The promise of an “anti-inflammatory diet” sounds almost too neat. Change what is on your plate, lower inflammation in your blood, and perhaps grow old more slowly. The first step is biologically plausible. The last is a much larger claim. Somewhere between them lies the question that matters: when a diet changes a biomarker, has it changed a person’s future?
There is good evidence that dietary patterns influence cardiovascular and metabolic health. Randomised trials also show changes in some inflammatory markers. What is less clear is how much of those benefits comes through inflammation, rather than through blood pressure, blood lipids, body weight and other pathways. A diet can be worthwhile without passing a test of biological age.
The attractive shortcut: treating inflammation as an ageing clock
Low-grade inflammation often increases with age, a pattern known as inflammaging. Researchers study markers including C-reactive protein (CRP) and interleukin-6 (IL-6), which are associated with several age-related conditions. But these markers are not interchangeable, and their concentrations can change with illness, body fat, medication and many other factors.
Ageing itself involves several interacting processes, from cellular senescence and altered nutrient sensing to changes in the immune system. Inflammation is part of this biology, not an independent master dial that can simply be turned down with a shopping list.
Suppose someone’s CRP falls after a dietary intervention. That is a measurable biological result. It does not, by itself, tell us whether the person will have fewer heart attacks, remain mobile at 85 or live longer. Those outcomes require different studies, often conducted over years rather than weeks.
The central distinction
A better blood test is not the same as slower ageing. A fall in CRP can be real and useful to investigate. It is not a validated conversion from milligrams per litre into extra years of healthy life.
The Mediterranean trials: an instructive mixed result
The Mediterranean dietary pattern has one advantage over lists of supposed anti-ageing superfoods: it has been tested repeatedly. Its common elements include vegetables, fruit, legumes, whole grains, olive oil, nuts and fish, with less processed meat and fewer heavily refined foods. It is a pattern, not a fixed menu or a supplement regime.
A 2026 systematic review pooled 33 randomised controlled trials involving 3,476 participants. Compared with control diets, Mediterranean-style interventions significantly reduced high-sensitivity CRP, IL-6 and IL-17. Yet ordinary CRP, tumour necrosis factor-alpha, IL-10 and total antioxidant capacity did not show statistically significant overall changes.
That discrepancy is more interesting than a tidy claim that the diet “fights inflammation”. Different biomarkers reflect different parts of the immune response; trials vary in participants, duration and what the comparison group eats. A positive result for one marker does not automatically extend to all the others. Even the meta-analysis authors called for better-quality studies.
An earlier review in older adults also linked stronger adherence to a Mediterranean pattern with lower CRP, although much of that evidence was observational. It reinforces interest in the pattern without settling the question of cause and effect.
A more demanding test: what happens to patients?
To find out whether a diet matters, we should look beyond blood samples. CORDIOPREV, a randomised trial in Spain, followed 1,002 people with established coronary heart disease assigned to either a Mediterranean or a low-fat diet. Over the trial, the primary composite cardiovascular endpoint occurred in 87 people in the Mediterranean group versus 111 in the low-fat group. Adjusted hazard ratios in the published analyses ranged from approximately 0.72 to 0.75 in favour of the Mediterranean diet.
A result that matters
87 versus 111 cardiovascular events. In CORDIOPREV, the Mediterranean-diet group experienced fewer major cardiovascular events than the low-fat group. The trial studied people who already had coronary disease. It did not measure whether their intrinsic rate of ageing slowed.
The result has a practical meaning that a cytokine measurement cannot provide. It concerns actual cardiovascular events. Yet it comes with limits: most participants were men, all had existing coronary disease, and the trial compared two whole diets. It cannot isolate the precise proportion of benefit attributable to lower inflammation.
The much larger PREDIMED trial also reported fewer major cardiovascular events in Mediterranean-diet groups, but its original report was withdrawn and reanalysed after deviations from random assignment came to light. The revised results remained favourable. Its history is a useful reminder that even famous nutrition trials require methodological scrutiny.
We can therefore say something stronger than “olive oil may lower an inflammation marker”: certain Mediterranean-style interventions have improved important cardiovascular outcomes in tested populations. We still cannot translate those improvements into a universal claim about slower biological ageing.
Two diets, both meeting healthy-eating guidelines
Food processing offers a second way to test easy assumptions. If people follow a healthy dietary guideline, does it matter how much of the food is industrially formulated? A 2025 randomised crossover trial published in Nature Medicine put that question to a small but unusually direct test.
Fifty adults were included in the intention-to-treat analysis. Each eight-week dietary period followed the UK’s Eatwell Guide, but one diet emphasised minimally processed foods and the other ultra-processed foods. Participants could eat according to appetite. Mean body-weight change was −2.06% on the minimally processed diet and −1.05% on the ultra-processed diet, a between-diet difference of about one percentage point over eight weeks (P = 0.024).
