#380 ‒ The seed oil debate: are they uniquely harmful relative to other dietary fats? | Layne Norton, Ph.D.
This episode features nutrition scientist and power athlete Layne Norton debating the evidence-based case against seed oils. He and Peter Attia explore scientific bias, historical RCTs, LDL oxidation, processing, and ancestral nutrition arguments to determine if seed oils are uniquely harmful.
Deep Dive Analysis
19 Topic Outline
Debate Format, Biases, and Evidence-Based Thinking
Four Core Arguments Against Seed Oils
The Minnesota Coronary Experiment (MCE) Analysis
Distinguishing Saturated, Mono, Polyunsaturated, and Trans Fats
Trans Fat Confounding and Limitations of MCE
The Sydney Diet Heart Study (SDHS) and Duration Problem
Evaluating RCT Evidence on Seed Oil Harm
The Rose Corn Oil Trial: A Small, Confounding Study
Trials Showing Cardiovascular Risk Reduction with PUFA
Mechanisms: Why PUFA Substitution Reduces CVD Risk
Mendelian Randomization and LDL Cholesterol's Causal Role
Lifelong LDL Exposure and Compounding Effects
Linoleic Acid, Inflammation, and Oxidized LDL
LDL Particle Number vs. Per-Particle Oxidation Risk
Industrial Processing and Solvent Extraction of Seed Oils
Evolutionary and Ancestral Diet Arguments Against Seed Oils
Balancing Processing Concerns with Metabolic Evidence
Practical Considerations for Dietary Fats and Cooking Oils
Prioritizing Interventions: Seed Oils vs. Caloric Intake/Activity
10 Key Concepts
Steelmanning
Steelmanning is the practice of attempting to present the strongest possible version of an opposing argument, even if one does not personally agree with it, to ensure a thorough and rigorous discussion.
Saturated Fat
A type of hydrocarbon where every carbon bond is 'saturated' with hydrogen atoms, meaning there are no double bonds. Saturated fats tend to be solid at room temperature and play a role in cardiovascular disease through their impact on LDL receptors and cholesterol synthesis.
Monounsaturated Fat
A fatty acid containing one double bond in its hydrocarbon chain. This double bond typically creates a 'cis' kink in the molecule, which increases the fluidity of cell membranes and lipoproteins compared to saturated fats.
Polyunsaturated Fat
A fatty acid containing multiple double bonds in its hydrocarbon chain. Like monounsaturated fats, these double bonds create kinks that enhance membrane fluidity, which is important for lipoprotein function and recognition.
Trans Fat
An unsaturated fat characterized by a 'trans' double bond, which allows its fatty acid tail to remain straight, structurally resembling a saturated fat. Trans fats are highly atherogenic and have been effectively banned due to their significant negative impact on cardiovascular health.
Reverse Causality
A phenomenon in observational studies where an observed association between two variables is actually due to the outcome causing the exposure, rather than the exposure causing the outcome. For example, very low cholesterol levels in sick individuals might be a symptom of poor health rather than a cause.
Mendelian Randomization
A powerful research method that leverages naturally occurring genetic variants, which are randomly assigned at birth, to infer causal relationships between an exposure (like LDL cholesterol levels) and an outcome (like cardiovascular disease). It effectively acts as a lifelong randomized controlled trial.
ApoB
A protein found on every non-HDL lipoprotein particle (such as LDL, VLDL, IDL). The number of ApoB particles is critical because any lipoprotein under 70 nanometers in diameter containing ApoB can penetrate the endothelium and potentially get retained in the arterial wall, initiating atherosclerosis.
Lipid Hypothesis
This hypothesis posits that non-HDL cholesterol particles, specifically those containing ApoB and under 70 nanometers, can penetrate the arterial endothelium. Once retained in the intima, these particles can be enzymatically modified, oxidize, aggregate, and trigger an inflammatory response, leading to atherosclerosis.
Naturalistic Fallacy
The logical fallacy of assuming that what is 'natural' is inherently good or desirable. In the context of diet, it's the idea that ancestral or unprocessed foods are automatically healthier, often overlooking the complexities of modern biology and disease progression.
11 Questions Answered
Podcast debates are often unhelpful because participants can make unsubstantiated claims or cite evidence out of context, and no human has the capacity to verify all literature in real-time, making it difficult to counter arguments effectively.
The four main arguments against seed oils are: negative mortality outcomes in some large randomized controlled trials (RCTs), increased LDL oxidation and inflammation, harmful byproducts from industrial processing, and a lack of evolutionary or ancestral precedent for their consumption.
