Caloric Restriction: Lessons From the CALERIE Trial

Caloric restriction (CR) is the oldest and most reliable intervention in aging biology: feed an animal fewer calories without starving it, and across many species, it tends to live longer. The obvious question is whether the same holds in humans. We can’t run a multi-decade randomized lifespan trial on people, so the field has had to settle for the next best thing. CALERIE (Comprehensive Assessment of Long-Term Effects of Reducing Intake of Energy) is the most rigorous attempt yet.

What CALERIE was

CALERIE Phase 2 was a multicentre, phase 2, randomised controlled trial run at three US sites. It enrolled 218 healthy, non-obese adults aged 21–50 and assigned them 2:1 to a 25% calorie-restriction target or an ad libitum control diet for two years, with results published in The Lancet Diabetes & Endocrinology in 2019. In practice, sustained restriction is hard: the restriction group achieved a mean reduction of only 11.9% in calorie intake, a useful reminder that adherence, not theory, sets the ceiling on what these studies can show. They sustained a mean weight loss of 7.5 kg.

Laboratory test tubes and chemistry equipment used for clinical trial sample analysis

The measured outcomes were biomarkers, not lifespan. The trial reported significant improvements across multiple cardiometabolic risk factors (including blood pressure, lipids, and insulin sensitivity markers) in young, non-obese adults. A separate 2023 analysis in Nature Aging applied DNA-methylation “clocks” to CALERIE samples and found that restriction slowed the pace of aging measured by DunedinPACE by roughly 2–3%, while the static-age clocks PhenoAge and GrimAge showed no significant change. The authors noted that a 2–3% slower pace, in other cohorts, corresponds to a 10–15% lower mortality risk. But that is an extrapolation, not a measured lifespan effect.

3D render of a DNA double helix representing the epigenetic clocks used to measure biological aging

The honest summary: CALERIE shows that moderate, sustained calorie reduction shifts human aging-related biomarkers in a healthy direction. It does not show that it extends human lifespan, and it was never designed to.

What CALERIE measured	Result
Calorie reduction achieved (target 25%)	11.9% mean
Weight change	−7.5 kg
Cardiometabolic risk factors	Significantly improved
Pace of aging (DunedinPACE)	2–3% slower vs control
Static-age clocks (PhenoAge, GrimAge)	No significant change

A bowl of muesli with oats, grains, and dried fruit representing a balanced, moderate-calorie meal

What’s worth holding onto

The cardiometabolic improvements were real, and overlap with what you’d expect from losing excess weight.
Adherence was the binding constraint: the group hit 11.9%, not the 25% target.
The biological-aging signal came only from the change-sensitive DunedinPACE clock, not the static-age clocks, a distinction worth keeping.

The takeaway

CALERIE is the strongest human CR evidence we have, and it points in a plausible, encouraging direction without proving the headline claim. Severe, lifelong restriction in lean people carries real risks (muscle and bone loss, cold intolerance, and quality-of-life costs) that the trial’s modest benefits don’t clearly justify for most. For now, the defensible reading is narrow: avoiding chronic overconsumption is good for metabolic health, and that’s a more grounded goal than chasing a lifespan effect the data hasn’t earned.