The Exoneration of Seed Oils Part 2

A Bayesian View of Seed Oil Consumption and Health

Seed oils. When it comes to controversial nutrition topics, a debate on this issue pretty much guarantees a heated exchange. In one camp, we have the vocal naysayers whose life mission is to warn of the perilous dangers of these supposedly noxious agents. Their rivals – including the majority of academics, dietitians, nutritionists, food policy-makers and the food industry – state that, not only are there no health concerns but, relative to intakes of other fats, seed oils may offer health benefits. The first group’s claim is that these substances – oils extracted from seeds that include sunflower, safflower, corn and rapeseed (canola) – are being added to food as a cheap way of making it tasty and that we should be cautious of the nefarious motives of the food industry because, as they correctly point out, we’ve only been consuming these ingredients for less than a century. Some also state that the human body is designed to process seed oils in large quantities in their concentrated form.

With so much noise surrounding this oddly polarising topic, I sought to glean other perspectives and review the evidence, and in July 2023, I published The Exoneration of Seed Oils: A Review of Misinformation and the Evidence [1]. Initially, I received little challenge, leading me to assume – somewhat arrogantly, in hindsight – that my job was done, at least for the time being. That was until December, when I uploaded a short video in which I expressed my views on the seed oil controversy. Of the numerous criticisms I received, many were predictably aggressive, conspiratorial and counter-scientific, and it’s likely that few of those who attacked me bothered to read my article to see how I arrived at my claims. Encouragingly, however, a couple of respondents offered more constructive critiques, and one individual gave me pause for thought in a well-rounded challenge. In Rebuttal to: The Exoneration of Seed Oils, Jeff Webster, Co-Founder of Hunter & Gather, compiled a second-person rebuttal to my article [2].

Now, it’s generally wise – if you want a quiet life, that is – to ignore social media criticisms, especially ad hominem attacks. But, when presented with a view that runs contrary to your beliefs, it can be useful to peruse the ideas and delve into the author’s claim. As Webster was sufficiently motivated to unpick my article, it’s likely that he presented an opinion to be taken seriously, possibly something that might change my view. Conversely, a failure to be convinced by his challenge would serve to further reinforce my position. Either way, it’s a win: I either learn something new or my prior position is validated. Maybe both. So, here, in Part 2, I’ll comment on Webster’s rebuttal, along with discussing other research involving seed oils and non-communicable disease (NCD) risk. And I’ll do this while reflecting on my initial position.

My primary motivation for writing and posting articles is to force myself to rigorously review relevant literature, consolidate my thoughts and open myself up to challenge, with the objective of gaining a greater-than-average understanding on a topic. To further explore the seed oil controversy, it seems wise that a review of my prior perspective should be Bayesian. As described in my article Bayesian Nutrition, Bayesian reasoning is a method for evaluating the degree of certainty in a proposition in order to work out how likely something is, taking into account what we already know, along with any new evidence [3]. Although imperfect, it mandates updating one’s beliefs to be better aligned with the available evidence. This means I should reassess my degree of confidence in the assertion that seed oils present no health concerns based on new evidence with which I’m confronted. However, biases are impossible to completely mitigate. I have an obvious one: I’m co-founder and Chief Sustainable Nutrition Officer of Huel. As several Huel products include sunflower or rapeseed oil in order to provide essential and beneficial fatty acids, I’m motivated to defend my position. Moreover, aside from the fact that I’ve a financial interest, I’m emotionally attached: I’m responsible for the decision to include seed oils. Even if I were to no longer financially benefit from Huel, it would still be hard to remove a bias stemming from a decision I’ve already confidently made. Biases work in all directions, however: proponents of the seed-oils-are-dangerous argument are equally committed. Jeff Webster’s company Hunter & Gather, for instance, sells a range of food products and supplements “all free from refined sugar, grains, and inflammatory seed/veg oils” [4], a bias that he admirably acknowledges in a disclaimer in his article. However, like me, he is not only financially committed but also emotionally motivated to defend his position. All we can do is be aware of our biases, try to mitigate them, aim to be transparent and defer to the evidence.

