The Influencer's DGAs & The Rancher's Pyramid

A nothing-Burger on the surface, more underneath

“Eat Real Food'“ is the message of the newly released 2025-2030 Dietary Guidelines for Americans (DGAs). Undoubtedly, this messaging, coupled with a smooth new website (realfood.gov) and already-consumer friendly talking points (‘eat more protein and less processed foods’) will be largely a hit and one of the more uncontroversial things this new administration does with respect to public health.

With the burden of diet-related disease that Americans face, it’s hard not to watch a press conference announcing new DGAs that place nutrition at the center of the country’s priorities and feel at least a little something. For anyone in nutrition and public health, the feeling of nutrition being a political hot potato politicians won’t touch for more than a few seconds has been a growing frustration over the past decade - every food advocate in the DC area that I know has opined about how impossible it is to get anyone - left, right or center - to take nutrition all that seriously after the insane politicization of Michelle Obama’s efforts. Couple that frustration (mostly on the Left) with the emergence of a Right-coded foodie coalition under MAHA and you’ve got a bipartisan excitement about doing something about nutrition for the first time in a while. Excitement only goes so far though and there are only so many tools that the government has to take action.

Enter scene: the new Dietary Guidelines. DGAs are a small but meaningful tool that the government has to take some action and it’s arguably the first real test of whether the MAHA coalition can get serious on nutrition (I don’t consider much of their nutrition efforts so far very serious). In addition to being a bit of a cultural rallying point, the DGAs hold some weight by influencing/constraining federal programs like the National School Lunch Program, the Senior Nutrition Program, and SNAP and WIC. While on their own they often don’t dictate policies that’ll vastly change the food supply, they’re the framework off which you can start to build policies and you’ll see them cited as an authoritative piece of evidence in impactful legislation like removing GRAS recognition for trans fat containing oils.

This edition of the DGAs have been long awaited, with much speculation and chatter over the past year. Today, we saw them in full:

• A 10-page DGA

• A 90 page Scientific Report

• A …drumroll…418 page Appendix

• We also saw a brand new (1992-inverted) Food Pyramid, the newest pictographic food guide update (RIP MyPlate, 2011-2026)

  

Note: That’s a lot to take in and i’m sure we’ll all be digesting (🥁) for a while - but here’s my first take from a decently thorough review.

_______________________________________________________________________

At the highest level, before we even look at what the guidelines say, we’re on uncertain grounds with this DGA approach. Normally, we have the Dietary Guidelines Advisory Committee process and report that gets transformed by USDA and HHS into the DGAs - that’s always a bit of a black box and where Committee recommendations get dropped/changed and there’s much speculation as to why. Nevertheless, at the other side of that box, a 100+ page report (2020-2025 was a 164 pg report) comes out that is intended to guide healthcare professionals and be used in policy and programs. Alongside the report we get updates to the Food Guides like MyPlate and new consumer friendly materials that aim to translate the guidelines (but the guidelines themselves aren’t intended to actually be read by Americans).

With this edition, we’re getting a 10 page report, clearly intended for the public and not for healthcare professionals, program administators or policymakers. The implications of this alone are vast but challenging to foresee the specifics. The high critiques of the DGAC report by the HHS Secretary resulted in a new scientific report and appendices that are a bit all over the place but seem to attempt to justify their deviations from it - it’s not clear whether their approach (more on these below) will be seen as rigorous enough to convince healthcare professionals to change their practice or industries to undertake pricey reformulations of products.

In some ways, this new DGA takes us back to the old DGA style of the original reports that were shorter and more consumer facing. History is repeating itself but we should remember why DGAs evolved as they did - short, vague, and non-quantifiable recommendations in the DGAs don’t hold much weight in policy. Saying ‘avoid too much sugar’ doesn’t actually say anything and when you look across federal programs, there’s no way to implement that, let alone enforce it. However, when you say, no more than 10% of your calories should come from added sugars, that allows for stricter guidelines for school lunches, encourages product reformulations, allows for stronger definitions of what can be labeled ‘healthy’ and facilitates front of package labels that serve to let consumers know products are high in something. These are small tools in changing the food supply and educating consumers but meaningful ones. There’s a real potential here that in choosing political theatre and a forward facing guideline, the new DGAs have defanged themselves quite a bit - time will tell.

