Can eating a little less actually slow how fast you age? In the CALERIE trial, healthy adults who cut their calories by about 12% for two years slowed their “pace of ageing” – measured by a DNA-methylation test called DunedinPACE – by roughly 2–3% compared with people who ate as usual. It is one of the few randomised human trials of an anti-ageing intervention, and its lessons are more useful – and more modest – than most headlines suggest.
Case study at a glance
- Study: CALERIE Phase 2, a randomised controlled trial at three US research centres.
- Who: about 220 healthy adults aged 21–50 without obesity.
- What: two years of calorie restriction (target 25%, achieved about 12%) vs. normal eating.
- Key result: DunedinPACE slowed by about 2–3%; two other “epigenetic clocks” did not change significantly.
- Bottom line: a real but small signal – and a reminder that how you measure ageing matters.
Why this study matters
Calorie restriction is the most studied way to extend lifespan in animals, from yeast to mice. But animal results often fail in humans, and we can’t wait 50 years to see whether a diet makes people live longer. CALERIE (Comprehensive Assessment of Long-term Effects of Reducing Intake of Energy) was designed to test, in a randomised way, whether calorie restriction changes markers of ageing in healthy, non-obese people.
What the participants did
Participants were randomly assigned in a 2:1 ratio to calorie restriction or to continue their usual diet. The restriction group received intensive support – dietitians, meal plans, group sessions – with a goal of eating 25% fewer calories than they needed to maintain weight. In practice, most people managed about 12% less, which still led to an average weight loss of around 10% in the first year.
That gap between target and reality is a lesson in itself: even motivated volunteers with professional support found strict restriction hard to sustain.
How “pace of ageing” was measured
Researchers analysed blood samples taken at the start, after one year and after two years, looking at DNA methylation – chemical tags on DNA that change with age. From these, they calculated three “epigenetic clocks”:
| Measure | What it estimates | Result in CALERIE |
|---|---|---|
| DunedinPACE | How fast you are ageing right now (like a speedometer) | Slowed by about 2–3% vs. control |
| PhenoAge | Biological age based on markers linked to disease and death | No significant change |
| GrimAge | Biological age tuned to predict mortality | No significant change |
DunedinPACE was developed by following the same people for decades and tracking how quickly their organs declined. It is designed to capture the rate of change, which may be why it picked up an effect after only two years when the “age” clocks did not.
What did the 2–3% slowdown mean?
A 2–3% slower pace of ageing sounds tiny. The researchers noted that, in other studies, differences of this size in DunedinPACE are associated with roughly a 10–15% lower risk of early death – comparable to the effect of some well-established health interventions. That is an extrapolation from observational data, not a measured outcome of the trial: CALERIE did not follow people long enough to see differences in disease or death.
Earlier analyses of the same trial also found improvements in cardiometabolic risk factors in the calorie-restriction group, including LDL cholesterol, blood pressure and insulin sensitivity.
What this case study does NOT show
- That calorie restriction makes people live longer – lifespan was not measured.
- That “biological age tests” sold to consumers are precise – two of the three clocks showed no change in the same blood samples.
- That more restriction is better – participants achieved half the target, and very low intakes carry risks such as muscle and bone loss.
- That the results apply to older adults, people with obesity or people with medical conditions – they were not studied here.
Practical lessons for you
- Moderate beats extreme. A sustainable reduction – such as stopping late-evening snacking or using a 12–14 hour overnight fast – is more realistic than a strict 25% cut. See our 16:8 beginner’s guide.
- Protect muscle. If you eat less, keep protein up and add strength training. Use our protein calculator.
- Be careful with test results. If you buy a biological age test, remember that different clocks can give different answers from the same blood. Read our guides to home testing.
- Combine pillars. Eating patterns work best alongside daily movement and good sleep.
Calorie restriction is not for everyone. It is not appropriate during pregnancy, for people who are underweight, under 18, frail, or have a history of eating disorders. Talk to your doctor before making major changes to how much you eat.
What to do tomorrow
Instead of counting calories, set one boundary: no food after dinner. That single habit typically removes a few hundred calories of evening snacking – a realistic version of what CALERIE participants achieved.
Frequently asked questions
What was the CALERIE trial?
A two-year randomised trial in about 220 healthy, non-obese adults testing whether calorie restriction slows markers of ageing.
Did calorie restriction slow ageing?
It slowed the DunedinPACE measure of pace of ageing by about 2–3%, but two other epigenetic clocks did not change significantly, and lifespan was not measured.
Should I restrict calories?
Strict restriction is hard and not right for everyone. A moderate approach, such as a 12–14 hour overnight fast with enough protein, is more realistic.
Turn this into a habit in 30 days
The 30-Day Longevity Reset turns the research into 15 minutes a day: eating window, movement, strength and calm – one simple step at a time. Join the waitlist for first access and a founding-member price.
Sources
- Waziry R et al. Effect of long-term caloric restriction on DNA methylation measures of biological aging in healthy adults from the CALERIE trial. Nature Aging, 2023. PubMed
- Kraus WE et al. 2 years of calorie restriction and cardiometabolic risk (CALERIE): exploratory outcomes of a multicentre, phase 2, randomised controlled trial. Lancet Diabetes Endocrinol, 2019. PubMed
- Belsky DW et al. DunedinPACE, a DNA methylation biomarker of the pace of aging. eLife, 2022. PubMed
This article is for education only and is not medical advice. Read our medical disclaimer.
