Thinking exercise only counts in big doses

Checked against 2 peer-reviewed sources, most recently on 26 September 2026.

Many people think exercise only counts in big doses, so they do none. In a large meta-analysis of cohort studies, half the usual weekly recommendation of moderate activity, such as brisk walking, was linked to a lower risk of early death than no activity outside work, and most of the lower risk on the curve was already there at that half. The fix: half the usual weekly target of brisk movement, starting today.

The one honest fix

Get about 75 minutes of brisk movement a week — around 11 minutes a day — starting today. Half the usual target already counts.

What the evidence says

In a dose-response meta-analysis of large prospective cohort studies, adults doing about 75 minutes a week of moderate activity such as brisk walking outside work - half the usually recommended minimum - had a 23% lower risk of all-cause death than adults reporting no such activity

−23% · RR 0.77, 95% CI 0.73-0.80

Meta-analysis of cohort studies · associational

Limitations & how to read this number

These studies followed people over time; they show that people who moved more died less often, not that the movement alone caused it. People who are active differ from inactive people in many ways, and illness can make people less active before it is diagnosed - the researchers left out studies with less than three years of follow-up to reduce that problem, but cannot remove it entirely. Activity was mostly reported by the participants themselves in questionnaires, and activity at work was not counted. The 75 minutes is a point read off a pooled curve, compared with doing no activity outside work at all; it is not a group of people who were told to do 75 minutes. The analysis reports weekly amounts, so it does not say whether the minutes are best done daily or all at once.

SourceGarcia L, Pearce M, Abbas A, Mok A, Strain T, Ali S, Crippa A, Dempsey PC, Golubic R, Kelly P, Laird Y, McNamara E, Moore S, de Sa TH, Smith AD, Wijndaele K, Woodcock J, Brage S (2023) · Br J Sports Med · 10.1136/bjsports-2022-105669

How to actually do it

Aim for about 75 minutes a week of movement brisk enough to raise your breathing - a fast walk counts. Spread over the week that is around 11 minutes a day. It needs no gym and nothing you don't already have. If you already do more, keep going; the point is that starting small is not pointless.

Supporting findings (2)

In the same meta-analysis, doubling that to the usually recommended 150 minutes a week was associated with a 31% lower risk of all-cause death, and doubling again to 300 minutes with 34% - most of the lower risk came with the first 75 minutes

−31% · RR 0.69, 95% CI 0.65-0.73

Meta-analysis of cohort studies · associational

These are three points on the same pooled curve from the same studies, all compared with no activity outside work, so they are not three separate findings. Differences at higher amounts were smaller and less certain, partly because fewer people report doing that much. The same cautions apply as above: the studies show a link, activity was mostly self-reported, and people who move more differ in other ways.

In a harmonised meta-analysis of eight cohort studies that measured activity with accelerometers in adults aged 40 and over, people in the second-lowest quarter of moderate-to- vigorous activity had a 36% lower risk of all-cause death than the least active quarter, and activity of any intensity, including light activity, was associated with lower risk, with the steepest part of the curve at the low end

−36% · HR 0.64, 95% CI 0.55-0.74

n=36,383 · Meta-analysis of cohort studies · associational

Activity was measured with a movement sensor worn at the hip for several days, which avoids relying on memory but captures only one short period. Participants were middle-aged and older (average age about 63), followed for a median of about six years, so some of the least active were likely already unwell; the researchers found similar results when they left out deaths in the first two years, but say this bias may persist. It compares groups of people, not people who changed how much they moved.

SourceEkelund U et al. (2019) · BMJ · 10.1136/bmj.l4570

Go deeper

Open what you care about

Why the first minutes matter most

When researchers pooled large cohort studies, the drop in risk was steepest at the low end: half the usual recommendation was linked to a 23% lower risk of early death, the full recommendation to 31%, and double it to 34%. A pooled analysis of eight cohorts that measured activity with movement sensors rather than questionnaires found the same shape, with light activity counting too.

Garcia L, Pearce M, Abbas A, Mok A, Strain T, Ali S, Crippa A, Dempsey PC, Golubic R, Kelly P, Laird Y, McNamara E, Moore S, de Sa TH, Smith AD, Wijndaele K, Woodcock J, Brage S 2023 · Ekelund U et al. 2019

Questions people ask

Does this mean 75 minutes a week makes you live longer?

Not proven. These studies followed people over time and show that people who did about that much activity died less often than people who reported none outside work. Active people differ in other ways too, so the link is not proof that the activity alone caused it.

Is more better?

Up to a point. Risk kept falling with more activity, but each extra hour was associated with a smaller difference than the first, and above about 300 minutes a week the differences were small and uncertain.

Sources

Garcia L, Pearce M, Abbas A, Mok A, Strain T, Ali S, Crippa A, Dempsey PC, Golubic R, Kelly P, Laird Y, McNamara E, Moore S, de Sa TH, Smith AD, Wijndaele K, Woodcock J, Brage S (2023). Non-occupational physical activity and risk of cardiovascular disease, cancer and mortality outcomes: a dose-response meta-analysis of large prospective studies. Br J Sports Med. Meta-analysis of cohort studies. doi:10.1136/bjsports-2022-105669
Ekelund U et al. (2019). Dose-response associations between accelerometry measured physical activity and sedentary time and all cause mortality: systematic review and harmonised meta-analysis. BMJ. Meta-analysis of cohort studies. doi:10.1136/bmj.l4570

Related fixes

Works alongside this