The Story So Far

If you’re interested in the seed oil debate, you might find it useful to read both The Exoneration of Seed Oils (which, from now on, I’ll refer to as “Part 1”) and Jeff Webster’s Rebuttal to provide contrasting viewpoints. In Part 1, I attempt to demonstrate that consuming seed oils provides no health hazard and possibly even offers some benefits relative to other fats. I also point out that the connection between seed oil consumption and adverse health might be due to the fact that a large amount of them are present in heavily processed junk foods where the oils will have been subject to oxidation through certain processes. The article also includes my musings on the origins of the negativity surrounding seed oils and concerns about the unscientific rationale made by those who claim they’re hazardous [5].

In his rebuttal, Jeff Webster challenges some of the research I present and submits other citations in support of his position. His assertion is that seed oils are not exonerated from blame in NCD, as was inferred from Part 1, and, at the very least, require further enquiry within the scientific literature [6].

Asking the Right Questions

Some of the unease regarding seed oil consumption lays caution to the industrial processes involved in their production. Other concerns stem from the fact that food-grade seed oils are recent phenomena and humans have not evolved to digest/process/utilise them. While both are interesting objections – and issues worth addressing – they aren’t relevant to this article, because either the research shows a link between seed oil consumption and NCD or it doesn’t.

What we should examine is the challenge made by the anti-seed-oilists regarding seed oil consumption and the incidence of NCD. When looking at the epidemiology, there does appear to be an association between seed oil consumption and worse disease outcomes. Over the past few decades, vegetable oil consumption in the West has gone up, and, at the same time, incidences of diseases linked to lifestyle, such as cardiovascular disease (CVD), metabolic disease and cancer, have also risen. So, as highlighted in Part 1, there is some validity to the claim that consumption of seed oils is linked with CVD, inflammation, metabolic disease and some cancers. While it’s easy to suggest a connection between these two variables, most of us are well aware that correlation doesn’t mean causation. Instead, when examining the research, we should compare outcomes and mechanisms. Being rich in mono- and polyunsaturated fats, seed oils are prone to oxidation*. Vegetable oils that have been repeatedly heated and cooled may well have undergone chemical alterations making it likely that they have a negative impact on health. Crucially, however, this link applies to the regular consumption of junk food in general. Moreover, as oils are calorie-dense and common ingredients in junk foods, they will have contributed to increased calorie consumption in Western diets over the past few decades.

Something I didn’t sufficiently emphasise is that seed oils are not innocuous in their contribution to negative health outcomes. Three reasons might explain this. Firstly, the fats they contain are prone to oxidation, and their presence in many foods involves them being heated-cooled-reheated several times, which damages the fatty acids and may contribute to poor health. It seems that the anti-seed-oils camp has taken this information, deduced that “seed oils are toxic” and gone on to scaremonger their audience. This is the issue when evaluating the evidence and drawing conclusions from it: correlation epidemiology can be problematic and is why we should be cautious of health policy that’s based on it. The second reason is that, due to their prevalence in highly palatable convenience foods, they can contribute to energy toxicity through calorie overconsumption. And thirdly, seed oil consumption can be a marker for overall poor diet quality: diets habitually high in damaged oils are also typically high in fat, sugar and salt, and low in fibre and micronutrients.

But what about oils extracted from seeds that haven’t been heated, cooled and reheated several times? Are they bad for you? What happens when saturated fats are substituted with polyunsaturates on a one-to-one ratio? The key question is, “Are seed oils uniquely harmful?” (When we revisit this question, to avoid the need for clarification, let's interpret it as referring to undamaged oils. Additionally, we’ll consider it without factoring in their contribution to calorie toxicity and association with poor diet quality).

Seed Oil Consumption and Disease Risk

In his rebuttal, Jeff Webster claims that there “is plenty of evidence linking an excess ratio of omega-6 to omega-3 with dysfunction”. I don’t disagree with this, and it’s something I covered in an earlier article, The Omega Ratio: Useful or an Unnecessary Hassle?, which I referenced in Part 1. Webster’s argument infers that it’s the ratio of dietary omega-6s to -3s that’s the crucial issue, implying that simply by reducing our intake of omega-6s, the ratio would improve and, well, job done. However, as I concluded in The Omega Ratio, it’s the underconsumption of omega-3s that needs addressing. Ensuring an adequate consumption of omega-3-rich foods is way more important than worrying about one’s intake relative to omega-6s [7]; i.e. it’s not the ratio that’s important, but the fact that most of us aren’t getting enough omega-3.

