Why “eat less” was never the whole answer
For nearly a century, the most reliable way to make a laboratory animal live longer has been to feed it less. Researchers first showed it in rats in the 1930s, and the finding has since been repeated in yeast, worms, fruit flies, and mice. The trouble is that very few people can actually live that way.
Severe calorie restriction is hard to sustain, and in humans it can bring bone fractures, hormonal disruption, menstrual problems, and heart rhythm disturbances. So scientists have been asking a more hopeful question: can you capture much of the benefit without going hungry?
A new review published this fall in The Innovation Nutrition says the answer is very likely yes. Jintao Huang, Dejian Huang, and Ziyun Wu, of Shanghai Jiao Tong University and the National University of Singapore, pulled together animal research, laboratory work on how cells sense nutrients, and more than thirty large human cohort studies and meta-analyses. They conclude that the quality of your food, and the balance among carbohydrates, protein, fat, and fiber, shapes how you age at least as much as the number of calories you eat.
In this article, I’ll walk you through what they found, how their recommendations shift as you move from young adulthood into midlife and beyond, and where I think you should read their conclusions with a careful eye. I’ll also show you how to put the most solid parts into practice starting this week.
Your cells are reading every meal
Your body was designed with a remarkable set of fuel sensors. Every time you eat, your cells take a reading and decide whether it’s time to build and grow or time to conserve, clean house, and repair. The review focuses on four of these sensing systems, and you should know them by name.
Two of them push toward growth. mTOR (short for mechanistic target of rapamycin) acts like a construction foreman, switching on protein building when amino acids are plentiful. The insulin and IGF-1 pathway carries a similar “times are good, keep growing” message, and it rises with high-glycemic foods and with large amounts of animal protein.
The other two push toward maintenance. AMPK switches on when energy is scarce, telling cells to burn fat and stop storing it. The sirtuins, especially SIRT1, help protect DNA and support healthy mitochondria, the tiny power plants inside your cells.
When the growth signals are turned down and the maintenance signals are turned up, the review explains, cells make more mitochondria, recycle worn-out parts through a cleanup process called autophagy, resist oxidative stress, and guard their genetic material. That shift is a large part of why calorie restriction lengthens life in animals. The encouraging news is that the foods you choose, and not only how much you eat, can move these same dials.
If you or someone you love has faced cancer, you may recognize mTOR and IGF-1. Many tumors exploit these same growth signals, which is one reason the topic of food and aging matters so much to my patients.
Carbohydrates: the kind matters more than the count
Few topics in nutrition stir up more argument than carbohydrates, and this review helps settle some of it. Across the large studies it summarizes, the relationship between carbohydrate intake and survival is shaped like a “U.” People who eat very little carbohydrate and people who eat a great deal both tend to die sooner than people in the middle.
One of the clearest examples comes from the ARIC study, which followed 15,428 American adults aged 45-64 for a median of 25 years. The lowest risk of death showed up at 50-55% of calories from carbohydrate. When the researchers pooled their data with seven other cohorts, covering more than 432,000 people, both low intakes (below 40% of calories) and high intakes (above 70%) carried greater risk than moderate intake.
Here is the part I most want you to notice. In that same analysis, replacing carbohydrate with animal fat and protein was linked to an 18% higher risk of death, while replacing it with plant fat and protein was linked to an 18% lower risk. Study after study in the review tells the same story, from the Nurses’ Health Study to large cohorts in Japan and Korea: low-carbohydrate diets built on meat tend to raise risk, while those built on vegetables, nuts, and whole grains tend to lower it.
Glycemic index matters too. Foods that send blood sugar and insulin surging push the growth dials we just discussed. Low-glycemic eating has extended life in aged mice, and in people it is associated with better odds of healthy aging, though direct proof of longer human lifespan is still lacking. A major series of analyses in The Lancet, built on 135 million person-years of data, rated the evidence for fiber and whole grains as considerably stronger than the evidence for glycemic index alone.