The MCE's results were likely confounded by the significant inclusion of trans fats (25-40%) in the margarine given to the polyunsaturated fat group, which are known to be highly atherogenic and could have offset any potential benefits from cholesterol reduction.
The primary confounder in early RCTs, such as the MCE and Sydney Diet Heart Study, was the inclusion of high amounts of trans fats in the polyunsaturated fat interventions, as their deleterious effects were not yet understood at the time these studies were conducted.
Mendelian randomization studies provide the most powerful evidence, acting as lifelong randomized controlled trials by observing the effects of genetic variants that influence LDL levels from birth, consistently showing a strong, dose-dependent reduction in cardiovascular disease risk with lower LDL.
Mendelian randomization studies reflect lifelong exposure to lower LDL, starting from birth, whereas statin trials typically involve individuals who have already accumulated years of LDL exposure, meaning the statins are 'pumping the brakes' on an already progressing disease rather than preventing its initiation.
While polyunsaturated fats are more prone to oxidation, most LDL oxidation occurs within the arterial intima, not in the plasma, where antioxidants are less available. Studies show that increased linoleic acid intake does not necessarily lead to increased arachidonic acid production or a net increase in inflammation, and PUFA-enriched LDL is less prone to aggregation.
Industrial processing of seed oils, including hexane extraction, is unlikely to cause harm. Hexane levels in the final product are typically well below one part per million, and the amounts required to cause even mild toxicity are astronomically higher than what could be consumed, with the body also capable of clearing it.
Evolutionary arguments against seed oils are often based on the naturalistic fallacy and romanticize the past. While our ancestors didn't consume seed oils, they also had much shorter lifespans and died from other causes before cardiovascular disease could manifest. Furthermore, ancestral LDL levels (e.g., Hadza tribe) are significantly lower than what many modern diets, including some 'ancestral' ones, would produce.
If concerned about seed oils, individuals can choose other cardioprotective fats like olive or avocado oil, but the overall context of the diet is more important. Frying oils in restaurants, especially if reused or in thin layers, can accumulate negative products, but focusing on overall caloric balance and activity levels offers much greater health benefits.
The impact of seed oils on health is significantly less than that of caloric imbalance and physical inactivity. Major health levers like maintaining a healthy weight, regular exercise, and managing overall diet quality contribute far more to longevity and disease prevention than specific concerns about seed oils.
32 Actionable Insights
1. Focus on Major Health Levers
Prioritize your health efforts on high-impact behaviors like managing caloric intake and increasing physical activity, rather than disproportionately worrying about minor dietary details like specific cooking oils.
2. Boost Strength and Activity
Prioritize improving your strength, lean mass, and overall activity levels (e.g., VO2 max, grip strength), as these are enormous predictors of longevity with massive impacts on mortality risk.
3. Monitor LDL and ApoB
Actively manage your LDL cholesterol and get your ApoB measured to ensure these critical markers of cardiovascular health are under control.
4. Manage All Cardio Risks
Recognize that cardiovascular health involves multiple interconnected factors beyond just LDL, including blood pressure, cardiovascular fitness, insulin sensitivity, and inflammation, and address them holistically.
5. Adapt Beliefs to Evidence
Be open to changing your deeply held beliefs and opinions when presented with new, compelling evidence, even if it contradicts your previous understanding or academic background.
6. Seek Converging Evidence
When assessing scientific claims, look for the overall consensus and converging lines of evidence from various study types to build stronger or weaker confidence in a statement’s accuracy.
7. Prioritize Net Health Outcomes
When evaluating health interventions or dietary components, prioritize the overall net effect on hard outcomes (like disease events) rather than getting fixated on individual mechanisms that might appear negative in isolation.
8. Reduce LDL Particle Count
Prioritize reducing the overall number of LDL particles (ApoB) in your bloodstream, as this is a more significant factor in preventing atherosclerosis than the individual oxidizability of each particle.
9. Prevent LDL Retention
To reduce oxidized LDL, focus on preventing LDL particles from penetrating and being retained within the arterial intima, as this is where most oxidation occurs.
10. Declare Personal Biases
Be upfront and vocal about your personal biases and beliefs, especially when they differ from consensus or evidence, to maintain transparency and intellectual honesty.
11. Beware Belief-Driven Bias
Be aware that personal beliefs can be as powerful, if not more powerful, than financial incentives in influencing adherence to evidence, especially in information-siloed social media environments.
12. Read Original Research
To accurately understand scientific findings, always read the original research paper rather than relying on headlines or social media summaries, which often misrepresent or oversimplify results.
