Nutrition Advice Shifts From Nutrient Targets to Minimally Processed Foods

The most consequential shift in the updated dietary guidelines, in Dariush Mozaffarian’s reading, is not the red-meat-and-butter story that dominated much of the coverage. Mozaffarian said that framing missed what he considers the largest change in 45 years of federal nutrition advice: for the first time, the guidelines explicitly tell people not to eat a broad set of foods.

Previous guidelines, in his view, mostly avoided naming foods to limit. They encouraged more of some categories and buried reduction advice inside nutrient targets such as sugar, saturated fat, or salt. The new version, as Mozaffarian described it, says not to eat highly processed foods, ready-to-heat foods, cookies, candies, sugar-sweetened beverages, refined grains, salty or sweet packaged foods, and processed meats with added chemicals. He estimated that this amounts to 50% to 60% of the food supply being put on the “do not eat” side of the ledger.

Mozaffarian also emphasized that the guidelines do not elevate butter and beef tallow as preferred fats. In his reading, they prioritize oils with essential fatty acids, which means plant oils and, in particular, seed oils, even if the guidelines do not use that politically charged phrase. Butter and beef tallow appear, he said, as options rather than the central recommendation.

The same reading underlies his account of the other major changes. The guidelines raise the protein target, go “really hard” on sugar, and move from the old “make half your grains whole grains” advice to a recommendation that all grains be whole grains and refined grains be avoided. For children under 10, Mozaffarian said the guidance calls for no sugar at all; for foods and meals, he cited targets of 2.5 grams of sugar per item and 10 grams per meal, which he said are far below what is common in school meals or average foods.

Jessica Knurick largely agreed that the guidelines themselves are better than the public controversy suggested, but she separated the document’s quality from its usefulness. She argued that dietary guidelines operate through three different lenses: public nutrition education, professional practice, and federal nutrition policy. The public-facing role has historically been weak. Knurick said fewer than 10% of Americans have ever followed the advice in the guidelines, so the documents have not meaningfully moved what people eat.

For registered dietitians and nutrition professionals, Knurick said the serving-size framework and food-group recommendations did not change enough to alter day-to-day counseling very much. The new whole-grain language illustrates the gap between ideal guidance and practical advice. If almost 90% of grains consumed in the United States are refined, telling people to eat only whole grains may be less useful in practice than helping them move from all refined grains toward half whole grains.

The policy lens is different again. The guidelines shape programs such as the National School Lunch Program, the National School Breakfast Program, WIC, and, when it existed, SNAP Education. For that work, measurable targets matter. Knurick’s assessment was that the new guidelines are more public-facing than some previous iterations, but that the “tangible markers” likely to move federal nutrition policy did not change as dramatically as the rhetoric around the document.

Ty Beal placed the seed-oil debate in that same gap between the text and the politics around it. The guidelines, he noted, do not say “seed oils,” even though the evidence reviews addressed them. Instead, they highlight olive oil, probably the least controversial oil, while recommending oils high in essential fatty acids. Those oils, he said, are seed oils. His distinction was that the guidance supports minimally processed foods made with such oils, not ultra-processed or deep-fried foods simply because they contain seed oils.

Beal saw unnecessary polarization in the debate. He said that when he talks with scientists such as Christopher Gardner and Deirdre Tobias, they agree on 90% to 95% of the substance. The fight is often over the margins, the graphics, and the marketing: how large the steak appears, how prominent the butter is, or whether a steak is shown next to birthday cake. Beal said a steak is healthier than cake, but that “just eat meat, meat, meat” is not what the guidelines say. In the text, plant-based proteins and animal proteins are treated as equivalent categories, not as a hierarchy that tells people to eat more meat than plant proteins.

Dariush Mozaffarian began his research career as a cardiologist asking what should replace saturated fat: carbohydrate, monounsaturated fat, polyunsaturated fat, protein, or something else. Over five years of work, he said, he came to a conclusion that surprised him: saturated fat was not a major cause of heart disease and should not be a priority metric for diet advice.

That does not mean saturated fat is biologically inert, or that foods containing it are irrelevant. His point was narrower: saturated fat, like total carbohydrate, total fat, or total protein, is too blunt a metric. Health effects depend on the food source, processing, quality, and dietary context. He called the 10% saturated-fat limit “made up” and said it should have been dropped.

