Researchers improved signs of biological aging after only four weeks of dietary changes

Biological age is meant to describe how old the body appears physiologically rather than simply how many years a person has lived. A new study suggests that one widely used estimate of biological age can change surprisingly quickly: after just four weeks on controlled diets, some older adults developed blood and metabolic profiles that looked younger than those of participants eating a diet higher in fat.

The study, published in Aging Cell, analyzed data from 104 adults aged 65 to 75 who had participated in the Nutrition for Healthy Living randomized trial. Rather than asking participants to modify their normal diets on their own, researchers supplied essentially all of their food for four weeks, giving the experiment unusually tight control over what people actually ate.

Participants were randomly assigned to one of four diets. Two were omnivorous, while two were semi-vegetarian, with protein coming mainly from plant sources. Each version was then either relatively high in fat or high in carbohydrates. All four diets provided about 14% of calories from protein.

The high-carbohydrate diets provided about 53% of calories from carbohydrates and 28–29% from fat, while the higher-fat diets provided roughly 37–41% of calories from fat and 41–43% from carbohydrates.

The researchers then asked whether these diets altered a measure called the Klemera–Doubal Method age, or KDMAge.

Unlike epigenetic clocks that examine chemical modifications to DNA, KDMAge combines ordinary physiological measurements that tend to change with age. The researchers used two versions of the calculation. One incorporated seven measures including C-reactive protein, cholesterol, blood pressure, glucose, albumin and kidney- and liver-related markers. A second, broader version incorporated 15 measurements, adding variables such as insulin, HDL cholesterol, triglycerides, waist circumference and body mass index.

The resulting score was compared with each person's chronological age. A positive difference means that someone's physiological profile looks older than expected for their age, while a negative value means that it resembles the profile typically seen in younger people. Previous population studies have linked higher KDMAge values with increased morbidity and mortality, although the measure is not a direct measurement of the biological mechanisms that cause aging.

After four weeks, the clearest difference appeared in participants assigned to the omnivorous, high-carbohydrate diet.

After accounting for their starting values, their physiological-age score was approximately 3.5 years lower using the seven-biomarker calculation and about 4.1 years lower using the broader 15-biomarker calculation compared with participants on the omnivorous high-fat diet. Both differences were statistically significant.

Participants eating the semi-vegetarian high-fat diet showed a similar reduction of about 3.5 years with the seven-marker calculation. The broader calculation also moved in the same direction, by around 3.2 years, although that result narrowly missed the conventional threshold for statistical significance.

The semi-vegetarian high-carbohydrate diet also produced younger physiological-age estimates, by roughly three years compared with the high-fat omnivorous group, but those differences did not reach statistical significance.

The high-fat omnivorous diet was used as the comparison because it most closely resembled what participants had been eating before entering the study. That group showed little meaningful change in its age-related physiological profile over the four weeks. Adjusting the analysis for sex did not substantially change the overall results.

At first glance, a biological-age estimate moving by several years in only one month might sound like rapid age reversal. The researchers explicitly caution against that interpretation.

Many of the measurements used to calculate KDMAge — including glucose, insulin, cholesterol, blood pressure and inflammatory markers — can respond to diet within days or weeks. A composite score built from them can therefore change rapidly even if the underlying processes responsible for long-term aging have not been altered.

In other words, the study shows that age-associated physiology is responsive to diet, not that four weeks of eating differently made participants several years biologically younger.

This distinction is important because biological-age algorithms compress many measurements into a single number expressed in years. A three- or four-year reduction in that number does not mean that four years of cellular aging were removed from the body. It means the combination of measured biomarkers shifted toward values more commonly observed in younger or physiologically healthier people.

The researchers nevertheless argue that the changes were unlikely to reflect only one isolated biomarker. Similar patterns appeared using two different biological-age calculations containing different sets of measurements, including some markers that respond quickly to diet and others that tend to change more slowly.

The diets associated with lower physiological-age scores also shared several characteristics. Compared with the high-fat omnivorous diet, they contained more fibre and more servings of fruit, vegetables and legumes. The researchers suggest that changes in metabolic and inflammatory health could contribute to the observed effect, although the experiment was not designed to identify which foods or nutrients were responsible.

The results therefore do not establish that carbohydrates themselves slow aging, nor that vegetarian diets are inherently anti-aging. In fact, the strongest statistically supported result occurred in the omnivorous high-carbohydrate group, while some effects in the semi-vegetarian groups did not reach statistical significance.

The study instead points to a broader possibility: changing the overall composition and quality of a person's diet can rapidly alter physiological systems that are themselves associated with aging.

Limits of the evidence

The intervention lasted only four weeks, so researchers do not know whether the lower KDMAge scores would persist if participants continued the diets for months or years, or whether they would quickly return toward baseline once normal eating resumed.

The analysis also comes from an existing randomized dietary trial rather than a trial originally designed specifically to test biological aging. The 104 participants were relatively healthy adults aged 65–75, so the results may not apply to younger people, people with significant chronic disease or the broader older population.

KDMAge itself is another limitation. It is a biomarker-derived estimate of age-related physiological status, not a direct measurement of lifespan, cellular damage or the rate at which a person is aging. Its components are influenced by diet, exercise, stress, medication and other short-term factors. The authors therefore describe the measure as a useful proxy rather than evidence that biological age has literally been reversed.

The study also cannot isolate one dietary component as responsible. The diets differed simultaneously in fat-to-carbohydrate ratio, fibre, fruit and vegetable intake, legumes and the balance between animal and plant protein.

What the experiment does show is that a composite measure associated with aging and long-term health can respond within weeks to a controlled dietary intervention. Whether maintaining those changes ultimately reduces age-related disease, preserves physical function or extends healthy lifespan will require much longer trials.