46 to 66 years". But you deliberately left out "There is a large body of evidence assessing effects of reducing saturated fat for at least two years. These studies provide moderate‐quality evidence that reducing saturated fat reduces our risk of cardiovascular disease" and "In this review, saturated fat reduction had little or no effect on all‐cause or cardiovascular mortality but did appear to reduce the risk of cardiovascular events by 17%". The conclusion is that reduction of saturated fat reduced risk of cardiovascular events in the participants that had or were at risk of cardiovascular disease by 17%. If you have any Cochrane reviews telling us the opposite that increased saturated fat consumption reduces cardiovascular events, let us know but no such finding has ever been published because saturated fat clearly increases CVD risk and yes it takes decades, not two years but those trials have shown us reducing CVD for around two years can reduce CVD events. Psychologist Guy (talk) 23:48, 9 June 2023 (UTC)

Here is a major study: [5]. The conclusion is: "In this cohort, substituting omega 6 linoleic acid for saturated fat did not provide the intended benefits, but increased all cause morality, cardiovascular death, and death from coronary heart disease." A recent meta-analysis -- included in the provided link -- addresses more recently raised concerns, and reaffirms the original result. Therefore exclusion of this result is an NPOV issue.

With every food investigated in nutrition there will always be an outlier study. There is no point in cherry picking 1 study or meta-analysis when every other meta-analysis and review says the opposite as you are ignoring consensus. The study you are quoting is from 2013. We since have many recent reviews published between 2014 and 2023 which conclude linoleic acid consumption lowers risk of CVD and mortality [6], [7], [8], [9]. Psychologist Guy (talk) 21:40, 6 June 2023 (UTC)

It is undeniable that the wikipedia article on saturated fat violates NPOV.

"Available evidence from adequately controlled randomised controlled trials suggest replacing SFA with mostly n-6 PUFA is unlikely to reduce CHD events, CHD mortality or total mortality. The suggestion of benefits reported in earlier meta-analyses is due to the inclusion of inadequately controlled trials. These findings have implications for current dietary recommendations.".

Source: doi.org/10.1186/s12937-017-0254-5 sbelknap (talk) 20:59, 9 September 2023 (UTC)

You already cited that paper before on the red meat talk-page. It's a bit silly to be talking about CHD mortality or total mortality from randomized controlled trials that are around a year and a half long. There is no long-term clinical data. Of course there will be no signifcant effect here. Total mortality is not something that can be studied within a year from a couple of short-term trials. The best evidence that we have for this are epidemiological studies that look at decades of data and show consistent results. Psychologist Guy (talk) 20:53, 10 September 2023 (UTC)

You have misstated the findings of the Hamley meta-analysis. Perhaps you have not read it. Is this one behind a paywall for you?

If you have access to the article, look at Figure 6.

Poor-quality RCTs show that switching satfat → PUFA *reduces* CHD mortality while the high-quality RCTs show switching satfat → PUFA *does not reduce* CHD mortality.

I propose that the findings of Hamley be included in this wikipedia article on Saturated Fat. sbelknap (talk) 22:21, 10 September 2023 (UTC)

Most controlled trials will not give us good results for CHD mortality or total mortality because nearly all of the controlled trials are short-term. Hopefully you realise this. We need long-term data, that is why we rely on epidemiological studies here.

Many of these short-term trials that the author you cite mentions are between 1 and 2 years. In a human lifetime that is a short time. It will not give us significant data about mortality. The paper you are talking about by Steven Hamley is online in full [10]. The review he is criticizing is the Hooper review [11]. The Hooper review reported that there was little or no effect of reducing saturated fat on all‐cause mortality or cardiovascular mortality. The reduction was in cardiovascular events. It is not our purpose here to be doing original research like Steven Hamley has been doing. The Hooper review is a high-quality source and is cited on the article. There is no reason to cite a weak source like Steven Hamley. If like Hamley you believe certain controlled trials are "inadequate" that is up to you, but this type of original research will not be put onto the article. It is not up to us to question reliable reviews here, we just report what they say. The Hooper review is cited on the article. Psychologist Guy (talk) 00:22, 11 September 2023 (UTC)

I've published two meta-analyses. I'm familiar with how this is done.