Among foods that contain saturated fat, Mozaffarian singled out processed meats as the consistently harmful category. He defined them as meats processed so they do not spoil: low-fat deli meats, bacon, sausages, and hot dogs, including many products labeled nitrite-free. He said such products can still contain nitrites when celery juice or beet juice is used, and that the European Union does not allow the same “nitrite-free” labeling approach. In his account, salt, nitrites, and other compounds in processed meats are clearly linked to cardiovascular risk, colon cancer risk, and diabetes risk.

Fresh red meat is different. Mozaffarian said unprocessed beef, pork, and lamb are not nearly as harmful as processed meats, and the disease links are weaker. But he also said there has “never, ever” been a study showing that people who eat more red meat have better health outcomes.

The more plausible risks from red meat, in his explanation, are not primarily about saturated fat or cardiovascular disease. The stronger links are to colorectal cancer and diabetes, diseases he associated with inflammation, carcinogenesis, and the microbiome. He pointed to heme iron as a signature compound: necessary in small amounts, but pro-inflammatory in excess. He also cited compounds in red meat that the microbiome can metabolize into carcinogenic compounds.

His practical interpretation was that one or two servings of unprocessed red meat per week is probably acceptable, while seven or 10 servings a week is not where people should be. But the comparison matters. A fresh steak is a worse choice than fish, almonds, or blueberries. It may be a better choice than a baked potato, dinner roll, most breakfast cereals, crackers, or other ultra-processed foods.

That comparison explains why some people improve when they move to a carnivore-style diet from an average ultra-processed American diet. Mozaffarian said they improve because they have removed the junk. But he argued that moving from there to a Mediterranean-style diet would likely improve health further. Knurick agreed that a Mediterranean-style pattern is about as close as nutrition gets to common ground: “nobody could disagree with that,” she said.

Preparation also matters. Asked about charred meat and colorectal cancer, Mozaffarian said high-temperature cooking of any food can create carcinogenic compounds and that people should not burn food, particularly meat. He compared it to fish: a fish sandwich or fish sticks can be linked to harm when the fish is highly processed, battered, fried, and cooked in reused oils. The food category alone does not settle the health effect.

Protein has become a marketing strategy as much as a nutrient. Allison Aubrey pointed to protein lattes at Starbucks, protein Pop-Tarts, protein Cheetos, and protein popcorn. The new guidelines, as discussed, raise the protein recommendation to 1.2 grams per kilogram of body weight; Aubrey translated that into about 77 grams per day for someone her size, noting that a five-ounce chicken breast contains roughly 30 grams.

Jessica Knurick argued that protein receives more attention than the evidence warrants for the average American diet. Using national diet data such as NHANES and What We Eat in America, she said protein has not recently been a nutrient of concern. Americans, on average, already consume roughly 1 to 1.2 grams per kilogram per day, which is in line with the new target. There are many nutrients of concern, she said, but protein is not generally one of them.

Her concern is that food companies are eager to capitalize on the rhetoric. At a natural-products conference she attended, she said, “pretty much everything” was ultra-processed food with added protein. Her practical advice was not to treat protein as a powder or fortification label, but to get it from food-based sources.

Mozaffarian agreed and reframed the target as a percentage of calories. The guidelines moved, in his shorthand, from about 15% or 16% of calories from protein to roughly 20%. The average American is already at 16% or 17%. The more important point, he said, is that extra protein is only beneficial for building or preserving muscle if a person is also doing regular strength training.

If muscles have not been worked, Mozaffarian said, the body does not turn excess protein into muscle; it stores the surplus energy, including as fat. He supported strength training as important, but rejected the idea that adding protein to drinks will produce a muscular body on its own.

There are exceptions. Protein needs can increase with age, especially for women over 60. Mozaffarian agreed that roughly 15% of Americans may not get enough protein, and that older adults, particularly older women, should pay attention to intake. The dispute was not over whether protein matters; it was over whether the broad cultural fixation reflects the population-level problem.

Ty Beal described sugar-sweetened beverages as the clearest policy target because the science is unusually uncontroversial. Liquid sugars, he said, are among the most harmful and problematic foods, and they are easy to define in legislation. That matters for programs such as SNAP, where food-category boundaries determine what can be purchased with benefits.

His position was that excluding sugar-sweetened beverages from SNAP would shift taxpayer dollars toward more nutritious foods by cutting out the worst offenders. But he treated that as the easy case. Policies aimed at school meals and ultra-processed foods are more complex because the rhetoric of better food must be matched with infrastructure.