Unexpected result
Two guideline-compliant diets produced different weight changes. In the 2025 crossover trial, participants lost weight on both diets, but roughly twice as much, in percentage terms, on minimally processed foods. This was a short weight-loss study, not a longevity experiment.
The finding deserves attention, not mythology. Participants began with overweight or obesity and high habitual ultra-processed-food intake. The trial was modest in size and duration. Whole menus differed in more than a single chemical property, including factors that may influence appetite and how easily people consume energy. The study does not prove that processing itself was the isolated cause or that one group will live longer.
Observational evidence reaches further in time, but offers less control over confounding. A 2025 meta-analysis of 18 prospective cohorts, 1,148,387 participants and 173,107 deaths reported an association between each additional 10% of diet classified as ultra-processed and a 10% higher relative risk of all-cause mortality (hazard ratio 1.10; 95% confidence interval 1.04–1.16).
That is not a prediction that adding one packaged snack will raise an individual’s risk by 10%. The estimate comes from populations observed over time. Heterogeneity was very high (I² = 91% for the incremental analysis); studies also differed in whether they measured food shares by energy or weight. Diet quality, income, smoking, health status and other behaviours are difficult to disentangle completely.
The sensible reading of these two kinds of evidence is cumulative. They strengthen the case for diets built largely from minimally processed foods, while leaving open which features of processing, formulation and food environment matter most. A carton of plain yoghurt and a sugary dessert are not nutritionally identical merely because both come in packaging.
The pattern that matters more than any superfood
The most practical conclusion is almost disappointingly ordinary. Build meals around vegetables and fruit, beans or lentils, whole grains, nuts, olive oil and other unsaturated-fat sources, with fish or other suitable proteins according to preference and need. The foods work as a replacement pattern: beans may displace processed meat; whole grains may replace refined starches; olive oil may replace some saturated fat.
This last point changes how we interpret almost every nutrition headline. “Less saturated fat” is incomplete advice until we ask what takes its place. Replacing it with unsaturated fat can improve cardiovascular risk factors; replacing it with refined carbohydrate does not imply the same benefit. Likewise, simply adding olive oil, nuts or berries to an already excessive calorie intake is not guaranteed to improve health.
A useful pattern should also be adequate, not merely restrictive. Healthy ageing depends partly on maintaining muscle, strength and independence. Protein and sufficient energy matter here, especially in older adults. ESPEN’s geriatric nutrition guideline recommends at least 1 g of protein per kilogram of body weight daily for older people, with individual adjustment. The recommendation concerns nutritional adequacy and function, not an “anti-inflammatory” protein target.
Exercise, particularly resistance training, also matters for preserving muscle. No fruit, supplement or inflammation score can substitute for the broader problem of remaining able to climb stairs, carry groceries or recover from illness.
What is worth limiting?
The strongest everyday priorities are to reduce habitual reliance on sugar-sweetened drinks and processed meat, avoid industrial trans fats where they still occur, and pay attention to the overall nutritional quality and energy density of ultra-processed products. This is more defensible than a list that labels every food containing sugar, saturated fat or an unfamiliar ingredient “inflammatory”.
Extra-virgin olive oil, nuts and oily fish fit established dietary patterns. But they are not interchangeable treatments, and most trials test combinations of foods. It is difficult to assign a whole-diet benefit to a single ingredient.
The false villains: dried fruit, smoothies and dairy
Some familiar foods acquire strange reputations when nutrition is reduced to inflammatory labels. Take dried fruit. A date or raisin has less water than its fresh counterpart, so calories and sugars are more concentrated per gram. That can matter for portions and blood-glucose management. It does not establish that dried fruit inherently drives chronic inflammation.
Smoothies have a similar problem of definition. A drink made from whole fruit, vegetables and plain yoghurt is not nutritionally identical to filtered juice or a sugar-heavy commercial blend. Blending alters texture and may make energy easier to consume quickly, but it does not make every smoothie harmful. Compare the ingredients and the role it plays in the whole diet.
Dairy products are heterogeneous too. Plain yoghurt, cheese, milk, cream and sweetened desserts should not be treated as one exposure. Palm oil, meanwhile, contains saturated fat but is not chemically equivalent to industrial trans fat. These differences matter when choosing replacements, not just when assigning labels.
The general lesson is to resist false precision. A single food rarely determines a person’s long-term inflammatory state, still less their future lifespan. What is habitual, and what that food displaces, usually tells us more.