Looking at the two principal omega-6 fatty acids in seed oils – linoleic acid (LA) and arachidonic acid (AA), which are essential and semi-essential, respectively – Webster points to a study showing that LA oxidises in LDL cholesterol, creating, he claims, a dangerous version of it [8]. The implication is that LA is the most abundant fatty acid in LDLs, the oxidation of LDL transforms LA into oxidised metabolites, which are atherosclerotic, the content of these oxidised metabolites increases with age and, because of this, there’s more chance of LA accumulating in LDLs. This means that with greater age, more LDL oxidation occurs, which is a precondition for atherosclerosis, and so, by reducing LA, the process is slowed down. This seems quite a chain of events to arrive at the conclusion that simply having less dietary LA will mitigate atherosclerosis by this process, particularly as it’s based on a single 1998 study.

Conversely, other research points to the consumption of LA and AA being correlated with reduced CVD risk [9]. For example, studies exploring diabetes and all-cause mortality have indicated LA to be beneficial and AA neutral [10]. From viewing the literature, it’s simply not the case that consumption of these fatty acids in the form of vegetable oils is dangerous.

Webster claims that my article is “centred around four core very old papers”. Two things here. Firstly, I don’t see why the fact that they’re “very old” is relevant, since they were key to the 2017 review by The American Heart Association Core Trials Review that I discussed in Part 1. This paper concluded: “Taking into consideration the totality of the scientific evidence, satisfying rigorous criteria for causality, we conclude strongly that lowering intake of saturated fat and replacing it with unsaturated fats, especially polyunsaturated fats, will lower the incidence of CVD” [11]. The review cited the four papers as the most robust evidence available on the topic on the basis of the quality of study design, execution and adherence, the reasons for which I detailed. Secondly, while my article did discuss the 2017 review, 29 other research studies were also referenced. It’s therefore a misrepresentation to claim that my article is “centred around” these four papers, particularly because I only discussed one of the four – the 1969 LA Veteran Study [12]. His claim implies that the crux of my argument relied on these four papers, which is simply not true. For instance, Part 1 also included some similarly robust but more recent studies, such as the 2010 systematic review and meta-analysis of six randomised controlled trials (RCTs)** that concluded that increased polyunsaturated fatty acid (PUFA) consumption in place of saturated fatty acid (SFA) significantly reduced rates of coronary heart disease (CHD)♰ [13], and a 2018 meta-analysis of 54 controlled studies that showed all types of vegetable oil to be more effective at reducing LDLs when compared to butter [14].

As well as the evidence presented in Part 1, there’s considerably more research to further support my assertions. When looking at the effects of vegetable oils and inflammation, the outcomes are either neutral or positive [15]. Maybe the fact that some of the fatty acids in seed oils can be oxidised means that their consumption has a negative effect on human health and inflammation, but this is more than compensated for by their positive effects, because, overall, when inflammation is measured in the RCTs, we see that inflammation is either the same or lower when saturates are replaced with unsaturated fats.

What about heart disease? A 2021 meta-analysis demonstrated that replacing SFAs with PUFAs or carbohydrates was associated with reduced CVD incidence in adults with type 2 diabetes [16]. Supporting evidence from research in the general population increases our confidence in these findings, showing either a similar or lower risk of CHD when substituting SFA with PUFA. For instance, a 2014 systematic review and meta-analysis of prospective cohort studies showed LA intake to be “inversely associated with CHD risk in a dose-response manner” [17]. In fact, as these are prospective cohort studies, it’s actually a much larger effect because the studies are longer in duration. Other research indicated that blood markers for CHD risk were no different when monounsaturated and polyunsaturated fats were exchanged, but that both, relative to SFA intake, were key to lowering total and LDL cholesterol [18]. Study after study clearly indicates that replacing saturated fat with polyunsaturated fat significantly reduces CHD risk.