Beal has two children in public schools and said their food is “really not good,” including cinnamon rolls for breakfast. He supports the direction of state efforts to improve school meals and reduce ultra-processed foods, but warned that without investment, underfunded schools will be left behind. Many schools lack staff, equipment, and functioning kitchens. Some cannot cook from scratch because there is no real kitchen to cook in.

He described a rare moment of bipartisan agreement: Democrats and Republicans are talking about ultra-processed foods and recognizing that the food system is “out of whack.” But his concern was equity. If healthier school food is mandated without new funding, the schools and families with the fewest resources will face the hardest transition. He argued that downstream healthcare savings may be real, but the system must invest upfront.

Jessica Knurick made the same point through the economics of school meals. The reimbursement rate discussed was $4.60, an amount that must cover not only food but the whole operation of serving the meal. Knurick said the number includes equipment and the people who serve meals. Some districts, supported by local resources, grants, donors, and private funders, can run school lunch programs that she said she would want to eat from. Other districts operate with little more than a few microwaves.

The aspiration to improve school nutrition is broadly shared, Knurick said. The hard question is how. New USDA rulemaking tied to earlier guidelines is beginning to impose added-sugar limits on specific items such as yogurt and milk and keeping total added sugar under 10%, with additional sodium targets coming later. She called those targets good, but said school nutrition staff are trying to meet them under severe constraints. Without funding from Congress, she argued, meaningful progress will be very difficult.

Dariush Mozaffarian added another policy lever: the FDA. He said the agency has a budget of about $7 billion, but only about $100 million goes toward food, and the human-nutrition budget is only around $8 million to $10 million. Despite that limited nutrition budget, he said, the current FDA is pushing regulation more actively than in the past.

The first major potential “stick,” in his view, is closing the GRAS loophole. GRAS stands for “Generally Recognized as Safe.” As Mozaffarian described it, the pathway allows companies to determine for themselves that an ingredient is safe, hire their own scientists, place the data in a file cabinet, avoid telling the FDA, and put the ingredient into the food supply. The pathway was created in 1958 for familiar ingredients such as vinegar, garlic, and turmeric, he said, but companies have since used it for 99% of ingredients entering the food supply. He said no other country has the same loophole and that a proposed rule to close it was sitting at the White House Office of Management and Budget.

The second regulatory move, which Mozaffarian said industry fears even more, is to establish a federal definition of ultra-processed food. States including California have defined it or proposed definitions, but a federal definition would create a basis for policy action.

On the “carrot” side, Mozaffarian argued that governments have long invested in sectors deemed important to national security: railroads, technology, clean energy. Food has not received a similar national innovation strategy. He wants federal loans, tax credits, consumer credits, and other tools aimed at healthier, equitable, sustainable foods. Beal added that procurement is another lever: school meals, hospitals, and military food can create demand at scale and reward companies for producing healthier food.

Dariush Mozaffarian leads the Food is Medicine Institute at Tufts, and he framed food-based healthcare as distinct from nutrition assistance programs such as SNAP, WIC, or school meals. In his description, food is medicine is medically indicated food used as therapy. He estimated that healthcare already spends $10 billion to $15 billion per year on such food.

The strongest early case, he said, is among patients with immediate, expensive health risks rather than the general population, where prevention may take decades to show a return. Aubrey raised the structural problem: what someone eats today may contribute to chronic disease 30 years later, and by the time Medicare pays the bill, the private insurer that might have funded prevention is no longer responsible.

Mozaffarian answered with the highest-cost patients. He said about 1% of Americans account for 25% of healthcare costs, and 5% account for 50%. These patients often have multiple comorbidities, malnutrition, frailty, disability, mental health issues, substance-use issues, food insecurity, and repeated hospitalizations. For such patients, he said, six months of medically tailored meals—10 meals per week, or roughly half their meals—can keep people out of the hospital and save money.

He cited the Massachusetts experience over three years, saying the savings paid for the program over the course of the program, making it effectively free. If patients were followed for six months after the program ended, he said, benefits persisted, saving thousands of dollars per patient per year.

The model, as confirmed in the exchange, involves meals delivered to someone’s home after hospitalization, surgery, or serious illness, followed by measurement of rehospitalizations and other outcomes. Mozaffarian emphasized local sourcing. Many food-is-medicine programs, including Massachusetts’ Community Servings, source from local farms, prioritize regenerative farms, cook from scratch, and deliver high-quality meals to people too sick to shop or cook.

