Only nine scientists were asked to advise on the 2025 Dietary Guidelines for Americans, and Dr. Tom Brenna was one of them. In this conversation, he pulls back the curtain on how dietary guidelines are actually made, why the process has never been the same twice, and what it was like to watch a decades-old saturated fat recommendation get carried forward despite a lack of evidence supporting it at the level of total mortality. This is the dietary guidelines conversation the internet has been missing: straight from someone who was actually in the room.
summerizer
Committee scope and public record
- The advisory committee handled narrow assigned topics and did not write the final Dietary Guidelines for Americans.
- Saturated fat was Brenna’s assigned topic in the 2025-2030 cycle, and he had also advised earlier guideline cycles.
- The committee’s scientific questions, methods, and evidence files were online for public review.
- The advisory committee delivered a circumscribed scientific document, and federal agencies drafted the final guidelines afterward.
- Brenna saw the final guideline draft only shortly before release and did not build the upside-down pyramid.
- The American Society for Nutrition backed the broad whole-food pattern while objecting to departures from the established review process.
Historical roots of the saturated-fat problem
- The butter-versus-margarine fight predates 1900 and sits behind much of the later saturated-fat story.
- Partial hydrogenation turned liquid oils into solid fats and created trans fats.
- Crisco came from partially hydrogenated cottonseed oil and became a major baking fat.
- Europe used partially hydrogenated whale oil and then partially hydrogenated fish oil in margarine through the mid-20th century.
- A 1940s margarine film on YouTube shows ingredient streams that include a drum marked "Hardened Whale."
- Many old diet trials that are used as saturated-fat evidence belong to that same era of hardened marine oils.
- One cited control-group example carried an estimated 40 to 50 grams per day of hardened marine oils.
- Those chemically altered fats were not ordinary butter or dairy fat and entered brain and retina.
Definitions, dairy, and policy contradictions
- "Saturated fat" became a code word for dairy in ordinary nutrition talk.
- Dairy evidence does not line up cleanly with the generic saturated-fat warning.
- The 10% saturated-fat cap stayed in place even while school-lunch law excluded fluid milk from the weekly saturated-fat limit.
- Whole milk in school lunch therefore sits outside the same saturated-fat budget that still governs the rest of the policy.
- Beef fat is not a pure saturated-fat block here because a large share of beef fat is monounsaturated.
- Fresh red meat and processed meat do not belong in one undifferentiated category here.
Mortality, stroke, and trial quality
- The 10% limit drives people away from nutrient-dense foods that carry zinc, selenium, and B12.
- Total mortality comes ahead of cause-specific endpoints in the evidence hierarchy used here.
- Randomized trials and prospective cohorts do not show an effect on total mortality in the 2025 review summary given here.
- What is called saturated fat shows a protective association with stroke in the prospective cohorts cited here.
- Cause-specific cardiovascular outcomes do not close the case when total mortality does not move.
- The old randomized diet trials were not double blinded, and participants knew which foods or oils they were getting.
LDL, triglycerides, and population context
- Sky-high genetically driven LDL is a real problem and sits in a different category from ordinary population shifts.
- The key open issue is whether moderate LDL reductions below 160 meaningfully change outcomes.
- Japanese population data serve as an example in which the cholesterol-heart disease pattern runs opposite the usual U.S. expectation.
- The share of the population above LDL 160 narrows to about 5% to 7%, not the larger number floated earlier.
- Elevated triglycerides above 200 reach a bigger share of the population in this exchange.
- Triglycerides here point toward carbohydrate load, insulin resistance, and low omega-3 intake.
- Fish oil lowers triglycerides here and becomes the example of a pseudo-pharmacologic intervention at very high levels.
Pregnancy, fish, DHA, and the developing brain
- Fish avoidance during pregnancy removes DHA during a period of rapid brain development.
- Children of women who eat some fish in pregnancy do better than children of women who eat none on most tests in this telling.
- The benefit curve rises and then plateaus across the intake range used here.