So where should your carbohydrates come from? The authors suggest that roughly 25-30% come from whole grains such as oats, brown rice, and quinoa; 20-25% from legumes such as white beans, lentils, and chickpeas; 15-20% from vegetables; 15-20% from tubers and roots such as sweet potato and yam; and 10-15% from whole fruits. They allow 5-10% from “natural” sweeteners such as honey and maple syrup. Since those are still concentrated sugars, I would treat that last category as optional. Refined carbohydrates and sugary drinks are not on their list at all.
One more point if you live with diabetes or prediabetes. In the Korean cohort the review cites, higher carbohydrate intake was linked to higher mortality among people with diabetes. For you, carbohydrate quality, portion size, and how your own blood sugar responds matter even more.
Protein: how much depends on how old you are
Protein is where this review gets especially interesting, because the right answer seems to change with age. In midlife, more is not better, and the source matters a great deal. Later in life, the picture flips.
The landmark study here was published in Cell Metabolism in 2014. Researchers followed 6,381 American adults aged 50 and older for 18 years. Among those aged 50-65, people who got more than 20% of their calories from protein had a 75% higher overall death rate and roughly four times the risk of dying from cancer, compared with people eating low protein. Those associations largely disappeared when the protein came from plants. Among people over 65, however, higher protein intake was linked to lower cancer and overall mortality.
A second large analysis, published in JAMA Internal Medicine, followed 131,342 nurses and health professionals for up to 32 years. Each additional 3% of calories from plant protein was associated with a 10% lower risk of death. Swapping 3% of calories from processed red meat for plant protein was linked to a 34% lower risk. Fish and poultry looked far more neutral than red and processed meat.
Why would plant protein behave differently? The review points to amino acid makeup. Animal proteins tend to carry more methionine and other sulfur-containing amino acids, which activate mTOR and stimulate growth hormone release. Plant proteins carry less, so they nudge the growth dials more gently.
You may read about dramatic results from restricting single amino acids. Cutting methionine to about 20% of normal extends the lives of rats and mice by up to 30-45%, and restricting isoleucine extends mouse lifespan by 7-33%, depending on sex. In short human trials, though, methionine restriction lowered IGF-1 by only 7-14% and did not measurably change biological age, and long-term safety is unknown. Glycine supplementation has also lengthened rodent lifespans, but it has not been tested for longevity in people. I don’t recommend any of these as do-it-yourself strategies. The gentle reduction in methionine that comes naturally with a plant-forward diet is a very different thing from severe experimental restriction.
After 65, the main danger shifts to sarcopenia, the loss of muscle that leads to frailty and falls. The review cites Japanese data from adults 85 and older in which higher protein intake was linked to lower mortality. If you are in this stage of life, adequate protein becomes protective, and the authors suggest getting it mainly from fish and plant sources.
Fats: the type matters more than the total
For decades, people were told to fear fat in general. The evidence gathered in this review points in a more useful direction: what kind of fat you eat matters far more than how much.
The strongest single study comes from the Nurses’ Health Study and the Health Professionals Follow-up Study, which tracked 126,233 people for up to 32 years. Those eating the most saturated fat had an 8% higher risk of death than those eating the least, while those eating the most polyunsaturated fat had a 19% lower risk and those eating the most monounsaturated fat had an 11% lower risk. Replacing just 5% of daily calories from saturated fat with polyunsaturated fat was associated with a 27% lower risk of death.
The source of the fat also matters. Several cohorts in the review found that monounsaturated fats from plants were protective, while the same type of fat from animal foods was neutral or harmful. Omega-3 fats from walnuts and fatty fish were consistently linked to lower mortality, and trans fats were consistently harmful.
The authors suggest keeping total fat at no more than 30% of your calories. Of that, they place about 60% in a “recommended” tier (vegetable oils, nuts, fatty fish such as salmon, and avocado), about 30% in a “moderate” tier of lean meats, and about 10% in a “limit” tier that includes butter, cheese, lard, beef tallow, palm oil, and margarine.
The review lists olive oil among its recommended fats, and the population data on olive oil are generally favorable. As I’ve written before, though, I think the story on oleic acid, olive oil’s main fat, is more complicated for people focused on cancer, so I encourage you to lean on nuts, seeds, avocado, and fatty fish as your foundation rather than pouring on oil freely.