1980 DGAs

_______________________________________________________________________

Before we dive into the guidelines, (spoiler alert) you should know that they’re not exactly always coherent with themselves, with the new Food Pyramid that was released, or with the messaging coming from different actors in the administration. It’s quite clear reading these guidelines and the supporting scientific evidence that this was pretty rapidly put together, has numerous authors across different sections (the voice changes are apparent throughout), and didn’t have 1 clear person with a strong knowledge of nutrition guiding the whole process and focused on coherence. As you go through them, there are some straight forward things, but also a number of contradictory recommendations and just plain flubs. I’m already seeing pretty vastly different takes and interpretations and that’s because the guidelines lend themselves to such ambiguities.

What the Guidelines Recommend:

At a high-level, despite much fanfare about major changes in press conferences and messaging, there’s not much that has changed when you look at the overall recommended food groups - eat 5 servings of fruits and veggies and drink 3 cups of milk per day, choose whole grains, focus on protein foods. There’s less specificity in this edition, with removal of the specific groupings for vegetables and types of protein foods - this could have implications down the road for allowing federal programs to be more flexible, but would also allow for feeding all poultry and no fatty fish (a big downfall given low omega 3 status of the USA). Most categories, though, stay roughly the same or similar (e.g., they reduce added fats slightly to ~18-22g/day (equivalent of ~4.5 daily servings)) although the reasons for differences aren’t really explained - these serving sizes typically come from food pattern modeling exercises that assess the likelihood of meeting estimated nutrient requirements and nutrient limits while consuming common American foods. The biggest thing that stuck out to me here was essentially no recommendation for refined grains - while we always hear about the benefits of whole grains, which offer more fiber and micronutrients, refined grains have historically constrained grain recommendations and prevented them from recommending all whole grains due to their mandatory folic acid fortification (an amazingly successful policy at reducing Neural Tube Defects) - food pattern modeling exercises don’t meet recommended intakes of folate for females of reproductive age without some refined grains in the diet. With the new DGA recommendation to consume fewer grains and all of them whole, it would be prudent for FDA to update its current folic acid fortification policy to include whole grain products and not just refined (note: the guidelines emphasize folate in other places for females of reproductive age and in pregnancy but active supplementation isn’t likely to replace fortification - most pregnancies are unplanned and the neural tube closes in the first month of pregnancy, before individuals know that they are pregnant and can take action to increase folate intakes; fortification ensures passive intakes of folic acid for anyone capable of getting pregnant and adequate folic acid at the time of neural tube closure).

When we move over to the written text of the DGAs, we also see a lot of the same from past editions as well as what we expected from the 2025-2030 DGAC report, but it’s also where deviations emerge. There is the usual ‘eat an appropriate amount of calories to maintain energy balance’, stay hydrated by drinking water and unsweetened beverages, eat a variety of nutrient dense foods and colorful vegetables, limit added sugars and choose whole grains. Reasonable, simple advice is given by life stage (I’m glad they clearly engaged AAP in writing the infant feeding guidelines); for those with chronic disease and vegetarians/vegans.

The major deviations aren’t surprising if you’ve followed the administration’s regular signaling to support ag/ranchers, and messaging on saturated fat, seed oils, and ultraprocessed foods. The major deviations are:

• Protein - aim for protein goals of 1.2-1.6 grams protein per kilogram of body weight - higher than the current Dietary Reference Intakes of 0.8g/kg for adults but within the Acceptable Macronutrient Distribution Range.

  • My take: This comes from a ‘rapid’ systematic review (no meta-analysis) in the Appendix focused on weight and body composition outcomes, that predominantly relied on studies where individuals were intentionally losing weight and/or undertaking exercise. This literature typically doesn’t inform general population guidance because active weight loss and/or undertaking significant exercise are covered in tailored obesity medicine and sports guidelines. The numbers themselves aren’t egregious, are well-within recommended macronutrient ranges and most folks in nationally representative already report consuming 1.0-1.2g/kg/d of protein. That said, there’s pretty limited evidence that high protein diets are some silver bullet for magical weight loss - most studies that suggest small benefits are complicated/confounded by active weight loss advice, modifying the energy density of the diet at the same time as increasing protein or using protein preloads prior to meals that don’t reflect what simply eating more high protein foods will do. The guidelines emphasize that the higher protein diets they recommend should come from meats, dairy and plant sources but the reality of this recommendation is that we will likely see it translated into more protein-boosted cereals, bars, baked goods, chips and supplements, a trend that has already overtaken the supermarket shelves.

