Seed oils derived from grains or legumes are no longer excluded

Reviewed and edited by Gregory Katz, MD 

Since 2010, the Whole30 elimination phase has specifically called out all forms of grains and legumes. This extended into the cooking oils derived from grains and legumes: Corn oil, peanut oil, rice bran oil, and soybean oil. If these “seed oils” were listed on the ingredient label or used in the preparation of your restaurant meal, that product or meal was out for the duration of your Whole30 elimination. 

Of note: Whole30 has never eliminated “seed oils” as a category in our program rules. You have always been able to cook with or consume canola, flaxseed, hempseed, sunflower, safflower, sesame, or grapeseed oil on the program. 

Today, you’ll still find medical doctors, dietitians, and wellness influencers touting the “toxic” and “inflammatory” health effects of seed oils (also known as “vegetable oils”). In most health and wellness spaces, it’s routinely advised that seed oils, which also include canola, grapeseed, safflower, and sunflower oils, should be avoided for health reasons. 

I, and many of Whole30’s medical advisory board members throughout the years, have also shared this opinion. In my 2012 book, It Starts With Food, I devoted a whole chapter to the “less healthy” effects of seed oils. Citing their higher relative quantity of polyunsaturated fats (PUFAs) and omega-6 fatty acids, I quite boldly stated, “Consuming seed oils promotes systemic inflammation.”

Over the years, the Whole30 has evolved in how we think about the program, and how we evaluate science, data, and clinical findings. Today, the goal of the Whole30 is to eliminate as few foods as possible while still delivering participants the best results. This philosophy serves to make the Whole30 more accessible by eliminating fewer foods. It has also encouraged us to lean more heavily on peer-reviewed, high-quality research than clinical experience and anecdotal evidence to determine the program rules.

In 2022 and 2023, we reevaluated the current scientific findings on seed oils, inflammation, and health. Specifically, we were looking to determine if consumption of PUFA/omega-6-rich oils is a risk factor for disease, independent of omega-3 fatty acid intake and other important factors such as caloric intake and diet quality.

We found a plethora of research on the subject, including umbrella and systematic reviews, meta-analyses, and human randomized controlled trials (RCTs). Much of the large-scale research reviewed works off of interventional (controlled) studies, not observational research. That is a rarity for nutritional research on this scale. (One such paper was a 2018 meta-analysis of 54 clinical trials.) This lends a much higher confidence level to our conclusions than had we only had single studies and/or observational data to review.

Based on this huge body of high-quality research, it’s clear that the evidence is quite one-sided in favor of PUFA-rich oils. We found no credible evidence that increasing PUFA consumption leads to increased inflammation or heart disease risk. In fact, there is almost universal agreement between studies that substituting PUFAs in place of saturated fat in the diet reduces heart disease risk significantly. To highlight the substantial nature of this data:

• 30 randomized clinical trials did not support that seed oils are inflammatory 
• Another 14 did not find arachidonic acid (an omega-6 polyunsaturated fatty acid) to be inflammatory—which is the entire premise of the “seed oils are unhealthy” argument 

As a result, the Whole30 Program Rules no longer eliminate any form of cooking oil, regardless of its derivation. 

Here is a summary of that research, our findings, and our conclusions.

Our review focused specifically on results from systematic reviews and meta-analyses of human randomized controlled trials (RCTs). A total of 1 umbrella review, 4 meta-analyses, and 2 systematic reviews of human RCTs were included. 

• A 2021 umbrella review of 37 guidelines, 108 systematic reviews, and 20 RCTs examined the effects of foods on LDL cholesterol levels and found high-grade evidence that foods high in omega-6 PUFAs and/or MUFAs and low in saturated fat (SF) are associated with moderate to large reductions in LDL cholesterol. 
• A 2015 systematic review of 12 clinical trials found that replacing oils high in omega-6 PUFAs with high oleic vegetable oils (i.e., oils rich in monounsaturated fat) did not result in changes in total and LDL cholesterol, triglycerides, HDL, apoA-1 (a component of HDL), or the ratio of total cholesterol to HDL cholesterol.
• A 2018 meta-analysis of 54 RCTs involving 2,065 participants compared the effects of 13 different oils and fats on blood lipids. The results showed that all oils were more effective in reducing total and LDL cholesterol compared to saturated fat-rich butter, while safflower, canola, and sunflower oil were the most effective 

Several meta-analyses have also investigated the effect of individual oils. 