Now, let’s look at research that explores other metabolic health markers. One RCT showed that a polyunsaturated-fat-enriched ketogenic diet was shown to induce a greater level of ketosis and improve insulin sensitivity without adversely affecting total and LDL cholesterol compared to a traditional saturated-fat-enriched ketogenic diet, suggesting that a polyunsaturated ketogenic diet may be superior [19]. Another suggested that replacing saturated and trans-fatty acids with unsaturated fats benefits insulin sensitivity and reduces type 2 diabetes risk markers [20]. And an RCT in which overweight or obese subjects were fed either SFAs or PUFAs for eight weeks concluded that there was very strong evidence that PUFAs consumed in place of SFAs significantly reduced liver fat [21].

But what about the risk of cancer and seed oil consumption? In his rebuttal, Webster claims that one of the Core Trial Review papers showed an “82% increase in the chance of cancer for those on the higher polyunsaturated oil diet” [22]. In subsequent private email correspondence, he stated the actual number to be nearer to 87%, as per a review of the data by the same lead authors [23]♰♰. Other research has looked at vegetable oils and cancer. A 2015 meta-analysis, for example, concluded that a higher intake of vegetable oils is not associated with a higher risk of breast cancer in humans [24], and a 2021 systematic review and meta-analysis that looked at potentially toxic elements concluded that “the consumption of vegetable oils is safe and does not pose risk to the health of consumers” [25]. However, there may well be an increased risk of cancer from the regular consumption of reused edible vegetable oils that have been repeatedly heated and cooled, i.e. in heavily processed junk food.

Consequently, seed oil consumption only seems to be a concern where the oils have been heated to very high temperatures, allowed to cool and reheated, several times. This, essentially, is what happens in most fast foods. If a large amount of your diet is based on fried foods, is it the seed oils that are causing the negative effects or is it more likely that your diet is just poor quality and that eating excessive calories is the problem? Body fat is pro-inflammatory – adipose tissue secretes cytokines, and these increase overall inflammation – so it’s more likely that it’s this that contributes to NCD risk. You could just say, “Junk food is bad for you” and leave it at that! To my knowledge, there’s no evidence that PUFA consumption, relative to SFA intake, causes negative outcomes. Granted, it’s highly likely that it’s advisable to consume more omega-3s, but there’s no need to remove (undamaged) omega-6s from your diet. Indeed, the research seems to show an improvement by replacing SFAs with PUFAs. (Note: I’m not saying saturated fats are the problem, a debate I’ll cover in a future article.)

Refuting the Trans Fat Analogy

At the risk of losing the reader by veering away from seed oils for a minute, I feel I should address Webster’s analogy with the change in health policy surrounding trans fats. Trans fats are produced when oils undergo hydrogenation, and from the 1920s until the 1990s, they were prevalent in certain foods, particularly margarines and some confectionery. Moreover, the margarines that contain them were promoted by the food industry as having health benefits. Since then, however, a plethora of evidence has revealed their consumption to be associated with increased risk of a number of diseases, in particular CVD, but also diabetes, cancer, certain allergies and even obesity [26]. Consequently, trans fats are now almost universally accepted as being hazardous to health.

Webster mentions Fred Kummerow, a nutrition scientist and biochemist, who “devoted his long life to the study of lipids, cholesterol and heart disease.” Invoking phrasing I used in my criticism of people in the anti-seed-oil camp in Part 1, Webster claims that Kummerow “epitomised ‘anti-conventional-science rhetoric’,” and, he says, “warned us about trans fats in 1957. But according to you, this would have made him an anti-establishment conspiracy theorist.”

A couple of things here. Firstly, although I’m unfamiliar with Kummerow's work and so can’t comment on how he felt about the establishment, I don’t see how my article implies what Webster suggests. Someone who disagrees with others’ views – including the mainstream narrative – and cites evidence for their position need not be conspiratorial. Sometimes, the mainstream view of the time and the science that feeds into it are simply wrong and only later revealed to be so. The second issue with Webster’s analogy is that, unlike the seed oil debate where, as demonstrated, a huge amount of research exists in the form of RCTs and meta-analyses showing that their consumption offers no hazard, no such data for trans fats existed in the mid-20th century. When researchers did start digging into the issue, the dangers were revealed. Consequently, Webster’s trans-fat analogy doesn’t stand up.