• Fats - the DGAs define healthy fats as those which are in meats, poultry, eggs, omega 3-rich seafood, nuts, seeds, full-fat dairy, olives and avocados. In the dairy section, an emphasis on choosing full-fat dairy is restated. When cooking, it states to use olive oil, butter or beef tallow. Confusingly, however, the DGAs state that saturated fat (SFA) consumption should not exceed 10% of total daily calories and that limiting processed foods will help reduce this goal.

  • My take: Given all the fanfare about seed oils and ending the ‘war’ on saturated fat, it’s unsurprising that this recommendation is what it is. I’m all for an endorsement of more omega 3-rich seafood for americans, and a focus on nuts/seeds, olive oils and avocados, but this recommendation language is clearly intended to deviate from the long standing advice to replace SFA intakes with polyunsaturated fat intakes (PUFA). The DGA recommendation is clearly endorsing high animal fat intakes, which will result in high saturated fat intakes, but confusingly retains the former caps on SatFat at 10% of kilocalories - the 2 struggle to coexist. For reference, 3 cups of whole milk will get you about 6% of your calories from saturated fat if you’re consuming a ~2000kcal diet, so that leaves little room for fatty meats and the residual saturated fats in plant foods to increase your intakes (Scroll to the Bottom where I made a pretty idealized DGA sample meal plan showing this). Few people actually calculate their intakes of saturated fat so the point of consumer facing material is to provide food-based recommendations that help meet the nutrient-based restrictions - here they just simply don’t really align well. The guidelines seem to try to square the circle by saying to limit saturated fats through limiting processed foods, but this really only tackles common processed food sources of saturated fats like mixed dishes with fattier cuts of meats, ice creams and grain-based products made with either butter or palm oil - given the emphasis on meat and dairy containing healthy fats and to cook with butter, this collective advice just doesn’t make a ton of sense and it’s pretty clear that the 10% cap was retained despite all of the recommendations and supporting Scientific Review in this DGA release wanting it gone. For federal programs that are supposed to adhere to these recommendations like school lunch, this will allow some but not much more fatty cuts of meat and whole dairy on the menu, but this will be a challenge for managers planning menus to meet. These recs are conflicting and it’s hard to tell who they are for and what impacts they will have.

• ‘Highly Processed’ - the DGAs recommend avoiding ‘highly processed’ foods (notably not using the term ‘ultraprocessed’/UPF) - these are described as packaged, prepared, ready to eat or salt and sweet foods that have added sugars and sodium, artifical flavors, dyes and preservatives, and low calorie sweeteners.

  • My take: The administration couldn’t get a UPF definition together in time for this report and thus settled on a new, undefined term of ‘highly processed’. Folks in public health nutrition and foodie circles have been trying for decades to say ‘eat good foods/don’t eat bad foods’ and make a scientific definition with the vibes of ‘junk food’- none have succeeded. This is the newest attempt. For more elite and educated folks, processed foods are like p*rn - they know it when they see it but no one can agree on the definition. Whether this mish mash definition spanning nutrients in the formulation, packaging, and additives will really land with consumers and impact health is anybodys guess. You might hear ‘highly processed’ and think poptarts and be shocked anyone could take issue with the recommendation- but this definition also fits your unsweetened almond/oat milk latte, canned beans and veggies, many yogurts, granolas, trail mixes, protein shakes, etc. Given the lack of a real definition here, ‘highly processed’ won’t likely hold much regulatory weight cause there’s nothing to implement or enforce - it’s a vibes recommendation that’ll appease the masses but functionally do quite little IMO. If they do try to enforce, there’ll be an uproar from school districts who won’t be able to afford paying for staff and infrastructure to make everything from scratch - I won’t hold my breath for Congress to put a massive infusion of $ to support unprocessed school lunches and if they tried, industries that go after those school lunch contracts and provide the food will surely be sending in the lobbyists.