What a practical healthy-ageing plate looks like
Start with a generous portion of vegetables. Add beans, lentils, fish, eggs, dairy or another protein source suited to your needs. Include whole grains or other minimally processed starches as appetite and energy requirements allow. Use olive oil, nuts and seeds as sources of unsaturated fat. Fruit can be a dessert, snack or part of a meal.
You need not optimise every lunch. A regular, affordable way of eating is more useful than an expensive anti-ageing ritual that lasts two weeks.
Portions still count. For example, 100 g of nuts (about 3.5 oz) can supply roughly 600–700 kcal depending on the variety. A smaller handful of about 20–30 g (0.7–1.1 oz) is often a more convenient snack, although people with higher energy requirements may need more. That is portion guidance, not evidence that larger amounts cause inflammation.
For someone who already eats well, the marginal benefit of making the menu still more elaborate may be small. The bigger gain may come from consistency, sufficient protein, manageable energy intake and replacing foods that dominate a less balanced diet.
Has any dietary experiment shown slower biological ageing?
The question is not entirely beyond experimental reach. In a secondary analysis of the CALERIE randomised trial, 220 adults without obesity were assigned to calorie restriction or usual eating for two years. One DNA-methylation algorithm, DunedinPACE, suggested a small slowing of the measured pace of ageing. Two other widely used epigenetic-age measures, PhenoAge and GrimAge, did not show statistically significant changes.
This was a calorie-restriction experiment, not a test of a Mediterranean or explicitly anti-inflammatory diet. It was also a study of biomarkers rather than deaths or decades of preserved function. The mixed results are scientifically informative precisely because the different measures did not all agree.
There is therefore an important gap between modifying inflammation, changing a molecular ageing measure and demonstrating that people remain healthier for longer. Closing that gap requires longer, well-designed studies with outcomes that matter in everyday life.
An evidence ladder for claims about food and ageing
When reading a headline about foods that “slow ageing”, it helps to ask which rung of the following ladder the study actually reached. None is worthless; each answers a different question.
| What researchers measure | What the finding can support | What it cannot establish by itself |
|---|---|---|
| Inflammatory markers: hs-CRP, IL-6 | A measurable biological response to the intervention | Fewer diseases or slower ageing |
| Weight, blood lipids, glucose and blood pressure | Changes in established risk factors | That inflammation caused the change |
| Heart attack, stroke or other clinical events | A direct health benefit in the population and comparison studied | Extension of lifespan in everyone, or a lower intrinsic ageing rate |
| DNA-methylation ageing measures | Changes in particular biomarker algorithms | Proven extra years of healthy life or survival |
Mike’s Balance evidence map: a qualitative synthesis of the studies cited here, not a validated scoring system. “Evidence of effect” and “evidence of longer life” are separate thresholds.
What the evidence cannot yet say
Lower CRP is not a longevity treatment. No trial discussed here shows that reducing CRP with a particular “anti-inflammatory diet” extends human lifespan. The most compelling support for a good diet comes from cardiovascular outcomes, other clinical risk factors and long-term dietary patterns, not a claim to have reversed biological age.
The bottom line: better odds, not a younger clock
An anti-inflammatory diet is an appealing name for a sensible dietary pattern. But the name implies a causal chain that has not been fully demonstrated. Eating differently may alter selected immune markers; some dietary interventions improve important cardiovascular outcomes; neither result automatically means that the underlying pace of ageing has been slowed.
That does not make the advice weaker. It makes it more useful. Rather than buying a promise to become biologically younger, we can choose a way of eating supported by outcomes that matter: fewer avoidable cardiometabolic problems, enough nourishment to preserve function, and the best possible chance of remaining well as the years accumulate.
If there is a quiet surprise here, it is that the strongest evidence for healthy ageing does not require a list of forbidden foods or a heroic dietary ritual. It requires paying attention to the whole pattern, the quality of the comparison and what a study actually measured.
References and further reading
1. Mediterranean Diet Reduces Inflammation in Adults: A Systematic Review and Meta-analysis of Randomized Controlled Trials. Nutrition Reviews. 2026.
6. CORDIOPREV: Long-term secondary prevention of cardiovascular disease with a Mediterranean diet and a low-fat diet. The Lancet. 2022.
7. PREDIMED: Primary Prevention of Cardiovascular Disease with a Mediterranean Diet Supplemented with Extra-Virgin Olive Oil or Nuts. Revised analysis. New England Journal of Medicine. 2018.
8. Dicken et al. Ultraprocessed or minimally processed diets following healthy dietary guidelines on weight and cardiometabolic health. Nature Medicine. 2025.
9. Effect of long-term caloric restriction on DNA methylation measures of biological aging in healthy adults from the CALERIE trial. Nature Aging. 2023.
10. ESPEN practical guideline: Clinical nutrition and hydration in geriatrics. Clinical Nutrition. 2022.