• A 2022 meta-analysis found that rice bran oil significantly reduced total and LDL cholesterol and triglycerides.
• A 2020 meta-analysis showed that canola oil significantly reduced serum lipids and other markers of cardiac risk, including apo B (the main protein found in LDL) and the ratio of apo B to apo A-1 compared to other edible oils and saturated fat. Notably, canola oil did not impact markers of blood glucose, blood pressure, or inflammation, and the greatest benefit was seen at ~15% of total energy intake from canola oil.

Arachidonic acid (AA), a polyunsaturated omega-6 fatty acid, is often cited as a potential contributor to chronic inflammation.

• A 2019 meta-analysis of 14 RCTs found no adverse effects on blood lipids, platelet aggregation and blood clotting, immune function, or inflammation in adults with increased AA intake (up to at least 1000-1500 mg/d).

Linoleic acid (LA), the omega-6 polyunsaturated fatty acid found in seed oils, has also been investigated for its impact on inflammatory markers.

• A 2017 meta-analysis of 30 RCTs and 1,377 subjects who consumed LA either via diet or supplementation showed that LA intake did not increase 11 different blood inflammatory markers, including tumor necrosis factor, interleukin-6, adiponectin, and monocyte chemoattractant protein 1, or C-reactive protein.* The results were not affected regardless of the source of LA (soybean, sunflower, sesame, corn, or safflower oil). 

*One limitation of this meta-analysis was that the longest duration of any of the included trials was 6 months. However, several of the studies involved unhealthy populations, and there was still no effect on inflammatory markers. These results seem to suggest that increased LA intake does not have a meaningful effect on inflammation.

Taken together, these findings strongly suggest that PUFA-rich omega-6 oils (particularly when consumed in place of saturated fat) have either a favorable or neutral effect on markers of health, including serum lipids (LDL-cholesterol, HDL cholesterol, and triglyceride levels); cardiovascular risk markers; and inflammatory markers. Additionally, replacing PUFA intake with MUFA-rich oils, such as olive oil, does not appear to result in any additional benefits for serum lipids. These studies also serve to refute the hypothesis that cardiovascular disease is driven by increased AA or LA levels due to PUFA-rich oil consumption.

These results are consistent with recent meta-analyses of observational studies:

• A 2019 meta-analysis of 30 prospective studies involving 68,659 participants found that higher LA levels were significantly associated with lower risks of total cardiovascular disease, cardiovascular mortality, and ischemic stroke, and a non-significantly reduced risk of coronary heart disease.
• A 2020 meta-analysis of 44 observational studies found that higher LA intake was associated with a modestly lower risk of all-cause mortality, cardiovascular disease, and cancer.

In summary, this body of research provides little support for the hypothesis that AA or LA increase cardiovascular disease risk. In fact, we found credible evidence that higher LA intake was associated with a modestly lower risk of all-cause mortality, cardiovascular disease, and cancer. In addition, this is strong evidence that consumption of omega-6 oils is not a major independent driver of chronic disease, but is likely protective against chronic disease, especially when used in place of saturated fat.

So why are so many doctors, dietitians, and influencers (perhaps the loudest category here) still stuck on the idea that seed oils are inherently “toxic” and “inflammatory”? 

First, the hypothesis that PUFA-rich seed oils are a major independent driver of chronic disease appears to be limited to observational and mechanistic data, both of which have their limitations. 