Apart from these major highlighted changes, the 10pg report has only a few other notables.

• There is a concerning watering down of guidance on alcohol, with no quantitative limitations given in this edition, defended in the scientific review document as being based on the 2024 NASEM report that found limitations in making causal inferences in this area (this will undoubtedly get touted as a win for the alcohol industry). As noted above, when you take out the quantitative limits, you defang yourself against using the DGAs to support any real policy.

• There are one-off eyebrow raising statements like that full-fat dairy ‘supports brain development’ in middle childhood - there is genuine scientific interest in the milk fat globule membrane from dairy fat and brain development in infancy, but little evidence to be enthusiastic at these later stages of development; trials in this age group have showed counseling to reduce total and saturated fat intakes reduce serum cholesterol without impacting typical neurological development. I’m not sure if they’re confusing the full fat dairy recommendations up to age 2 with these recommendations for 5-10yo.

• There are some odd approaches, like encouraging consumers to get to know all of the dozens of ways added sugars can appear on a label, when added sugars already appear quantified on a label and could be a focus - I’m not surprised that this admin is advocating popular instagram influencer-style advice that stresses consumers into memorizing ingredient lists but it’s highly unlikely to be useful to most and distract from the overall goal of minimizing added sugars altogether. Unfortunately, this DGA removes the daily cap on added sugar of <10% kilocaries and replaces it with vague <10g sugar per meal and <5g sugar per snack, without defining what a meal and a snack look like or how many are expected per day. Again, in choosing consumer-facing advice, they’ve walked back recommendations that were more explicitly enforceable in policy. It’s likely these absolute grams per eating occasion limits will still result in <10% kcal fortunately, assuming they are enforcable (3 meals + 2 snacks on a 2000 calorie diet = 40g max, which is 8% of calories).

• As another nod to influencers, there’s a random section on ‘Gut Health’ that mentions eating somewhat unconventional and specific fermented foods (e.g. kimchi, sauerkraut, kefir and miso) but doesn’t provide anything quantitative that’ll carry forward (I found it odd they didn’t even mention the most common fermented food consumed by americans…yogurt).

• There is a notable absence of information on choosing pasteurized dairy products as in past guidelines. Fortunately, the new guidelines don’t say to consume raw milk.

• As the Academy of Nutrition & Dietetics points out, there’s limited considerations in the DGAs for individuals who choose to be dairy free or have lactose intolerance. The use of fortified, processed dairy alternatives is largely at odds with the avoiding ‘highly processed’ foods advice. This is particularly concerning for how it’ll impact school lunch and non-dairy alternative availability.

• There’s an abject dismissal of the ‘health equity’ lens of the DGAC 2025-2030 approach. This isn’t too surprising with the anti-DEI backlash of the admin, but its a shame given that guidelines are not prescriptions and people can incorporate science-based advice in many ways that interface with culture, values, preferences, and economic situations. To tackle public health nutrition seriously, you should consider the ways that race and ethnicity play into not only food culture but who is advertised to and what is advertised, how past injustices shape economics and local food environments, etc. Passing over this opportunity to call out reality and embrace the diversity that is intrinsic to food to quote the foodie elite’s mantra of “Eat Real Food” is pretty big flop of this DGA. For sure, some will prefer this not only for anti-DEI reasons but because it chooses a clearer ‘eat this, not that’ framing over the more flexible and conditional language of past DGAs, but that makes these Dietary Guidelines for (A Subset of) Americans.

_______________________________________________________________________

The Scientific Review & Appendix

To understand the process behind these DGAs, you have to go dig through the mix of the 90pg scientific review document and the 418 page appendix. Even then, how the final recommendations in the DGAs were determined is pretty opaque - they appear to carry over some recommendations from the DGAC report but deviate on a few fronts, citing everything from concerns about a ‘health equity lens’ to conducting new narrative and systematic reviews on select topics - but the exact way this all lead to the specific recommendations were received isn’t always very clear. Having reviewed them all I think we can discern a bit about how we got to what we got.

