Exercise and Longevity: What the Evidence Says

Exercise is the intervention this site should have covered first, because it is the one where the human evidence is deepest and the practical questions, how much, how hard, what kind, have actual numeric answers. This is the map of the silo: the doses and thresholds live here, the aerobic specifics in zone 2, the lifting specifics in strength training.

How much does activity actually buy you?

The best single answer comes from Arem’s pooled analysis of six cohorts, 661,137 adults, published in JAMA Internal Medicine. Against people reporting no leisure-time activity:

Two features of that table matter more than the exact percentages. The steepest part of the curve is at the bottom: going from nothing to “some” captures most of what going from “some” to “a lot” adds. And the top of the curve is flat, not inverted; the marathon-will-kill-you genre has no support here.

This is cohort data, so the usual caveat applies: healthy people move more, and statistics cannot fully unbraid that. But the dose-response shape, the consistency across six cohorts, and the mechanistic support make exercise the least controversial claim on this site.

Fitness is the number the cohorts are proxying

Self-reported minutes are a rough proxy for the thing that actually predicts survival: cardiorespiratory fitness. In Cleveland Clinic treadmill data on 122,000 patients, fitness stratified long-term mortality more strongly than smoking, diabetes or coronary disease, with no upper limit: the elite-fit outlived the merely high-fit.

The practical reading is not “become elite”. It is that fitness is trainable at any starting point, the biggest relative gains go to the least fit, and a treadmill or field test gives you a number to move, which beats counting minutes. How to train it cheaply and sustainably is the whole subject of zone 2, and the target ranges come from your own numbers via the zone 2 calculator.

Strength is a separate line item

Momma’s meta-analysis of 16 cohorts asked a different question: does muscle-strengthening activity predict mortality independently of aerobic activity? It does: 10 to 17% lower risk of all-cause mortality, cardiovascular disease, cancer and diabetes, with the maximum benefit at roughly 30 to 60 minutes per week and a J-shaped curve past it.

Independence is the interesting word. The cohorts suggest you cannot fully substitute one for the other: cardio does not build the muscle that keeps you off the floor at 80, lifting does not build the engine. The combination, both boxes ticked, showed the lowest risk of all. The lifting dose, oddly small, gets its own examination in strength training.

The weekly minimum, assembled

Putting the two evidence streams into one calendar:

1. 150 to 300 minutes of moderate aerobic work, most of it easy enough to talk through. Three to four sessions of 45 to 60 minutes covers it.
2. Two strength sessions, 30 to 60 minutes total across the week, compound movements.
3. Everything above this is preference, not prescription. The curve has flattened; train more because you like it.

For a beginner the honest entry point is lower: the −20% row of Arem’s table is reachable with a daily half-hour walk, and it is the single largest jump in the whole dataset. Zero to something is the trade of a lifetime; the rest is refinement.

What exercise does not need

It does not need to be optimised before it exists. The person comparing polarised training models while training zero hours a week has the order backwards, and I say that as someone who has caught himself doing exactly this with training plans the way others do with supplement stacks. Consistency for eight weeks beats the perfect program abandoned in two.

It also does not need gear. The evidence above was generated by people walking, cycling and lifting ordinary weights. A heart-rate strap is genuinely useful for zone work; nothing else on the accessories wall changes the curve.

When to check with a doctor first

Chest pain or pressure on exertion, fainting, unexplained breathlessness disproportionate to effort, or a strong family history of sudden cardiac death: these go to a doctor before a training plan, not after. The same applies to starting vigorous training past 50 after decades of inactivity; a check-up first is cheap insurance, and the moderate end of the spectrum remains open in the meantime.

What we do not know

How much of the cohort association is causal remains genuinely unresolvable at this scale; randomised trials of decades of exercise cannot be run. The independent contribution of intensity versus volume is still being untangled, which is why the zone 2 article treats intensity claims more sceptically than volume claims. And the J-shape at high strength-training volumes is unexplained: artefact, confounding, or real, nobody knows yet.

Where to start this week

If you currently do nothing: 30 minutes of brisk walking daily, and that is the whole plan for a month. If you have the aerobic base: add the two lifting sessions. If both are in place: compute your zones with the calculator and structure the aerobic hours properly. And guard the recovery side of the ledger, because the adaptations you are training for are built during sleep, not during the session. If you add one supplement for the strength side, the only one with strong evidence is creatine.