It’s worth pointing something else out, too. Even when the epidemiology linking trans-fat consumption with NCD risk was gaining momentum and feeding into policy, the position was challenged. In 1997, Samuel Shapiro, an epidemiologist, warned of a possible bias, which he referred to as “confounding by indication”. As, prior to the 1990s, public health policy had been promoting the health benefits of margarine over butter, many individuals with CVD may have been more likely to have followed official health policy and opted for margarine over butter, leading to a higher intake of trans fats. If this were true, then the identified positive association between trans fat intake and CVD risk may have been exaggerated by the people in this high-risk group making what they believed to be a positive behavioural change [27]. At the time, Shapiro questioned the theory that trans fats contribute to CVD. Of course, now we have mountains of evidence demonstrating that trans fats have dangers associated with their consumption. So, why, you may ask, do I bother mentioning this now seemingly moot issue? Because it shows that we should always question correlation epidemiology and look to RCTs when making assertions. This, I believe, is where the all-seed-oils-are-bad mob are going wrong.

A Bayesian Take

So, where are we? After reviewing the additional evidence presented in this article, along with that in Part 1 and Jeff Webster’s rebuttal, how should we answer the question “Are seed oils uniquely harmful?” when it comes to their consumption and influence on NCD risk? Looking at it through a Bayesian lens, before I read Webster’s challenge, my confidence that seed oils are not uniquely harmful was very high; let’s say my Bayesian prior – i.e. before considering new information – was 95 percent. Did my confidence waiver after reading Webster’s piece before I had time to carefully consider his arguments? Absolutely! I reckon it reduced my confidence to 85 percent. However, after poring over RCTs and meta-analyses that I hadn’t previously come across, my confidence has risen to a level considerably greater than my prior: say, to 98 percent.

Now, the more astute among you will be thinking, “But, James, those percentages are just subjective assessments. Sure, I see you’ve cited well-designed studies that contain objective data, but you’ve just arbitrarily come up with some figures!” And you’re correct. I’ve not used Bayes’ equation. But that’s the point. Bayesian reasoning used in this way is a subjective assessment, but one based on information that the reasoner has objectively reviewed. The percentages serve as a comparative illustration of where I stand on this issue in relation to previous positions.

But, has the research presented here merely been cherry-picked to back up my previous assertion in order to stop me looking silly? I have, after all, not commented on everything Webster presents. Have I sufficiently averted my own biases?

Truthfully, I don’t know. But I do feel I’ve done a pretty good job at trying to, and, by way of both articles – which include some extremely robust research – I’ve presented a very strong case to exonerate seed oils. For those who remain unconvinced and still feel that seed oils are uniquely harmful, I ask: What would it take for you to be persuaded otherwise?

Ideological?

There is an aspect of my article, however, where I should reconsider my position. Let me explain…

Webster ends his rebuttal by saying, “You end with, ‘any assertion [negatively about seed oils] is likely ideological and indicates an alignment with a particular dietary camp.’ How is your article any different?” Well, it is different because I presented well-designed peer-reviewed evidence to back up my assertion, and it’s this kind of behaviour that differentiates me from those whose thinking is ideological. So, no, in this respect, I don’t think I was being ideological.

Nevertheless, his comment got me thinking. In many cases, when people post their views on this issue that accuse me of being “paid by Bill Gates and Pfizer”, their perspective is most certainly ideological since they provide no evidence for their assertions, which, of course, I know to be bullshit. Others who’ve attacked me with comments like, “Seed oils are ultra-processed oils that were originally used alongside petrol and diesel, you can’t seriously be telling me they’re good? Either you’re paid to have this opinion or you’re no real nutritionist,” and who provide no intelligent rationale for their statements or any credible argument against my position, again, are being ideological. There do, however, appear to be some people who, like Jeff Webster, can present an evidence-based assertion of seed oil harms. In this respect, I accept that they are not ideological. I just think they’re mistaken in their opinion and that they’re either blinded by their biases or they’ve failed to sufficiently review the considerable research on this topic.

By way of both articles, I believe I’ve presented sufficient evidence that clearly demonstrates that the consumption of seed oils in their undamaged form not only presents no health hazard, but, relative to other fats, may even offer benefits. So, can I now sit back and relax in the knowledge that my job defending seed oil is done? I think not …

Postscript

Since Jeff Webster’s rebuttal, he and I have corresponded privately, primarily because I didn’t wish to misrepresent his position on a couple of points. I’m grateful for both his input and his initial challenge.