- The Seychelles example goes up to about 100 ounces of fish per week without a detectable mercury-harm signal.
- The people who eat the most fish in the world do not produce the iodine-deficiency picture that is contrasted with fish avoidance.
- Brain DHA accretion continues well beyond birth and runs to roughly age 20, with remodeling probably out to 25.
- Breastfeeding keeps delivering omega-3 and DHA after birth.
- DHA entered infant formula in the U.S. in 2001.
Omega-3, omega-6, and oil profile
- DHA is a 22-carbon, 6-double-bond omega-3 fatty acid.
- Omega-3 and omega-6 fatty acids use the same enzymes, so imbalance pushes one against the other.
- High linoleic acid intake acts as a metabolic suppressor of omega-3s everywhere in the body.
- The current U.S. diet sits on the high-linoleic side here.
- Some omega-6 is required, and the problem here is excess rather than zero intake.
- High-omega-6 oils create a larger metabolic demand for omega-3.
- Lower omega-6 intake makes a given omega-3 intake go farther.
- Monounsaturated replacement fats rank highest here, and high-oleic seed oils sit near olive oil in fatty-acid profile.
- Plant names no longer tell the whole story because modern oil crops were bred into multiple distinct fatty-acid profiles.
- Lower-omega-6 high-oleic oils also gained shelf-life and frying-life advantages here.
Precision nutrition, genetics, and special vulnerability
- Precision nutrition fits this field better than one number for everybody.
- Pregnancy, lactation, brain development, and later brain maintenance have different fatty-acid demands than a generic middle-aged adult target.
- A fatty-acid-desaturase-cluster insertion-deletion variant operates here as a requirement-shaping genetic signal.
- In the seAFOod work used here, 2 grams per day of EPA for 1 year cut colorectal polyps by about 50% in a certain genotype.
- Plant omega-3 sources such as flax, chia, and walnuts are not the same as long-chain omega-3 sources here.
- Vegetarian eating patterns that keep dairy hold omega-6 exposure lower than patterns that swap dairy fat for high-omega-6 oils.
- Feeding those oils to pregnant animals disrupts multiple brain measures in the offspring.
- Early human deficiency work in infants put babies on omega-6-deficient diets, produced skin lesions, and reversed those lesions when omega-6 returned.
- ADHD evidence remains unresolved here, and fish oil is not offered here as a fix.
References
- [04:21] Proposed Scientific Questions — https://www.dietaryguidelines.gov/sites/default/files/2022-07/Proposed%20Scientific%20Questions_508c_Final.pdf
- [04:33] Part C. Methodology — https://www.dietaryguidelines.gov/sites/default/files/2024-12/Part%2520C_Methodology_FINAL_508c.pdf
- [05:32] American Society for Nutrition Calls for Strong Science in National Nutrition Guidance — https://nutrition.org/american-society-for-nutrition-calls-for-strong-science-in-national-nutrition-guidance/
- [11:21] Food sources of saturated fat and cardiovascular disease: A systematic review — https://nesr.usda.gov/sites/default/files/2024-12/Food-sources-saturated-fat_cardiovascular-disease-2025DGACSystematicReview.pdf
- [16:49] Margarine From Oil (1940-1949) — https://www.youtube.com/watch?v=yWwhX-8-MsE
- [29:14] Whole Milk for Healthy Kids Act of 2025 – Implementation Requirements for the National School Lunch Program — https://www.fns.usda.gov/nslp/wmfhka-implementation
- [01:03] Prenatal methylmercury exposure from ocean fish consumption in the Seychelles child development study — https://doi.org/10.1016/S0140-6736(03)13371-5
- [01:10] Fatty acid desaturase insertion-deletion polymorphism rs66698963 predicts colorectal polyp prevention by the n-3 fatty acid eicosapentaenoic acid: a secondary analysis of the seAFOod polyp prevention trial — https://doi.org/10.1016/j.ajcnut.2024.06.004