One animal finding caught my attention. When mice were fed a high-fat diet for life, males lived about 144 fewer days on average, while females showed no significant change. Sex differences like this appear again and again in the review, and they are a reminder that one-size-fits-all advice has limits.
Fiber: the most consistent finding of all
If one thing stands out, it is fiber. The more fiber people eat, the longer they tend to live, and the relationship holds steady across cohorts in the United States, Europe, Japan, and Korea.
Across the meta-analyses the authors summarize, people with the highest fiber intake had about 10-23% lower risk of death than those with the lowest. The large Lancet analysis I mentioned earlier found benefits for death, heart disease, stroke, type 2 diabetes, and colorectal cancer, with the greatest gains at intakes of 25-29 grams a day or more. Fiber from whole grains, cereals, and vegetables showed the strongest protection.
Fiber works largely through the trillions of microbes in your gut. Soluble fibers such as those in oats, beans, and apples are fermented into short-chain fatty acids, including butyrate, which strengthen the gut lining and calm inflammation. In mice, a high-fiber diet reproduced many of the molecular signatures of calorie restriction while the animals ate freely. Butyrate itself has extended healthspan and lifespan in mouse models of accelerated aging.
The authors suggest building your 25-30 grams a day from whole grains (about 30%), legumes (about 20%), vegetables (about 20%), fruits eaten with their skins (about 15%), nuts and seeds (about 10%), and tubers such as sweet potato (about 5%). A cup of cooked black beans or lentils alone gets you a long way there.
If your diet is low in fiber now, increase it gradually over a few weeks and drink plenty of water. Your gut microbes need time to adjust, and a sudden jump can leave you gassy and uncomfortable.
Where keto fits
Many of my readers ask about the ketogenic diet, so I was glad to see the authors address it directly. They make an important distinction that often gets lost.
A true therapeutic ketogenic diet, first developed in the 1920s for hard-to-control epilepsy, limits carbohydrate to 5% of calories or less (about 25-50 grams a day), with roughly 75% from fat and 20% from protein. Its goal is a steady state of nutritional ketosis. Most popular “low-carb” plans allow 50-130 grams of carbohydrate a day and rarely produce true ketosis. That difference matters, because most human studies of low-carb eating reflect the popular version, while the animal studies showing longer life used the strict one.
In mice, long-term ketogenic feeding has extended median lifespan. But a 2024 study in Science Advances found that continuous ketogenic feeding in mice increased senescent cells, the worn-out “zombie” cells that drive inflammation, in several organs, including the heart and kidney. An intermittent schedule with planned breaks did not produce the same buildup.
Overall, the review notes that very-low-carbohydrate diets in human cohorts, especially those built on animal foods, are linked to neutral or modestly higher mortality. Long-term adherence is poor, and stopping abruptly can cause a metabolic rebound. The authors also note that robust support for keto in cancer, Alzheimer’s, and Parkinson’s disease remains limited. Their conclusion, which matches my own, is that keto is best used as a time-limited therapeutic tool under medical supervision, not as a lifelong longevity plan.
Your plate across the decades
The most practical part of the review is a set of age-adjusted targets. The authors are candid that the evidence behind them varies, and they label each age group accordingly. All percentages below refer to daily calories.
For young adults (19-35), they suggest 50-55% of calories from carbohydrates, built mainly on whole grains; 10-20% from protein, with less red meat; 15-25% from mostly healthy fats; and 20-40 grams of fiber a day. They rate the evidence for this group as high.
For midlife (36-64), the advice shifts toward quieting growth signals. Carbohydrate stays at 50-55%, mostly whole grains. Protein drops to 10-15% and becomes more plant-based. Fat falls to 10-20%, and fiber rises to 40-60 grams a day. The authors single out ages 40-50 as a critical window because cardiometabolic risk, early muscle loss, and epigenetic aging tend to accelerate in that decade.