* For a brief explanation of oxidative damage to oils, see Part 1.
** For a brief explanation of randomised controlled trials (RCTs) and meta-analyses, see Part 1.
♰ More accurately, the term is coronary artery disease (CAD) due to the coronary arteries being the affected tissue. However, the commonly used term in many research papers is coronary heart disease (CHD), so I’ve used that here.
♰♰ I struggled to interpret the comparative figures from this trial. When I shared a draft comment on this point privately with Jeff, he felt that I was misrepresenting his position and so helped craft this sentence. To clarify, he informed me that there were 17 cancer deaths out of 178 total in the non-seed-oil group and 31 cancer deaths out of 174 in the seed-oil group. So, 17/178 = 9.6%; 31/174 = 17.8%: a difference nearly 86% greater than the non-seed-oil group. However, the death rate in each group as a percentage of total was 4.8% and 8.8%, respectively: an 83% difference. Although I feel like I should pass comment here, anything I say will likely be incorrect.

References:
1. Collier, J. (2023) ‘The Exoneration of Seed Oils’, Thought for Food, 20 July. Available at: https://jamescollier.substack.com/p/the-exoneration-of-seed-oils (Accessed: 8 March 2024).
2. Webster, J. (2023) ‘Rebuttal to: The Exoneration of Seed Oils’, Tools to Thrive with Jeff Webster, 18 December. Available at: https://toolstothrive.substack.com/p/rebuttal-to-the-exoneration-of-seed (Accessed: 8 March 2024).
3. Collier, J. (2024) ‘Bayesian Nutrition’, Thought for Food, 19 January. Available at: https://jamescollier.substack.com/p/bayesian-nutrition (Accessed: 8 March 2024).
4. Hunter & Gather (2024) About Us – Hunter and Gather Foods. Available at: https://hunterandgatherfoods.com/pages/about-us (Accessed: 8 March 2024).
5. ibid (1).
6. ibid (2).
7. Collier, J. (2023) ‘The Omega Ratio’, Thought for Food, 17 July. Available at: https://jamescollier.substack.com/p/the-omega-ratio-useful-or-an-unnecessary (Accessed: 8 March 2024).
8. Jira, W. et al. (1998) ‘Strong Increase in Hydroxy Fatty Acids Derived from Linoleic Acid in Human Low Density Lipoproteins of Atherosclerotic Patients’, Chemistry and Physics of Lipids, 91(1), 1-11.
9. Marklund, M. et al. (2019) ‘Biomarkers of Dietary Omega-6 Fatty Acids and Incident Cardiovascular Disease and Mortality’, Circulation, 139(21), 2422-36.
10. (a) Wu, J. H. Y. et al. (2014) ‘Circulating Omega-6 Polyunsaturated Fatty Acids and Total and Cause-Specific Mortality: The Cardiovascular Health Study’, Circulation, 130(15), 1245-53; (b) Wu, J. H. Y. et al. (2017) ‘Omega-6 Fatty Acid Biomarkers and Incident Type 2 Diabetes: Pooled Analysis of Individual-Level Data for 39 740 Adults from 20 Prospective Cohort Studies’, The Lancet: Diabetes & Endocrinology, 5(12), 965-74; (c) Virtanen, J. K. et al. (2018) ‘Serum n-6 Polyunsaturated Fatty Acids and Risk of Death: The Kuopio Ischaemic Heart Disease Risk Factor Study’, The American Journal of Clinical Nutrition, 107(3), 427-35.
11. Sacks, F. M. et al. (2017) ‘Dietary Fats and Cardiovascular Disease: A Presidential Advisory from the American Heart Association’, Circulation, 136(3), e1-e23.
12. Dayton, S. et al. (1969) ‘A Controlled Clinical Trial of a Diet High in Unsaturated Fat in Preventing Complications of Atherosclerosis’, Circulation, 40(1s2), II-1–II-63.
13. Mozaffarian, D. et al. (2010) ‘Effects on Coronary Heart Disease of Increasing Polyunsaturated Fat in Place of Saturated Fat: A Systematic Review and Meta-Analysis of Randomized Controlled Trials’, PLoS Med, 7(3), e1000252.