Hamley's meta-analysis is a separate analysis and is not a commentary on the Hooper review. He does critique prior meta-analyses, including Hooper. This is a routine part of a meta-analysis & systematic review. Hamley does a more comprehensive analysis than Hooper, as he studies how different types of studies compare.

In the methods section of Steven Hamley's meta-analysis, he states, "I followed the PRISMA (www.prisma-statement.org) guidelines [52] throughout the design, implementation, analysis, and reporting of this meta-analysis."

Hamley does follow the PRISMA guidelines. Hooper does not follow the PRISMA guidelines. For these and other reasons, the Hamley review is of higher quality than the Hooper review.

This article could be improved by including the points made by Hamley. sbelknap (talk) 04:02, 11 September 2023 (UTC)

Ramsen and Hamley both deal with the question whether SFA should be replaced primarily with omega 6. Ramsen concludes that

"RCT that substituted mixed n-3/n-6 PUFA in place of TFA and SFA reduced CHD risk. By contrast, n-6 specific PUFA interventions tended to increase CHD risk.".

Hamley concludes that

"By contrast, n-6 specific PUFA interventions tended to increase CHD risk."

Both authors are looking for the best way to replace SFA and conclude that a replacement with primarily omega 6 does not reduce risk. Based on these two studies it seems that replacing SFA with a mix of omega 3 and omega 6 is the best strategy to reduce CVD risk. CarlFromVienna (talk) 09:26, 11 September 2023 (UTC)

You are correct that Hamley & Ramsen have different conclusions. Thanks for sharing your original research. As you know, editors are not to cite their own original research for content in wikipedia articles. Neither Hamley nor Ramsen draw the conclusion you state, that "replacing SFA with a mix of omega 3 and omega 6 is the best strategy to reduce CVD risk."

The way ahead with this article is to cite Hamley's findings. sbelknap (talk) 11:40, 11 September 2023 (UTC)

Hamley seems to be using some subjective criteria about what he classifies as a "adequately controlled trial" and the ones which are "inadequate" in his view. The adequate trials all seem to support his POV. For example this 1965 trial that used corn oil [12] was using high-dose corn oil in supplement form. The conclusion of this trial is that there was no benefit. This is not surprising. The subjects were consuming corn oil capsules, not a healthy diet. This is in contrast to another trial he cited that involved a plasma-cholesterol-lowering diet that included soybean oil, fish and vegetables [13], the conclusion was "the cholesterol-lowering diet reduced the incidence of total CHD". Yet Hamley excludes this trial as "inadequate". Why would we cite an article by Hamley that does original research?

I would agree that replacing SFA with a mix of omega 3 and omega 6 is a good strategy to reduce CVD risk. The scientific consensus is telling people to consume a mixture of n-3/n-6 PUFA in place of SFA. The idea to include just n-6 or "mostly" n-6 is not the consensus view. Hamley seems to be misrepresenting scientific consensus. The consensus view is very clear, we do not need to add one opinion piece from Hamley challenging it. See WP:Fringe. Psychologist Guy (talk) 11:58, 11 September 2023 (UTC)

Here, you misstate Hamley's methods, describing his criterion for "adequately-controlled" as being subjective. You also mischaracterize Hamley's article as "an opinion piece." Hamley is conducting and reporting meta-research and review in his article. If one follows his process, one obtains the same meta-research result. Perhaps you are unfamiliar with PRISMA. The failure of the Cochrane group to distinguish RCT quality is a deviation from PRISMA, which perhaps accounts for why the Cochrane meta-analysis does not claim adherence to PRISMA.

In the methods section, Hamley describes the criterion he used:

"Clinical trials were identified from earlier meta-analyses. Relevant trials were categorised as ‘adequately controlled’ or ‘inadequately controlled’ depending on whether there were substantial dietary or non-dietary differences between the experimental and control groups that were not related to SFA or mostly n-6 PUFA intake, then were subject to different subgroup analyses."

Hamley excludes the Oslo Diet Heart study because of a dietary difference between the experimental and control groups that was unrelated to SFA or (mostly) ω-6 PUFA intake—trans fatty acids: that is, in the Oslo Diet Heart study, "only the experimental group (the high n-6 PUFA group) received advice to avoid major sources of industrial trans fatty acids (TFA), such as common/hard margarines, shortenings and/or hydrogenated oils." This info is presented by Leren et al in the primary source. There is nothing subjective or original about Hamley's exclusion of the Oslo Diet Heart study. In fact, this critique of the Oslo Diet Heart study is well-known.