He presented this as a way to reverse what he called an extractive food system. Since the Green Revolution, he said, the system has extracted from soil, water, air, the Earth, human health, and the economy without accounting for externalities. Food is medicine, in his framing, can create a “virtuous cycle of nourishment”: nourishing the Earth through regenerative practices, nourishing bodies and minds, and reducing healthcare spending. It also creates political interest because hospitals buying from local farms attract the attention of local politicians and members of Congress who care about farms in or near their districts.

The equity dimension became explicit when an audience member asked how to talk about nutrition with people living in poverty who lack access to Starbucks, Trader Joe’s, Whole Foods, or even grocery stores with fresh produce. Jessica Knurick answered that SNAP once had an education arm staffed by registered dietitians who helped people assess what was actually available to them and how to cook on a budget. She said many places face food apartheid or food deserts, and that 13.5% of the country lives in food-insecure households.

Her point was that “eat whole foods” and “eat real food” are not equally actionable for everyone. Nutrition goals must be viewed through access and equity. As with whole grains, the right public-health advice may be less about telling people to achieve the ideal immediately and more about meeting them where they are and increasing what is feasible.

Mozaffarian agreed that the ultimate goal is to change the structure of the food environment from farm to plate so that healthy food is the generally available default. In the meantime, he said food-is-medicine programs often prioritize people facing social and structural challenges: low income, low education, rural residence, and similar barriers.

He told the story of a single mother in Iowa with severe diabetes, low income but not low enough to qualify for SNAP. Her eyesight and kidneys were failing, neuropathy was beginning, and her doctor warned of blindness, dialysis, and possible amputation. Through a charitable produce prescription program, she received about $60 per month in healthy produce for six months, along with nutrition education. Mozaffarian said her hemoglobin A1C fell from about 12 to about 7, she reduced some medications, and her organs stabilized.

When he asked what mattered most, he expected her to name an organ saved. Instead, she said the program changed her relationship with her son. Before, she would pick him up from school, buy him a meal at McDonald’s, skip buying one for herself because it was too expensive, and they would eat separately at home. With the produce prescription, he said, the child asked to go to the grocery store. They chose unfamiliar fruits and vegetables they previously could not risk buying, then cooked and ate together.

For Mozaffarian, that story captured why food is not only medicine. It is also dignity, culture, flavor, family, relationships, and soil.

The old federal alcohol advice—up to one drink per day for women and up to two for men—was removed from the new guidelines, which instead say to consume less. Dariush Mozaffarian said he had heard rumors, without knowing whether they were true, that there had been a move to make the advice stricter, closer to Canada’s recommendation of no more than one drink per day for everyone.

Mozaffarian’s scientific position was clear: in his view, there is no safe level of alcohol. Any amount, he said, incrementally raises risk across outcomes including homicides, depression, accidents, liver disease, and other consequences. He said the idea that moderate drinking, including red wine, protects against heart disease “probably really hasn’t panned out.” One or two drinks per week may not be awful, he said, but zero is better for most people.

He did not know why the guidelines went silent on numeric limits. He speculated that they might have been pulled either because some wanted stricter advice or because others wanted looser advice. Separately, and explicitly as his own “guesstimation,” he said this may be the first dietary-guidelines process not visibly influenced by industry. Historically, he said, USDA has taken committee recommendations behind closed doors and produced final guidelines where industry “fingerprints” could often be seen. This time, he guessed, the document reflected people in a room deciding what they thought was best, with both pros and cons.

Jessica Knurick was less forgiving. Aubrey referred to Dr. Oz saying, in announcing the recommendations, that “we’re human and alcohol is a social lubricant.” Knurick said that if such comments had not been made, she might have wondered whether numeric limits were removed to reduce confusion or to make the advice stricter. But given the stated rationale, she read the omission as clear.

Ty Beal separated the science of risk from the science of communication. He agreed with Mozaffarian that less alcohol is better, but said he did not know whether the current language or the prior numeric language would lead to better outcomes. His interpretation was that the guidelines’ authors wanted a concise document with fewer specific numbers: consume less; if you drink, do not drink too much. He stressed that he was not involved in writing the guidelines, only in the research work that was handed over to the administration.