Observational epidemiological studies are not experiments testing an intervention, and thus this research is not designed in a way that makes it possible to evaluate causality.  Because these studies are observing associations rather than testing hypotheses, the outcomes measured are influenced by other variables. These other variables are called confounders, which can have a profound influence on the outcomes we care about. Confounding is ubiquitous in non-controlled research, particularly in observational studies on nutrition. The many uncontrolled variables introduced into the equation can influence the results. The “healthy user” bias is a common example. Many participants who consume healthy diets also engage in other healthy lifestyle practices, such as regular exercise, which can skew the conclusions drawn from such studies. 

Observational research is often used for large-scale epidemiological studies because it is easy to perform. It is wonderful for laying the groundwork to show an association, but it is not best practice to base an entire argument off these studies. Observational studies alone are incapable of proving causality. However, when robust RCTs overwhelmingly reinforce the findings from meta-analyses of dozens of observational studies (as it does here, contradicting the idea that seed oils promote inflammation) it strengthens the hypothesis that seed oils, particularly when used in place of saturated fat, can be beneficial for health. 

In contrast, mechanistic studies occur in a highly granular and controlled lab setting, often outside of the human body. Mechanistic studies are valuable in understanding biological pathways, but their reductive nature limits their utility. They are also useful in hypothesis generation, to help us design clinical trials. However, their conclusions alone are highly unreliable. What happens in a carefully controlled lab setting is completely different from what happens in a highly complex biological system. As a result, mechanistic studies provide virtually no information about clinical outcomes or actual disease risk itself. As an example, one mechanistic study often cited in this discussion found that seed oils raise some blood markers associated with heart disease in the short term—but that gives us no information about the health outcome we care about: risk of actually developing heart disease. 

Though there are many proposed mechanisms by which seed oils have been suggested to be harmful to humans, robust RCTs contradict those findings. Therefore, the mechanistic data doesn’t capture the entirety of the biologic process of the disease. When the real world data from randomized trials contradicts the mechanistic data, it means that the model suggested by the mechanistic data is either wrong or incomplete. 

Observational and mechanistic arguments often sound compelling, but cannot be extrapolated to prove causation, and do not evaluate clinical or functional outcomes (like symptom reduction or disease progress). Controlled clinical human trials (like those we have reviewed here) are the sweet spot; they occur in the human body, and the setting is controlled to give you the most reliable evidence. High-quality randomized controlled trials (RCTs) and meta-analyses of RCT eliminate as much “noise” as possible and determine the true effects in actual humans and on measurable health outcomes.

For further details, review the hierarchy of evidence-based medicine.

In addition, we’ve observed much cherry-picking and misrepresentation of data to support claims that seed oils are bad for health. Sometimes, this is done in the name of selling a product or gaining influence and attention. (“Seed oils are toxic” is a far more provocative claim than “seed oils can, in the right context, be protective against heart disease.”) In many cases, however, the data is mis-summarized or misunderstood in good faith—as we did in 2012 in It Starts With Food.

As an example, in two often-cited studies (the Sydney Diet-Heart Study or the Minnesota Coronary Experiment) confounding factors like the presence of trans fats in the diet or the health status of participants have been overlooked or oversimplified. These studies are often cited as evidence that consumption of seed oils containing relatively high quantities of PUFA increase the risk of cardiovascular disease. However these studies (from the 1960s and ‘70s) both have serious methodological shortcomings, and have critical variables (like trans fats or the health status of participants) which can never be fully accounted for. As such, they cannot reasonably support what many have summarized them to say about the health effects of seed oils. 

We also recognize that many people have personal anecdotes about eliminating seed oils from their diets and receiving health benefits. However, individual anecdotes alone are not sufficient evidence to make nutrition claims. In addition, it is likely they have misattributed their improved health to the removal of seed oils instead of other factors (confounding variables). For example, eliminating seed oils usually means one is reducing their consumption of ultra-processed, hyperpalatable foods. For many potential reasons, reducing consumption of ultra-processed foods itself has health benefits:

• The reduction in salt, sugar, and/or fat
• The reduction in calories
• The resulting weight loss that may occur
• The replacement of ultra-processed foods with whole foods
• The increased satiety factors of whole foods

As such, it is impossible to say whether the removal of seed oils or any (or all) of those confounding variables were responsible for that individual’s health improvements. 

