Ninety minutes a week. That is the dose where the returns on lifting weights top out, and a Harvard team that tracked 147,374 health professionals for three decades found nothing waiting past it. Somewhere in the band between 90 and 119 minutes of resistance training a week, the all-cause mortality curve bottoms out at a 13 percent lower risk of dying, a hazard ratio of 0.87. Push to 120 minutes and beyond, and the line goes flat. The extra hours buy nothing the body will pay you back for.

The result, published in June in the British Journal of Sports Medicine, is a ceiling, and the ceiling is the useful part. Headline writers will read it as one more reason to train. An operator should read the plateau: two hours a week is the whole prescription, and effort past it is spend with no conversion.

The extra hours buy nothing the body will pay you back for.

The cohorts were the Health Professionals Follow-Up Study and the two Nurses’ Health Studies, with 35,798 deaths recorded across the follow-up. At the 90-to-119-minute band, cardiovascular death ran 19 percent lower, a hazard ratio of 0.81, and death from neurological disease, Alzheimer’s and Parkinson’s among them, ran 27 percent lower at 0.73. The neurological figure is the largest and the softest. No drug on the market moves it that far in healthy adults, and no observational study can prove the lifting is what moved it here. Grade it honestly: this is Level 2 evidence, association rather than cause, drawn from a mostly white, professional cohort of people who already tend to look after themselves.

The one place the data rewards more work is stacking. Participants who paired the strength dose with high aerobic volume, thirty to forty-five MET-hours a week, reached a hazard ratio of 0.55, roughly 45 percent lower mortality. Cardio and resistance stack rather than substitute, and the combination holds the biggest number in the paper.

The ceiling behaves the way a paid-acquisition channel does. There is a point where the next dollar stops converting, and the operator’s job is to find it and move the budget somewhere it still returns. Two gym sessions is that point for strength training. The hours past it are the ones a disciplined operator reallocates, to sleep, to Zone 2, to sitting in a room with people who matter.

The single highest-return intervention in the June data carries no invoice and fits in three sessions a week.

That reallocation is the quiet rebuttal to a boom running three floors up from where cè’s readers work. Singapore’s longevity clinics are selling founders the premium edition of the same question. Chi Longevity, run out of the Four Seasons by the geroscientist Andrea Maier, prices its packages from S$4,250 to S$18,000, gene panels and cognitive tests and a six-month retest included, and Fortune sized the wider industry at 4.6 trillion dollars. Maier is also the one who said the quiet part on the record: public enthusiasm has outpaced scientific validation. The single highest-return intervention in the June data carries no invoice and fits in three sessions a week.1

The move is to bank the free thing first, and to treat the six-figure package the way a founder treats an unproven tool someone wants in the stack: interesting, unvalidated, made to earn its line before it gets one. The gym costs a membership and two evenings. Everything above that is a purchase before it is a protocol.

Footnotes

  1. The caveat the clinics will hand back: the exact minute-count is soft. Change the cohort, or change what counts as a session, and the sweet spot could land at 80 minutes or 130. The plateau is the robust finding; the decimal place is not. ↩