These objections to inclusion of Hamley's meta-analysis in this wikipedia article on saturated fat are without merit. Hamley is more careful in his meta-analysis than other authors, he adheres to PRISMA & he reaches conclusions that are important and relevant here. Hamley's work is a high-quality secondary source. Inclusion of Hamley's work would improve this article. sbelknap (talk) 12:26, 11 September 2023 (UTC)

The title of this discussion is „no link between SFA and CVD“. Both studies you brought up do not deal with this question but with the question if SFA should be replaced with primarily omega 6. That is a different discussion. Do you have any study that actually deals with the question at hand and supports that there is no link between SFA and CVD? If not this discussion can be closed. CarlFromVienna (talk) 15:30, 11 September 2023 (UTC)

Have you read these studies? They compare a control diet to a diet that substitutes PUFA for SatFat. The PUFA diet was high in ω-6 fat and was not pure ω-6 fat. Thus, the intervention diet reflects standard diet recommendations at the time these studies were done. It is still routine to prescribe a PUFA diet high in ω-6. That is what happens when one eats grains, legumes, and many other plant-based foods as these foods contain PUFAs high in ω-6 fats.

A reduction in satfat *requires* an increase in some other macronutrient: carbohydrate, fat, or protein. For purposes of this line of inquiry, protein & carbohydrate is kept reasonably constant, it is the fat component of diet that is manipulated. To maintain constant total fat intake, some type of dietary fat is increased and someother type of dietary fat is decreased. It is not possible to satisfy your demand for a study that *only* manipulates satfat. That's not a thing.

I propose including discussion of Hamley's meta-analysis in this wikipedia article on saturated fat. sbelknap (talk) 18:15, 11 September 2023 (UTC)

Hamley's criteria for "adequately-controlled" and "inadequately-controlled" is indeed subjective. He just made up reasons to exclude certain trials as "inadequate", when such trials were actually reliable. For example he put the LAVAT trial (Los Angeles Veterans Administration Trial) [14] on the "inadequately-controlled" list based on his bizarre reason that there was insufficient vitamin E intake in the control group. The trial is one of the most rigorously controlled trials performed on the subject (they commissioned food companies to design foods to keep the subjects blinded). It's well known that vegetable oils contain higher vitamin E than animal foods so subjects consuming vegetables oils are going to have higher Vitamin E levels. There is no way round this. It's ridiculous to exclude the trial for this reason. He has other stupid reasons for dismissing other trials.

Steven Hamley is a conspiracy theorist who delivered his paper at the Low Carb Down Under event. He spends a lot of time attacking the American Medical Association. I doubt you are going to gain a consensus here by getting other users to agree with you. I am not convinced by Hamley's methods. This is not the place to promote fringe science. He is not a reliable medical source. You should also disclose here that you have had interaction with Steven Hamley which is an obvious conflict of interest. Psychologist Guy (talk) 20:25, 11 September 2023 (UTC)

The criterion for defining adequacy of control used by Hamley is objective and adequately described in the methods section of his meta-analysis. The lack of sufficient vitamin E constitutes an instance of a "substantial dietary or non-dietary difference between the experimental and control groups that were not related to SFA or mostly n-6 PUFA intake." Assuring adequacy of all essential nutrients in the control group is essential to excluding deficiency of a nutrient as a confounding variable. The scientists who designed LAVAT erred in this way. Excluding LAVAT from his meta-analysis meets his objective exclusion criteria for an adequately-controlled trial. Hamley was not the first one to critique the excluded trials for these design flaws. Many of these critiques were made by others.

Your ad hominem ("conspiracy theorist", "fringe science") against Hamley is hard to take seriously. Hamley gives enough detail in his paper that any competent meta-researcher could duplicate his results.

I don't know Hamley, never met him. I've never had any private exchange with Hamley. I don't recall any public exchange on twitter with Hamley. As Twitter (now X) is a public forum, this is possible, of course. Regardless, this does not constitute a COI. I have no other COI.

As you fellas are anons, there is no way to independently assess your COIs.