The scientific review states off the top in its Preface that it is assuming that ‘natural’, minimally processed foods are the basis of a healthy diet and that strong causal evidence is needed to recommend against these foods or to recommend foods & ingredients that are ‘highly processed’. In addition to this baseline assumption, the report goes on to discuss that we need rigorous science, defined as randomized controlled trials to prove causal links between nutrition and outcomes that are actual clinical endpoints of disease incidence and mortality, not surrogate/intermediate endpoints like LDL-cholesterol, blood pressure, blood glucose, etc. This might all sound reasonable and like rigorous science is being emphasized and it is indeed a very rigorous standard - it is basically the standard in many areas for pharmaceutical trials, given long learned lessons that observational evidence and surrogate endpoints can be unreliable predictors of changes in disease risk/outcomes. Applying this standard to nutrition would be extremely rigorous if you applied it equally across the board and it would also make the guidelines simple because you’d have few if any guidelines related to chronic disease risk: indeed, nutrition largely lacks well-powered randomized controlled trials that define dose response relationships between diet and actual disease endpoints. Relating nutrition to chronic disease risk has nearly always relied on the synthesis of shorter term controlled trials with surrogate endpoints (of which we only have a few: LDL-C, blood glucose, blood pressure and bone mineral density) along with prospective cohorts where we assess diet through either self reported measures or biomarkers and relate this to chronic disease endpoints like heart attacks, strokes, type 2 diabetes, cancer and total mortality. This reality has always fueled controversy in nutrition. If these more limited lines of data all become suddenly too weak to inform recommendations, then you just don’t have recommendations, apart from general Diabetes Prevention Program style ‘lose weight’ (through low fat diet recs, mind you) recs or attempted inferences from trials like PREDIMED that replaced liters of regular olive oil with extra virgin olive oil in the context of a Spanish style diet.

So this DGA claims to hold nutrition research to these rigorous standards but also made a lot of recommendations, so how did they? You can start to envision how, if you combine this evidentiary standard with the ‘appeal to nature’ fallacy basis of assuming ‘natural’ foods are good until rigorously proven otherwise and anything ‘highly processed’ is only bad until rigorously proven otherwise. Dairy and meat fat is good until proven otherwise - the fact that they are high in saturated fats and raise LDL-C isn’t rigorous enough controlled trial data because it’s a surrogate outcome; the fact that substituting them with unsaturated fats and whole grains in cohort studies reduces chronic disease and mortality isn’t rigorous enough data because the data is self-reported and isn’t a randomized trial; and the fact that existing controlled trials with disease endpoints have legitimate limitations prevents them from being rigorous enough data to conclude anything— so voilà, you haven’t met the evidentiary standpoint to recommend against consuming natural saturated fats. Similarly, when you start from the perspective that high-linoleic acid seed oils are intrinsically bad until exonerated because they’re ‘highly processed’, there is also no level of evidence to recommend them. The review and appendix actually spend considerable space detailing limitations of the epidemiology that consistently finds biomarkers of higher linoleic acid are also associated with reduced risks of chronic disease and have been used to support traditional recommendations to replace saturated fats with linoleic acid-rich PUFAs. The Review/Appendix contain novel, unpublished analyses pointing out that linoleic acid is enriched in the cholesteryl ester fraction of blood and that this can make total plasma levels of linoleic acid look low as a % of total fatty acids in the blood when triglycerides are high (triglycerides being mostly enriched in saturated and monounsaturated fatty acids), creating a confounded situation where low linoleic acid levels aren’t indicative of low intakes and harmful but rather a proxy for high triglycerides, an ApoB-containing lipoprotein linked to CVD risk. While this is physiologically and epidemiologically a reasonable critique, the Appendix spends many words on this limitation and doesn’t grapple with the still sizable body of cohorts that have measured linoleic acid in that cholesteryl ester fraction, as well as in RBCs and adipose tissue that still find protective effects of higher levels of linoleic acid against diabetes , CVD and mortality. That evidence, of course, is still only prospective cohort data (ie not a large randomized controlled trial) and can still be thrown out by the very rigorous standards laid out, but it does beg the question why so much time was spent on these nascent, unpublished analyses around confounding by triglycerides.