For older adults (65 and up), protein rises again to 15-20%, mostly from plants and fish, to protect muscle. Carbohydrate is set at 50-60%, fat at 10-15%, and fiber at 20-40 grams a day. The authors rate this evidence as moderate, because it rests mainly on frailty and muscle outcomes rather than on longevity trials.
Their companion “longevity pyramid” is different from the food pyramids you may remember. Its foundation is not a food group at all. It is hydration, adequate sleep, regular physical activity, stress management, consistent daily rhythms, and social connection. Above that come whole grains and legumes, then a wide variety of vegetables, then polyphenol-rich fruits, then high-quality protein at about 0.8-1.2 grams per kilogram of body weight, with healthy fats at the top. I appreciate that the authors put the way you live underneath everything else. Food works best on top of good sleep, movement, and close relationships.
Making it personal
The second half of the review argues that the future of longevity nutrition is precision: fitting the diet to the person, not the person to the diet. Several threads stand out.
Your blood sugar response is your own. In the PREDICT 1 study, published in Nature Medicine, researchers fed 1,002 adults identical standardized meals and found enormous differences in how their blood sugar, insulin, and triglycerides responded, even between identical twins. Much of that variation came from modifiable factors such as the gut microbiome, sleep, physical activity, and what else was eaten with the meal. A continuous glucose monitor can show you how your own body handles the foods you eat most often.
Your gut microbes are partners in the process. The review notes that people who live to 100 tend to have microbiomes rich in butyrate-producing bacteria and lower in inflammatory species. Much of the benefit attributed to fiber, plant protein, and polyphenols appears to run through these microbes.
When you eat matters. Time-restricted eating, in which you eat all your food within a consistent 6-10 hour window, aligns meals with your body’s daily rhythms. In a small 12-week pilot study of 19 people with metabolic syndrome, a 10-hour eating window led to weight loss, lower blood pressure, and lower atherogenic cholesterol, even in people already taking statins and blood pressure medication. The review also notes early evidence that finishing your last meal earlier in the day, before about 4-6 pm, may bring more metabolic benefit than a late eating window. Researchers have not yet tested long-term effects on lifespan in people.
Ultra-processed foods deserve their own warning. These industrial products now supply more than half the calories in many wealthy countries. An umbrella review in The BMJ, covering nearly 10 million people, linked greater intake to higher risk of 32 health problems, with convincing evidence for cardiovascular death and highly suggestive evidence for a 21% higher risk of death from any cause. The review’s authors recommend limiting these foods at every stage of life, and I agree.
Men and women may need different plans. In study after study, animals responded differently to the same diet depending on sex. High-fat feeding shortened life in male mice, but not females; isoleucine restriction helped males more, and glycine helped females more. The authors argue that sex-specific recommendations should become standard, though human data on this question are still thin.
Reading the review with a careful eye
This is a thoughtful and useful paper, but you should understand its limits so you can use it wisely.
First, nearly all of the human evidence is observational. Large cohort studies can show that people who eat a certain way tend to live longer, but they cannot prove the food is the reason. People who eat more beans and vegetables also tend to smoke less, exercise more, and see their doctors more often. The researchers adjust for these factors as well as they can, but no adjustment is perfect.
Second, effects that look dramatic in animals usually shrink in people. Inulin, a type of fiber, extended the lives of male fruit flies by more than 50%, while human trials have shown only modest improvements in blood sugar, lipids, and microbiome diversity, with no longevity data. The authors themselves note that the longer an animal naturally lives, the smaller the lifespan gain from any diet change tends to be.
Third, a few of the specific numbers deserve caution. The 40-60 grams of fiber a day suggested for midlife is, by the authors’ own description, a hypothesis drawn from animal research rather than established guidance; current guidelines sit at 25-30 grams. Their protein chart also sits a little uneasily beside their own text. The text stresses plant protein, yet by my count the chart draws roughly 60% of protein from animal foods such as eggs, fish, shellfish, poultry, pork, beef, and dairy. If you follow the evidence the authors summarize, I would tilt that balance further toward beans, lentils, soy, nuts, and whole grains.
Fourth, this review focuses on healthy aging, not active cancer treatment. If you are in treatment, fighting weight loss, or losing muscle, your protein and calorie needs may be considerably higher than these targets, and your plan should be tailored with your care team.