13. Improve Overall Diet Quality
Recognize that negative health outcomes often stem from an overall low-quality diet high in ultra-processed foods (like chips and fries), rather than solely from the presence of seed oils.
14. Reduce Seed Oils, Cut Junk
Restricting seed oils can be beneficial by proxy, as it often leads to reducing consumption of low-quality, ultra-processed foods like chips and sugary dressings, but avoid being overly restrictive.
15. Limit Saturated Fat
If you choose to avoid seed oils, ensure you are still actively reducing saturated fat in your diet by opting for leaner cuts of meat and other lower saturated fat protein sources.
16. Replace Saturated Fat Wisely
When reducing saturated fat intake, consider substituting it with fiber-dense sources of carbohydrates, as this is likely to lead to a reduction in cardiovascular disease risk.
17. Opt for MUFAs
If you are concerned about seed oils, opt for monounsaturated fats like olive oil or avocado oil as a cardioprotective alternative to saturated fats.
18. Avoid Reused Frying Oil
Be aware that frying in thin layers of oil or reusing oil repeatedly, especially in restaurant settings, can quickly lead to the accumulation of oxidized and negative byproducts.
19. Beware ‘Health Washed’ Foods
Be critical of marketing claims that frame unhealthy foods (like fries) as ‘healthier’ due to specific ingredient changes (e.g., using lard instead of seed oils), as this can lead to overconsumption based on a false perception of health.
20. Consult Research Interpreters
Recognize that interpreting scientific research requires specific skills; seek out experts who can cut through the noise and provide accurate, contextualized information.
21. Enhance Debate Rigor
When engaging in debates or discussions, pre-submit all evidence and agree on source materials upfront, only referencing pre-submitted information to ensure rigor and prevent misinterpretation.
22. Word Carefully, Avoid Misinterpretation
When communicating complex information, especially in public forums, be extremely careful with wording to prevent misinterpretation, as audiences may draw incorrect conclusions.
23. Question Data Presentation
Understand that data can be selectively presented or analyzed to support a desired conclusion, so critically evaluate how data is framed and presented.
24. Respect Study Limitations
When evaluating research, acknowledge that all studies have limitations, and use this understanding to carefully interpret findings and avoid overgeneralization.
25. Grasp Longevity’s Evolution
Recognize that evolutionary biology prioritizes reproduction and genetic propagation, not necessarily maximum individual longevity, which explains why health declines after reproductive age.
26. Prioritize Evidence-Based Net Effect
When evaluating dietary components or health interventions, base your decisions on the overall net effect supported by the best available evidence, rather than relying on naturalistic or evolutionary arguments alone.
27. Disregard Trace Chemical Fears
Do not be overly concerned about trace amounts of processing chemicals like sodium hydroxide in refined oils, as the quantities are negligible and often chemically transformed, requiring impossibly large consumption to cause harm.
28. Moderate Seed Oil Use
Do not fear using seed oils like safflower or canola oil in moderation (e.g., in salad dressings) if you prefer their taste, as the data does not suggest significant harm in such contexts.
29. Address Beyond LDL
Even if LDL is pharmacologically controlled, continue to consider other health benefits and risks associated with dietary choices, such as the impact of frying oils on other health markers.
30. Restaurateurs: Drop ‘No Seed Oils’
If you are a restaurateur, remove ’no seed oils used’ claims from your menu, as this marketing is often based on misinformation and can be misleading to consumers.
31. Check Chemical Bioaccumulation
When evaluating the safety of trace chemicals in food, consider whether they bioaccumulate in the body or if the body can clear them quickly enough to prevent negative outcomes.
32. Support Ad-Free Content
Become a premium member of Peter Attia’s podcast/website to access exclusive content and benefits, which helps fund their work without relying on paid ads.
7 Key Quotes
The scientific method is perfect. It is a perfect method, but it is done by people who are not.
Lane Norton
People think that funding or money is by far the biggest driver of people essentially like not sticking with the evidence. And I would say that in some cases, that's true. But I think that personal beliefs are actually just as powerful, if not more powerful.
Lane Norton
If you torture the data enough, it will confess what you want it to show.
Lane Norton
If an outcome exists, there's a mechanism or mechanisms to explain it. But those are like single stocks. An outcome, like a cardiovascular disease event, that is like a mutual fund.
Lane Norton
If you live long enough, everyone at some point will get some form of cardiovascular disease. It's just a time and exposure issue.
Lane Norton
We're stepping over $100 bills, picking up pennies.
Lane Norton
My own nomenclature on this is that we're majoring in the minor and minoring in the major.
Peter Attia