Jessica Knurick said the misinformation pattern that most frustrates her is the hunt for a single magic bullet or a single additive that explains chronic health problems. Her alternative is dietary pattern. What matters is what someone eats day after day: adequate fiber, vegetables, whole grains, and a diet mostly built from whole foods. More than 90% of Americans, she said, do not consume adequate fiber and vegetables. If that baseline is strong, occasional less-healthy foods are not likely to be “incredibly detrimental.” Diet does not work by completely eliminating one food or always eating another.

Dariush Mozaffarian challenged another common belief: that unhealthy food tastes better. He said events that ask hotels for healthy food often receive egg whites and boiled lentils, as if blandness were the same as healthfulness. His counterexample was a ripe peach with slivered almonds and a little honey or unsweetened whole-fat whipped cream, compared against a KitKat, Doritos, Cheetos, Twinkies, or similar ultra-processed food. His argument was that the real barriers to healthy food are effort, preparation, cost, and opportunity cost—not taste.

He also offered a deliberately provocative example: a healthy natural ice cream can be healthier than bread. Ice cream, he said, contains fat, protein, vitamins, sphingolipids, and phospholipids; it is not a superfood, but it is not terrible. Bread, by contrast, is starch—“glucose, glucose, glucose, glucose, with salt.” The point was not to declare ice cream a health food in general, but to push against deprivation-based thinking and nutrient clichés. A healthy diet, he said, is about celebrating nourishing food rather than restricting oneself.

Ty Beal focused on “plant-based” as a health halo. A plant-based Twinkie, donut, pie, or jelly bean is not equivalent to whole grains, nuts, seeds, legumes, beans, lentils, peas, fruits, and vegetables. In the highly refined versions, he said, phytonutrients are gone, the food matrix is destroyed, and additives are used to create hyper-palatable, energy-dense products. His preferred distinction was not simply plant-based, but plant-rich, minimally processed, whole plant foods.

Allison Aubrey described wearing a continuous glucose monitor for two months and finding the data actionable. One morning, after eating a scone on an empty stomach at work, she saw her blood sugar spike sharply. Drawing on a study she had covered years earlier, she experimented with meal sequencing: starting the day with fiber and protein, and saving refined carbohydrate for the end of a meal or later in the day. She reported a much steadier glucose line and said she felt more energetic after avoiding morning refined carbohydrates such as scones, bread, or plain oatmeal.

Dariush Mozaffarian said he was not surprised. He argued that the biggest change in the food supply since the rise of the obesity epidemic in the 1980s has been refined grains, not sugar. Sugar intake, he said, is not that different over the last 50 years, and sugar-sweetened beverage intake is not that different from 100 years ago to today. The bigger change, in his account, is refined carbohydrate: wheat starch, corn starch, rice starch, and potato starch, along with ultra-processing, additives, contaminants, and other changes.

His own CGM experiment produced two “aha” moments. The biggest glucose spike came from popcorn eaten on an empty stomach during a movie with his family. Popcorn is officially a whole grain, he said, but mostly starch; his glucose rose to about 150 or 160. Had he eaten dinner first, he suspected it would not have had the same effect.

The other experiment involved a dark chocolate macadamia nut cookie with a cappuccino in the afternoon, after lunch. He expected a spike, but saw no change, even when he repeated the test with two cookies. The timing, the prior meal, the cappuccino, and the cookie’s macadamia nuts, chocolate, and butter all blunted the glucose response, he said. He added that 70% dark chocolate does not spike his sugar and called it a health food, though Aubrey noted it might keep someone awake at night.

Mozaffarian cautioned that blood sugar is not the only thing that matters. Steak and hot dogs will not raise blood glucose much, but they may have other effects through inflammation or the microbiome. For glucose specifically, he said, starch is the big problem in the U.S. food supply.

Jessica Knurick was more cautious about CGMs for people without metabolic disease. She said some people are data-oriented and may benefit from the feedback, but for others the data can contribute to disordered eating. She would not recommend that everyone wear one.

Ty Beal agreed. He warned that health-conscious people may overreact to normal fluctuations, treating a rise to 100 or 120 as a reason to overhaul their diet. Blood sugar rises and falls, including in athletes, and that is not necessarily a problem. He saw CGMs as useful tools but urged caution. He also pointed to lower-tech ways to manage glucose response: eat vegetables, fiber, and protein with starches, and go for a walk after meals.