If the Review had set out their rigorous causality standard and held it genuinely, I’d spend less time critiquing it here - but they didn’t. They of course couldn’t say all diet research is too limited and we need to massively invest in nutrition research like never seen before to conduct huge trials and generate higher confidence data- their whole basis is that nutrition is the key determinant of chronic disease risk, that you need to take personal responsibility to reduce your risk and that you’ve been lied to by past administrations who’s recommendations caused your health issues (not conspiracy - this is all over the current DGAs and announcements). These new DGAs still make plenty of recommendations, and ipso facto, none of them fit their very rigorous evidence standards. Do we have rigorous blockbuster trials on reducing added sugars and refined grains showing they reduce rates of obesity, type 2 diabetes, cardiovascular disease and mortality? Nope. This body of evidence is actually meaningfully weaker than the body of evidence supporting past recommendations to replace SFA with PUFA. How about to support their protein recs? Nope. When it comes to their protein advice, where they are changing something from the DGAC, we get a more detailed scientific review that relies on evidence that identifies small, somewhat inconsistent effects of higher protein diets on body weight and composition measures - not rigorous disease endpoints. The link to disease endpoints is actually super weak, even for age-related muscle loss as highlighted in the systematic reviews set to inform the forthcoming update of the Macronutrient DRIs. When it comes to why we don’t get the recommendations to use vegetable oils when cooking and choose more PUFA-rich foods, there is an additional section of the review that focuses heavily on oxidized lipids, emphasizing theoretical harms of consuming high polyunsaturated fat rich oils from acute single meal studies measuring oxidized lipid metabolites in blood and observational data associating self-reported fried foods with disease risk. This section also emphasizes a single 4 week study comparing fried meat vs boiled/steamed meat that found worsened BMI and glycemic control with frying (the type of oil used to fry the meat wasn’t specified in the study). Again, BMI and glycemic control are surrogate endpoints and we can’t even link these to being fried in PUFA rich oils - the effects of this study as well as the fried food epidemiology could very well be due to frying increasing other non-lipid compounds like dietary advanced glycation endproducts (AGEs). I’m not arguing for fried food consumption here by any means - rather, I’m emphasizing how easy it is to knock down any form of evidence you don’t like when you have an inconsistent evidence standard - it lets you emphasize hot topics with no causal data linked to health outcomes like more protein and handwaving about processed foods, non-nutritive sweeteners, gut health and fermented foods when you see fit, and dismiss large bodies of evidence when you’d like to make the opposite conclusions.

Ultimately, the Appendix is a lot of work by individual authors who are all published in these topics and if you know them (their funding sources are already being critiqued), the conclusions they would come to were pretty obvious. There is no illusion from reading the Review and Appendix that these new DGAs resulted from a rigorous and transparent process that pre-registered questions to be addressed, reviewed the data, and got the experts in a room to set down a common measuring stick by which they’re assessing the evidence- the approach is little more than a gish gallop to support the preformed conclusions that the HHS Secretary, MAHA advocates and influencers have been pushing since the moment they got into office. It’s actually a shame, because there is new work in these Appendices that should be discussed and debated about in the field but their premature and non-transparent launch to justify predestined conclusions probably won’t have them generating the rigorous discussions that they should for some time.

_______________________________________________________________________

The Pyramid

I shouldn’t be surprised that we are living in a world where a South Park episode has come to life and we’ve flipped the evidence pyramid upside down. In reality, this evidence pyramid really shouldn’t be depicted as a pyramid - pyramids are used to convey relative amounts of foods that should be consumed, with foods to be minimized displayed at the narrowing point and the base detailing the majority of what your diet should be. The 1992 Food Pyramid that gets rightly critiqued for being too grain-heavy at the bottom had sweets and oils at the top, intended to be used sparingly.

You’d think looking at the new DGA pyramid that this intended to be a reduced carbohydrate diet pyramid, with a strong base of fatty meats and dairy and olive oil, and low starch vegetables, with sparing fruits and nuts and even fewer grains. However, when you read the text, it describes 2-4 servings of whole grains that are clearly not represented in the pictures. Despite grouping plant-protein foods in equivalently alongside animal sources in the text of the DGAs and the Protein Foods Category of the recommended food groupings, you could blink and miss any plant protein sources in the pyramid, drowning in a sea of animal-based options. The animal based options also show somewhat normal or quite large serving sizes (a single serving of yogurt, large cuts of steak, a whole carton of milk, large wedge of cheese, a whole turkey, an egg, a large cut of salmon) whereas the plant proteins just show 1 nut of each type and a can of green beans - no lentils and chickpeas, no soy products (another slight to soybean farmers?). These generous animal portion sizes serve to further emphasize that you’re going to be going way over that 10% calorie saturated fat cap if you consume this dietary pattern. You’ll also likely come nowhere close to recommended fiber intakes and have pretty low intakes of other potential beneficial phytochemical bioactives. As dietary guidelines have drifted towards emphasizing more plant-based foods across the globe in the past decade, citing both health and sustainability rationale, its hard not to see the messaging and fanfare, and the visuals as being a hard pivot back towards a heavy reliance on animal foods. You could say that’s for health, but the administration have been anything but quiet that these changes are for the economic future of ranchers/farmers.