Regardless, my proposal stands. So far, the objections raised to inclusion of Hamley in this wikipedia article range from specious to laughable. I've countered each one fully. Enough of this. Lets craft some text on the Hamley article to put in the article. sbelknap (talk) 20:53, 11 September 2023 (UTC)

The criteria Hamley uses is laughable, nobody else would use criteria like this. The LAVAT trial involved commissioning food companies to create entirely new products with vegetable oils in them to keep subjects entirely blinded. This isn't a design flaw. Show me another trial that went this far, the methodology and control was top notch. That trial is one of the most well controlled that we have on this subject. If vegetable oil is going to be used in a trial, the Vitamin E content will always be higher than that of an animal-based diet. I don't know how you would get round this, you would have to invent and alter a vegetable oil which would be very hard to do, almost impossible. If you want to believe Hamley is an objective researcher that is up to you, but it is obvious he isn't. This is very much WP:Fringe. He never published a Meta-regression. The Hooper 2020 meta-analysis includes a meta-regression and is does not employ outrageous criteria [15]. It is a much more reliable source. Nobody outside of the low-carb community takes Hamley seriously. We shouldn't be promoting him on Wikipedia. Psychologist Guy (talk) 21:21, 11 September 2023 (UTC)

To be sure, a near order-of-magnitude difference in dietary vitamin E in LAVAT is no small thing! FTFA:

Vitamin E: in LAVAT, α-tocopherol intake in the con- trol group was 9.4-fold lower than the experimental group (22.6 mg vs. 2.4 mg) [48] and only 16.0% of the current RDA (15 mg) [49]. Based on the average energy intake of the control group reported in the vitamin E paper (2400 kcal) [48] and the estimated energy intake (3150 kcal) and vitamin E (11.54 mg of α-tocopherol equivalents) per capita in the United States food supply between 1959–1968 [50], the vitamin E intake of the control group would be expected to be about 8.79 mg of α-tocopherol equivalents. This was not discussed by any of the meta-analyses.

sbelknap (talk) 21:11, 11 September 2023 (UTC)

And how do you suggest that could be modified? Vegetable oils are higher in Vitamin E than animal fats. Of course a diet high in animal fat will be lower in Vitamin E. This isn't a control issue. There is no way round this. It's ridiculous for this reason alone for Hamley to discount the trial as "inadequate". It is not a valid reason. I have read a lot of meta-analyses and I have never seen any nutritional researcher dismiss a trial as "inadequate" based on the reason cited here. The Hooper meta-analysis is a much better source. Psychologist Guy (talk) 21:44, 11 September 2023 (UTC)

Seriously?

The issue isn't that the PUFA diet had more Vitamin E. Instead, it is that the control diet was *deficient* in vitamin E at 16% of the RDA. Thus, the effect of vitamin E deficiency serves as a confounder. Vitamin E deficiency is relevant to the effect variable, as glycemia drives atherosclerosis. The deficiency of vitamin E in the control group is a serious design flaw. nutritionj.biomedcentral.com/articles/10.1186/s12937-023-00840-1

Hamley was correct in excluding LAVAT from his meta-analysis. sbelknap (talk) 22:22, 11 September 2023 (UTC)

It isn't a design flaw. It is just a consequence of a diet being based on animal fats. Like I said, there is no way round this. Any control diet with high amounts of animal fat will be low in vitamin E. This doesn't mean the trial is inadequately controlled. Even if you disagree with this, hopefully you agree that the Hamley meta-analysis does not show anything significant. Take a look at figure 2 [16]. Look where the summary point is for total CHD events for Hamley's adequately controlled trials. The risk ratio is 1.02 with a p value of 0.84. This is non significant. There is a clear noninferiority in the treatment effect. The difference between the diets is non-significant in Hamley's meta-analysis. Psychologist Guy (talk) 00:51, 12 September 2023 (UTC)

The control diet in LAVAT was the domiciliary diet at the Los Angeles Veterans Administration Center. This was not a diet "based on animal fats" — the domiciliary diet at that time was only 40% total fat, as described in the primary source articles.