Finally, the authors are refreshingly honest about what comes next. They call for trials lasting 3-5 years that measure biological age clocks such as GrimAge, PhenoAge, and DunedinPACE, and they describe their own frameworks as testable hypotheses rather than settled rules.
Putting it into practice
You don’t need to count every gram to benefit from this research. The strongest, most consistent findings point to a handful of habits you can start this week.
Build your meals on whole plant foods. Make whole grains, beans, lentils, vegetables, and whole fruits the base of your plate, and let refined flour, added sugar, and sugary drinks fade out. This alone moves you toward the low-glycemic, high-fiber pattern that keeps showing up in the data.
Choose your protein thoughtfully. In midlife, aim for moderate protein that comes mostly from plants such as beans, lentils, soy, nuts, and seeds, with fish as your main animal source and processed red meat kept to a minimum. After 65, make sure you are getting enough protein to protect your muscles, and pair it with strength training.
Trade saturated fat for unsaturated fat. Walnuts, almonds, seeds, avocado, and fatty fish such as salmon are good everyday sources. Keep butter, fatty meats, and fried foods for rare occasions.
Work your way up to at least 25-30 grams of fiber a day. Add one extra serving of beans or whole grains at a time, give your gut a few weeks to adjust, and drink plenty of water.
Give your eating a daily rhythm. Try keeping your meals within a consistent 10-hour window and finishing dinner earlier in the evening. If you take insulin or medications that lower blood sugar, talk with your doctor before changing your eating schedule.
Cut back on ultra-processed foods. If a product has a long list of ingredients you would never keep in your own kitchen, choose something simpler.
And measure rather than guess. The review’s own precision-nutrition framework begins with baseline testing, including fasting glucose, insulin, lipid particles, and inflammatory markers. A comprehensive metabolic blood panel, repeated over time, shows you whether your changes are actually moving your numbers in the right direction.
The bodies we have been given are fearfully and wonderfully made, and caring for them well is good stewardship. This review is a welcome reminder that the most powerful longevity tool is not a pill or an extreme diet. It is the steady, everyday choice to eat real food, mostly plants, in a rhythm your body was designed for.
References
- Berry SE, Valdes AM, Drew DA, et al. Human postprandial responses to food and potential for precision nutrition. Nat Med. 2020;26(6):964-973.
- Huang J, Huang D, Wu Z. Beyond dietary restriction: Towards next-generation precision diets for healthy aging. The Innovation Nutrition. 2026;1(3):100034.
- Lane MM, Gamage E, Du S, et al. Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. BMJ. 2024;384:e077310.
- Levine ME, Suarez JA, Brandhorst S, et al. Low protein intake is associated with a major reduction in IGF-1, cancer, and overall mortality in the 65 and younger but not older population. Cell Metab. 2014;19(3):407-417.
- Reynolds A, Mann J, Cummings J, Winter N, Mete E, Te Morenga L. Carbohydrate quality and human health: a series of systematic reviews and meta-analyses. Lancet. 2019;393(10170):434-445.
- Seidelmann SB, Claggett B, Cheng S, et al. Dietary carbohydrate intake and mortality: a prospective cohort study and meta-analysis. Lancet Public Health. 2018;3(9):e419-e428.
- Song M, Fung TT, Hu FB, et al. Association of animal and plant protein intake with all-cause and cause-specific mortality. JAMA Intern Med. 2016;176(10):1453-1463.
- Wang DD, Li Y, Chiuve SE, et al. Association of specific dietary fats with total and cause-specific mortality. JAMA Intern Med. 2016;176(8):1134-1145.
- Wei SJ, Schell JR, Chocron ES, et al. Ketogenic diet induces p53-dependent cellular senescence in multiple organs. Sci Adv. 2024;10(20):eado1463.
- Wilkinson MJ, Manoogian ENC, Zadourian A, et al. Ten-hour time-restricted eating reduces weight, blood pressure, and atherogenic lipids in patients with metabolic syndrome. Cell Metab. 2020;31(1):92-104.e5.