_______________________________________________________________________

The Flubs

It’s not worth belaboring much but the feeling that limited nutrition expertise was involved in bringing this DGA together has been there for months - we’ve had RFK Jr mocking ingredient lists with hard to pronounce chemicals that include basic vitamins; we’ve had ridiculous statements about seed oils in infant formula (the source of essential fatty acids); we’ve had clear confusion about what the difference between the Dietary Guidelines Advisory Committee Report is vs what the Dietary Guidelines are. The list goes on and on and it doesn’t stop with this DGAs. A couple examples:

The ‘healthy fats’ component of the DGAs says to cook with oils containing essential fatty acids like olive oil and that other options can include butter and beef tallow. This is wrong but I’m not exactly sure in which way its wrong. The entire Appendix makes the argument that we shouldn’t cook in essential fatty acids because they’re readily oxidizable - so this should probably say ‘low in essential fatty acids’. That probably sounds bad to consumers who expect to hear that they should eat ‘essential’ fats. If ‘essential’ was correct, then it’s also wrong - because olive oil is very low in polyunsaturated essential fatty acids. That’s why its been recommended in the Review/Appendix - because it presumably will result in less oxidized fats because it is mostly monounsaturated with a bit of saturated and even less poly. This might sound nitpicky but it screams in the face of anyone who has taken basic nutrition 101 and I’m not sure how it got through final reviews without one person who knows basic food composition or that what they’ve said is inconsistent with 1 of their key topics.

In their scientific review talking up the harms of food processing, there are both deeply unfair comparisons and just flat out wrong numbers included in their example of how processing harms the nutritional value of yogurt. The minimally processed, full fat yogurt that is shown in panel A is actually a high protein greek yogurt - its nutrition facts panel looks otherwise correct. The processed low fat yogurt in B looks super high in added sugars (which you can easily find in the grocery store but not all low fat yogurts are inherently this way) and is very low in protein - but you don’t lose protein when you skim yogurt. The product isn’t listed but you can guarantee that this comparator is a regular, non-Greek yogurt (greek yogurt is strained, concentrating the protein). The labels they’ve chosen here make processed low-fat yogurt look egregiously bad, when its actually not hard at all to find a low or no sugar added greek yogurt that is skimmed in supermarkets. In panel C, things just get unserious - This is a no fat greek yogurt that is listed as having 150 calories per serving, but has 0g fat, 9g carbohydrates and 12g of sugar - that should tally up to ~80 calories, not 150. They’re obscuring a perfect example of howing processing can actually keep protein high, added sugar and saturated fat low, while providing all the same essential micronutrients for fewer calories. Again, to a nutrition knowledge person, this screams amateur hour.

In the Daily Servings chart, there is also some absurdity in the ‘protein foods’ serving size examples. The equivalent servings here are all over the place and vary in protein: 3oz poultry (26g protein), 1 egg (6g), 1 oz nuts (6g), or 1/2 cup lentils (12g). Each differs in protein considerably and no matter what mix you eat, 3-4 of these servings only gets you close to the 100g target protein goal if its exclusively coming from meat. It’s pretty clear no one did any modeling of these recommendations to check that they actually met their own recommendations.

Also in the Veg*n section, it encourages monitoring of nutrient status for several nutrients, including calcium, which you need stable isotopes to measure anything close to the ‘status’ of, and iodine, which individual status assessment requires multiple complete 24 hour urines to assess (no one does this clinically). So their recommendation is for the public/professionals to monitor the status of nutrients they can’t monitor the status for - helpful!