Finally, we recognize that sharing an opposing position to a commonly held belief is controversial, and comes with inherent risk. When we shared our updated research on MSG in 2022 (and reversed our position on the substance), we faced significant backlash from our community, healthcare professionals, and in some cases, the media. Whether it’s true or not, if the vast majority of influential people take one position, it can be hard to go against the grain. (In our view, taking on that challenge would only add to a brand or influencer’s credibility and integrity.)

We are fond of reminding people that context matters. Consuming seed oils in the context of a whole-food, nutrient-dense diet like that of the Whole30 elimination phase may provide health-promoting benefits. In the literature, this is especially clear when some saturated fats are replaced with PUFA-rich oils. Consuming seed oils in the context of a diet rich in ultra-processed foods, low in beneficial omega-3 fatty acids, and in the presence of other health-detracting lifestyle factors (like being sedentary or smoking) is very different.

• Consumption of ultra-processed foods (which often contain PUFA-rich oils) is associated with higher mortality rates in the general population. This is likely due to multiple factors, including high levels of calories, sodium, and sugar. We do not recommend over-consuming seed oils found in ultra-processed foods. 
• Inadequate consumption of omega-3 fatty acids (relative to total fatty acids) may also increase your risk of cardiovascular disease and mortality. If you include PUFA-rich seed oils in your diet, you should also conscientiously include foods rich in omega-3 fatty acids, like cold-water fatty fish, and/or an EPA/DHA supplement. 
• In addition, the repeated heating of PUFA-rich oils, or use of PUFA-rich oils for high-heat cooking (like deep frying), promotes oxidation (damage) and degrades the oil’s natural antioxidant protections. 

Altogether, these findings suggest that PUFA-rich oils should be: 

• Consumed in or along with whole foods, rather than ultra-processed foods 
• Consumed in conjunction with adequate amounts of omega-3 fatty acids
• Cooked at lower temperatures, and not reheated 

In addition, though most of the RCTs referenced here looked at replacing saturated fats with PUFA-rich seed oils, that does not mean you have to completely eliminate saturated fats in your whole foods-based diet. Feel free to include a variety of fats and cooking oils in your diet (including ghee or coconut oil) while following the guidelines above.

Finally, you don’t have to consume seed oils to enjoy a health-promoting diet. If you have the desire, access, and financial privilege to use oils lower in PUFA and higher in heart-healthy monounsaturated fats (like extra-virgin olive oil, avocado oil, or cultured/fermented oils), please do so! However, contrary to what your favorite wellness influencer might tell you, you don’t have to fear the canola oil used in your restaurant meal, or the sunflower oil found in your salad dressing. 

More important, for those who don’t have the same privileges, rest assured that cooking your way through the Whole30 using canola, rice bran, or sunflower oil (and our best practices) is both perfectly compatible with the program and health-promoting, according to science.