The adequately-controlled trials showed that the PUFA-enriched diet did not reduce CHD events compared to placebo. That is the point. This is why citing Hamley would improve this wikipedia article. sbelknap (talk) 02:13, 12 September 2023 (UTC)

As all major nutritional societies recommend swapping SFA for PUFAs, can you point to one that has reflected Hamley in their research? If not, Hamley is only a primary source with no impact. Why then would we allow him to make an impact here if he cannot even make one in nutritional science? CarlFromVienna (talk) 06:18, 12 September 2023 (UTC)

Here on wikipedia, we are to rely on high-quality secondary sources.

Hamley is a high-quality secondary source. It has > 150 citations in the peer-reviewed literature.

According to the two United States National Academies of Science review, the Dietary Guidelines for Americans are not evidence-based.

In its most recent iteration, the Dietary Guidelines for Americans no longer presents the conclusion that saturated fat is harmful. There is a diversity of views and this article would be improved if it reflected that.

The current version of this article gives the position of the nutritional societies. It does not reflect the state of the relevant scientific literature, including Hamley. Lets fix that. sbelknap (talk) 14:34, 12 September 2023 (UTC)

Most recent Dietary Guidelines for Americans [17] says "Chapter 1 explains the importance of limiting intakes of saturated fat to support healthy dietary patterns. Staying within saturated fat limits and replacing saturated fat with unsaturated fat is of particular importance during the adult life stage" (page 102), also "For those 2 years and older, intake of saturated fat should be limited to less than 10 percent of calories per day by replacing them with unsaturated fats, particularly polyunsaturated fats." (page 44). Guideline 4 is to "Limit foods and beverages higher in added sugars, saturated fat, and sodium, and limit alcoholic beverages". Your claim that "the Dietary Guidelines for Americans no longer presents the conclusion that saturated fat is harmful", is odd because they are telling people to limit saturated fat to support health. You dislike scientific consensus I get that but this isn't the place to challenge medical consensus. These sorts of discussions are not productive (a vote may be better). The Hamley source is unlikely to be included on the article because it is unreliable. Psychologist Guy (talk) 15:55, 12 September 2023 (UTC)

Going over the history of this topic, the Hamley source has been mentioned before on Wikipedia in two places. The consensus was to not include it as it is unreliable. I have raised this issue at WikiProject Medicine to gain a new consensus. Hopefully there will be input from other editors so this can finally be resolved. Psychologist Guy (talk) 17:01, 12 September 2023 (UTC)

The way wikipedia works is that engaged editors decide how to proceed.

Now that I've addressed every objection that has been raised here among engaged editors, lets craft some text and add it to the saturated fat article.

Unless there is some other objection? sbelknap (talk) 18:15, 12 September 2023 (UTC)

I am puzzled by the claim that the Hamley article is unreliable. No engaged editor has presented any plausible reason why they hold this opinion. Each objection raised has been shown to be wrong. What is going on here?

AFAIK, this is the best meta-analysis that has been done so far on substituting PUFA for satfat. sbelknap (talk) 20:11, 12 September 2023 (UTC)

"Saturated fat and cardiovascular disease" listed at Redirects for discussion

The redirect Saturated fat and cardiovascular disease has been listed at redirects for discussion to determine whether its use and function meets the redirect guidelines. Readers of this page are welcome to comment on this redirect at Wikipedia:Redirects for discussion/Log/2023 August 28 § Saturated fat and cardiovascular disease until a consensus is reached. Mdewman6 (talk) 22:54, 28 August 2023 (UTC)

"Arterycloggingsaturatedfat" listed at Redirects for discussion

The redirect Arterycloggingsaturatedfat has been listed at redirects for discussion to determine whether its use and function meets the redirect guidelines. Readers of this page are welcome to comment on this redirect at Wikipedia:Redirects for discussion/Log/2023 August 28 § Arterycloggingsaturatedfat until a consensus is reached. Mdewman6 (talk) 22:57, 28 August 2023 (UTC)

"Cardiovascular disease and saturated fat" listed at Redirects for discussion

The redirect Cardiovascular disease and saturated fat has been listed at redirects for discussion to determine whether its use and function meets the redirect guidelines. Readers of this page are welcome to comment on this redirect at Wikipedia:Redirects for discussion/Log/2023 August 28 § Cardiovascular disease and saturated fat until a consensus is reached. Mdewman6 (talk) 23:08, 28 August 2023 (UTC)