While we’re looking at this section - I’ve had concerned vegans and vegetarians reach out about the long list of shortfall nutrients this very animal food-forward DGA cites. That list comes from modeling that the actual DGAC committee did but you have to go into the supplementary documents to find it. This list of shortfall nutrients comes simply when you just remove eggs and dairy from the main USDA pattern, with no substitutions. You lose all of the nutrients from those foods and eat nothing else in this modeling, so you also end up with a bunch of calories leftover. The new guidelines focused on this modeling effort that was never intended to represent a diet that anyone actually eats, and ignores the modeling where the DGAC substituting dairy and eggs with fortified soy beverages that substantially reduces the list of nutrients where intakes are low. Even then, there are still calories leftover for most age groups so smart choices can be made to fill gaps. There are no nationally representative data on what veg*ns eat or those that are common in the veg*n food landscape so there’s no fair way (like we do for omnivorous eating patterns) to guestimate what a general veg*n pattern might fall short on - the new DGAs use of this data totally distorts this reality to make veg*n diets look way worse off than they are. It is still true that veg*n diets can run low on B12, iron, vitamin D, iodine, choline and omega 3’s and smart supplementation and planning is key. But this guideline, with its emphasis on animal products and philosophical bias against processed foods (thus not recommending fortified soy beverages), unsurprisingly is likely to confuse and offer little practical advice for veg*ns.

_______________________________________________________________________

Conclusion

I had worried these DGAs might say something bad or dangerous. Seeing the final version, I’m not sure it says much of anything concretely. Depending on which part of this release you choose to focus on, you can conclude that not much has changed or things have changed dramatically - there are multiple levels of contradictions and errors that facilitate ambiguity. I can’t imagine a lot of what’s new will hold much policy weight or be cogent enough to implement, and I don’t imagine the short forward facing consumer version will do much - core recs in the new DGAs are minimally different than older versions that also recommended low sugars and high fruit/veg/whole grain intakes that never took hold at the population level.

I’m left feeling how I’ve always felt about DGAs - they’re a small tool to enact change but we won’t get anywhere without a functional Congress. Congress is going to need proposed policies that carefully detail all of the levers across the food value and supply chain that influence what ends up on our plates and in our stomachs (from funding research and innovation in Ag through to subsidies, taxation and advertising laws). There’s no silver bullet policy that’ll change the food system, let alone a silver bullet federal guidance document. I’ve not seen any evidence Congress has much of a pulse, let alone a hunger to start selecting economic winners and losers across the food system. If there’s any real impact of these DGAs, it’ll be the cultural movement of MAHA getting folks to pay attention, but it’s not clear that really has longevity - we know from decades of nutrition intervention studies that short-term dietary changes don’t persist in the vast majority, particularly when the environment doesn’t change and the default option is still overwhelmingly one misaligned with health.

_______________________________________________________________________

In something that couldn’t be more 2025-2026 coded, the Scientific Review also panders to influencer culture around ‘T’ optimization and has a very non-exhaustive quick overview of diet and testosterone. Can’t imagine how that got pushed to get its own section!

_______________________________________________________________________

DGA Style Meal plans:

There’s no 1 diet plan that will be perfectly representative here for any one metric, but I tried to keep things pretty ‘whole foods’ and minimalist (note: this meal plan would be a stretch for the average American to eat or looks nothing like the average of our food environment), while hitting recommended food groups, the new protein goals (~100g) and choosing the fattier animal products. It’s a struggle to stay below the 10% cap.

It’s a good time to mention that the previous 10% cap of on saturated fat wasn’t driven by concerns about CVD risk - it was driven by modeling exercises that showed higher than 10%E SFA resulted in imbalanced overall dietary patterns that didn’t meet various other nutrient targets. I didn’t model all of the micronutrients of course but you can see with nutrients like fiber that the amounts vary quite a bit and can fall short of the recommendation (14g fiber per 1000 k calories) if you don’t make specific swaps. I suspect it’d be even harder (close to impossible except for very walnut enriched diets) to meet the DRI range for PUFA (5-10%E) with this new DGA - another area besides protein where the guidelines are bucking the mandate to meet the DRIs.

Full Fat Dairy only

Swapping 2 of the full fat dairy servings and fatty meat approaches the 10%