1. Raatz SK, Conrad Z, Jahns L. Trends in linoleic acid intake in the United States adult population: NHANES 1999-2014. Prostaglandins Leukot Essent Fatty Acids. 2018 Jun;133:23–8.
2. Guyenet SJ, Carlson SE. Increase in adipose tissue linoleic acid of US adults in the last half century. Adv Nutr. 2015 Nov 13;6(6):660–4.
3. Dayton S, Hashimoto S, Dixon W, Pearce ML. Composition of lipids in human serum and adipose tissue during prolonged feeding of a diet high in unsaturated fat. J Lipid Res. 1966 Jan;7(1):103–11.
4. DiNicolantonio JJ, O’Keefe JH. Omega-6 vegetable oils as a driver of coronary heart disease: the oxidized linoleic acid hypothesis. Open Heart. 2018 Sep 26;5(2):e000898.
5. Ramsden CE, Hibbeln JR, Majchrzak SF, Davis JM. n-6 fatty acid-specific and mixed polyunsaturate dietary interventions have different effects on CHD risk: a meta-analysis of randomised controlled trials. Br J Nutr. 2010 Dec;104(11):1586–600.
6. Schoeneck M, Iggman D. The effects of foods on LDL cholesterol levels: A systematic review of the accumulated evidence from systematic reviews and meta-analyses of randomized controlled trials. Nutr Metab Cardiovasc Dis. 2021 May 6;31(5):1325–38.
7. Schwingshackl L, Bogensberger B, Benčič A, Knüppel S, Boeing H, Hoffmann G. Effects of oils and solid fats on blood lipids: a systematic review and network meta-analysis. J Lipid Res. 2018 Sep;59(9):1771–82.
8. Huth PJ, Fulgoni VL, Larson BT. A systematic review of high-oleic vegetable oil substitutions for other fats and oils on cardiovascular disease risk factors: implications for novel high-oleic soybean oils. Adv Nutr. 2015 Nov 13;6(6):674–93.
9. Pourrajab B, Sohouli MH, Amirinejad A, Fatahi S, Găman M-A, Shidfar F. The impact of rice bran oil consumption on the serum lipid profile in adults: a systematic review and meta-analysis of randomized controlled trials. Crit Rev Food Sci Nutr. 2022;62(22):6005–15.
10. Amiri M, Raeisi-Dehkordi H, Sarrafzadegan N, Forbes SC, Salehi-Abargouei A. The effects of Canola oil on cardiovascular risk factors: A systematic review and meta-analysis with dose-response analysis of controlled clinical trials. Nutr Metab Cardiovasc Dis. 2020 Nov 27;30(12):2133–45.
11. Calder PC, Campoy C, Eilander A, Fleith M, Forsyth S, Larsson P-O, et al. A systematic review of the effects of increasing arachidonic acid intake on PUFA status, metabolism and health-related outcomes in humans. Br J Nutr. 2019 Jun;121(11):1201–14.
12. Su H, Liu R, Chang M, Huang J, Wang X. Dietary linoleic acid intake and blood inflammatory markers: a systematic review and meta-analysis of randomized controlled trials. Food Funct. 2017 Sep 20;8(9):3091–103.
13. Marklund M, Wu JHY, Imamura F, Del Gobbo LC, Fretts A, de Goede J, et al. Biomarkers of Dietary Omega-6 Fatty Acids and Incident Cardiovascular Disease and Mortality. Circulation. 2019 May 21;139(21):2422–36.
14. Li J, Guasch-Ferré M, Li Y, Hu FB. Dietary intake and biomarkers of linoleic acid and mortality: systematic review and meta-analysis of prospective cohort studies. Am J Clin Nutr. 2020 Jul 1;112(1):150–67.
15. Blanco-Rojo R, Sandoval-Insausti H, López-Garcia E, Graciani A, Ordovás JM, Banegas JR, et al. Consumption of Ultra-Processed Foods and Mortality: A National Prospective Cohort in Spain. Mayo Clin Proc. 2019 Nov;94(11):2178–88.
16. Handakas E, Chang K, Khandpur N, Vamos EP, Millett C, Sassi F, et al. Metabolic profiles of ultra-processed food consumption and their role in obesity risk in British children. Clin Nutr. 2022 Nov;41(11):2537–48.
17. Khan SU, Lone AN, Khan MS, Virani SS, Blumenthal RS, Nasir K, et al. Effect of omega-3 fatty acids on cardiovascular outcomes: A systematic review and meta-analysis. EClinicalMedicine. 2021 Aug;38:100997.
18. 18. Szabo Z, Marosvölgyi T, Szabo E, Koczka V, Verzar Z, Figler M, et al. Effects of Repeated Heating on Fatty Acid Composition of Plant-Based Cooking Oils. Foods. 2022 Jan 12;11(2).