Longevity has become a market, and markets need products. That has produced a great deal of enthusiasm for compounds that extended lifespan in organisms with a three-week lifespan and no human data at all.
The interventions that do hold up in humans are known, unglamorous, and mostly cannot be sold in a bottle.
Cardiorespiratory fitness
VO2 max is among the strongest predictors of all-cause mortality ever measured — stronger than smoking, diabetes or hypertension in several large cohorts. The step from the bottom quartile to merely below-average is associated with the largest reduction in risk of anything on this list.
- Most of the weekly volume at an easy conversational pace
- A small amount of genuinely hard interval work, which is what moves the ceiling
- The person who does nothing has the most to gain from doing a little
Muscle mass and strength
Grip strength predicts mortality. Muscle is the largest site of glucose disposal in the body, so losing it degrades metabolic health directly. And sarcopenia is what turns a fall in old age into a permanent loss of independence.
- Resistance training twice a week, progressively loaded
- Adequate protein, and more of it with age rather than less — the anabolic response blunts as we get older
- Start now. Peak muscle mass sets the height from which the decline begins
Metabolic health
Insulin resistance sits upstream of cardiovascular disease, type 2 diabetes, Alzheimer's and several cancers. It is measurable early — fasting insulin, triglyceride-to-HDL ratio, waist circumference — and it is reversible, which cannot be said of most of what it leads to.
Sleep, and the rest
- Sleep, seven to nine hours, consistently. Both short and disrupted sleep track with mortality
- Not smoking, and keeping alcohol low. Both are larger effects than anything in a supplement aisle
- Social connection, which shows an effect size in the mortality literature comparable to smoking cessation and which nobody sells
- Purpose and engagement, consistently present in every long-lived population studied
The Blue Zones data has been challenged on record-keeping grounds and some of the specifics may not survive. The consistent findings across them are still the unremarkable ones: constant low-level movement, real food, strong social ties, and not eating to excess.
The compounds
Rapamycin has the strongest animal evidence of anything in the field and is now in human trials — genuinely interesting, with real immunosuppressive risk. Metformin's longevity case rests on observational data that has weakened under scrutiny. Resveratrol has largely failed to replicate. NAD+ precursors raise the biomarker and have not yet shown much else.
None of them substitute for the list above, and taking them instead of it is the most expensive way to get nothing. If the training, the sleep and the metabolic markers are all in order, then a bet on an experimental compound is a reasonable thing to consider. In that order.

No comments yet