Reviews are graded A to D against the published methodology. No claim in this journal currently holds grade A on an interventional basis, and one prognostic claim does. Grades attach to the specific claim stated at the top of each review, so the same compound may carry different grades for different assertions about it.
Several of these reviews turn on the difference between a marker that predicts an outcome and a marker that causes one. The same distinction decides what a biological age test result can support, which is set out with the test types, the costs and the questions to put to a provider before paying.
Several of these compounds are sold as supplements, and a supplement on a British shelf has not been assessed for whether it works. What the rules do check before sale, who enforces them and which health claims are allowed is set out in our explainer on whether supplements are regulated in the UK.
- 01
NAD precursors: nicotinamide mononucleotide and nicotinamide riboside
Raising blood NAD metabolites is well demonstrated. Changing the trajectory of human ageing is not the same claim, and it is the one the supplement market makes.
Grade D · Intervention claim - 02
Metformin for healthspan
The best characterised drug in the field, and still without a trial that tests the claim being made for it. The observational signal is real and so are the reasons to distrust it.
Grade C · Intervention claim - 03
Rapamycin and mTOR
The most convincing animal lifespan data in the field, attached to an immunosuppressant with real risks and no long human ageing trial. Both halves of that sentence matter.
Grade D · Intervention claim - 04
Senolytics: the current state
Clearing senescent cells is one of the better ideas in ageing biology. Whether it can be done safely and usefully in a person is an open question with early answers only.
Grade D · Intervention claim - 05
VO2 max as a longevity marker
The single measurement most consistently associated with how long people live. That is a prognostic claim, and it should not be confused with proof that raising it causes the benefit.
Grade A · Prognostic claim - 06
Resistance training and mortality
Strong randomised evidence for function, strength and falls. Weaker observational evidence for mortality, resting on self-reported exposure and an odd shape at the top of the range.
Grade B · Prognostic claim - 07
Sleep architecture and ageing
Sleep changes measurably with age and poor sleep tracks poor outcomes. Which way the causal arrow points, and whether treating sleep changes anything downstream, is much less clear.
Grade C · Prognostic claim - 08
Time-restricted eating
One of the few interventions here with genuine randomised human evidence. The trials are short, and when calories are matched the additional effect of the window is small.
Grade C · Intervention claim - 09
Calorie restriction in humans
Calorie restriction has a genuine two-year randomised human trial behind it. Weight and several risk markers moved; lifespan, disease and function were never measured, and bone and lean mass losses were reported.
Grade C · Intervention claim - 10
Absolute and relative risk in longevity research
The same result can be reported as a third fewer events or as two people in a hundred. Which figure a study leads with decides how large its finding sounds, and the gap is where most healthspan headlines are made.
Method · How to read a claim