14. Schwingshackl, L. et al. (2018) ‘Effects of Oils and Solid Fats on Blood Lipids: A Systematic Review and Network Meta-Analysis’, Journal of Lipid Research, 59(9), 1771-82.
15. (a) Nydahl, M. et al. (1994) ‘Similar Serum Lipoprotein Cholesterol Concentrations in Healthy Subjects on Diets Enriched with Rapeseed and with Sunflower Oil’, European Journal of Clinical Nutrition, 48(2), 128-37; (b) Keogh, J. B. et al. (2005) ‘Flow-Mediated Dilatation Is Impaired by a High-Saturated Fat Diet but Not by a High-Carbohydrate Diet’, Arteriosclerosis, Thrombosis, and Vascular Biology, 25(6), 1274-9.
16. Schwab, U. et al. (2021) ‘Dietary Fat Intakes and Cardiovascular Disease Risk in Adults with Type 2 Diabetes: A Systematic Review and Meta-Analysis’, European Journal of Nutrition, 60(6), 3355-63.
17. Farvid, M. S. et al. (2014) ‘Dietary Linoleic Acid and Risk of Coronary Heart Disease: A Systematic Review and Meta-Analysis of Prospective Cohort Studies’, Circulation, 130(18), 1568-78.
18. Ginsberg, H. N. et al. (1994) ‘Effects of Increasing Dietary Polyunsaturated Fatty Acids Within the Guidelines of the AHA Step 1 Diet on Plasma Lipid and Lipoprotein Levels in Normal Males’, Arteriosclerosis and Thrombosis: A Journal of Vascular Biology, 14(6), 892-901.
19. Fuehrlein, B. S. et al. (2004) ‘Differential Metabolic Effects of Saturated Versus Polyunsaturated Fats in Ketogenic Diets’, The Journal of Clinical Endocrinology and Metabolism, 89(4), 1641-5.
20. Risérus, U. et al. (2009) ‘Dietary Fats and Prevention of Type 2 Diabetes’, Progress in Lipid Research, 48(1), 44-51.
21. Rosqvist, F. et al. (2019) ‘Overeating Saturated Fat Promotes Fatty Liver and Ceramides Compared with Polyunsaturated Fat: A Randomized Trial’, The Journal of Clinical Endocrinology and Metabolism, 104(12), 6207-19.
22. Dayton, S. et al. (1969) ‘A Controlled Clinical Trial of a Diet High in Unsaturated Fat in Preventing Complications of Atherosclerosis’, Circulation, 40(1s2), II-1–II-63.
23. Pearce, M. L. and Dayton, S. (1971) ‘Incidence of Cancer in Men on a Diet High in Polyunsaturated Fat’, The Lancet, 297(7697), 464-7.
24. Xin, Y. et al. (2015) ‘Vegetable Oil Intake and Breast Cancer Risk: A Meta-Analysis’, Asian Pacific Journal of Cancer Prevention, 16(12), 5125-35.
25. Ghane, E. T. et al. (2022) ‘Concentration of Potentially Toxic Elements in Vegetable Oils and Health Risk Assessment: A Systematic Review and Meta-Analysis’, Biological Trace Element Research, 200(1), 437-46.
26. (a) Mensink, R. P. and Katan, M. B. (1990) ‘Effect of Dietary Trans Fatty Acids on High-Density and Low-Density Lipoprotein Cholesterol Levels in Healthy Subjects’, New England Journal of Medicine, 323(7), 439-45; (b) Willett, W. C. (1993) ‘Intake of Trans Fatty Acids and Risk of Coronary Heart Disease Among Women’, The Lancet, 341(8845), 581-5; (c) Mensink, R. P. et al. (2003) ‘Effects of Dietary Fatty Acids and Carbohydrates on the Ratio of Serum Total to HDL Cholesterol and on Serum Lipids and Apolipoproteins: A Meta-Analysis of 60 Controlled Trials’, American Journal of Clinical Nutrition, 77(5), 1146-55; (d) Mozaffarian, D. et al. (2009) ‘Health Effects of Trans-Fatty Acids: Experimental and Observational Evidence’, European Journal of Clinical Nutrition, 63(S2), s5-21.
27. Shapiro, S. (1997) ‘Do Trans Fatty Acids Increase the Risk of Coronary Artery Disease? A Critique of the Epidemiologic Evidence’, American Journal of Clinical Nutrition, 66(S4), 1011-7s.