Three seconds a day. Four minutes a day. Five minutes a day. Forget your step count—speed, not steps. Every headline this week promises the same thing: that the price of protecting an aging body has collapsed to something you can pay while the kettle boils.
Here is the number nobody put in a headline: 560 to 610 minutes a week.
That is what a British Journal of Sports Medicine analysis of UK Biobank accelerometer data suggests you need—of moderate-to-vigorous physical activity—for a 30% reduction in cardiovascular disease risk. Not 150 minutes. Closer to ten hours.
Not 150 minutes. Closer to ten hours. Both figures are real, and holding them in the same hand is the only honest way to read the week.
Feel the whiplash. That is the point. Both figures are real, and holding them in the same hand is the only honest way to read the week.
Last week this newsletter walked through the resistance-training dose-response curve and where cancer-mortality benefit seemed to peak at surprisingly low doses. This week the minimal-dose claims got louder and more specific, and it is worth saying plainly what they do and do not show.
Start with the seductive arithmetic, because it isn't hype. The FAST-2 trial, published in PLOS One out of Penn State, was a genuine randomized controlled trial: 97 adults, average age 74, four minutes a day of push-ups, chair stands, band rows, and stair stepping.
After 12 weeks, the exercise group gained 4.2 additional chair-stand repetitions, held a one-legged stand 3.6 seconds longer, and shaved 2.3 seconds off their sit-to-stand time versus controls.
The three-second study is real too—a randomized trial in the Journal of Strength and Conditioning Research, 120 sedentary adults aged 30 to 65, three seconds of maximal-effort lifting per day, three sets, three times a week. Twelve weeks later: 12% more leg-press strength, 9% more bench press.
These are not junk findings. They are randomized, controlled, and published. If you stopped reading here, you'd conclude the guidelines are a scam.
But look at what these trials actually measured, and look at who they measured it on.
FAST-2 measured chair-stands and balance time. The three-second study measured leg-press and bench-press strength. Both are strength and mobility outcomes. Both are surrogate markers—proxies for function, captured over 12 weeks in small, screened samples.
Now look at the 560-minute figure. That came from wrist accelerometers on roughly 150,000 people, linked to actual hospital records of cardiovascular events. That is not a surrogate. That is disease, tracked across a population.
This is the outcome mismatch that the headlines quietly launder. Strength and mobility are not cardiovascular protection. A better chair-stand does not clear an artery. And a 12-week gain in a screened trial is not the same evidence class as a mortality signal read across a decade of a national biobank.
Conflating them is what oversells the tiny dose. "Four minutes improved mobility" becomes, in the retelling, "four minutes is all you need"—a claim no one in either study made.
There's a reason the distinction bites. A perspective paper out of Cambridge assembled the stakes in blunt numbers: low muscular strength is associated with roughly a 200% increase in all-cause mortality risk, and very low cardiorespiratory fitness with about a 400% increase. Smoking, for comparison, sits near 50%.
So both endpoints matter enormously—strength and cardiorespiratory fitness are among the largest mortality signals we have. But they are different endpoints, moved by different doses. The minimal-dose trials speak almost entirely to the first. The 560-minute figure speaks to the second.
This reopens the thread from last issue. We noted then that cancer-mortality benefit appeared to peak at modest resistance doses while cardiovascular protection kept demanding more. This week sharpens the same asymmetry: the outcome you're chasing sets the dose. There is no single "enough."
Which is why the baseline matters more than any headline. The FAST-2 participants were doing 18 minutes of total exercise per week before the trial. Eighteen minutes.
That is not a neutral starting line. As San-Millan's group argues from muscle biopsies, the sedentary state isn't a blank baseline—it looks more like early bioenergetic downshift, with substantially less mitochondrial machinery to burn fuel. "Healthy sedentary," they suggest, may be close to an oxymoron. That's the one-line nod; the mechanism is a story for another issue.
The practical upshot is that from 18 minutes a week, almost anything is an improvement. Four minutes a day of resistance work against that floor is a real, defensible on-ramp. The Cambridge author's own framing is that as little as 15 minutes a day is associated with reduced all-cause mortality—and that risk keeps declining as the minutes climb.
That last clause is the whole argument. The dose-response curve does not flatten at the minimum. It keeps falling well past it.
So here is a defensible reader threshold, and it is dual-track.
Track one: if you are genuinely sedentary—closer to 18 minutes a week than to 150—use the minimal doses as designed. Four minutes of daily chair stands and band work. Three-second maximal lifts. Bodyweight squats every 45 minutes, which in one comparison controlled post-meal blood sugar better than a single 30-minute walk. These buy strength, balance, and functional independence, and the trial evidence for that is legitimate.
Track two: treat "something is better than nothing" as a floor to escape, not a destination to settle into. Because the mortality and cardiovascular endpoints—the 200% and 400% numbers—respond to more, not less.
More what? The Cambridge summary is compact: more variation, more time, more intensity. The 560-minute figure is a ceiling most people will never reach, but it names the direction of travel honestly. And intensity matters independently—vigorous activity yields more protection per minute than gentle movement, which is precisely what the kettle-boiling routine tends to skip.
That's the trap in the "exercise snacks" framing. Ekelund's five-minute finding is real, and cutting 30 minutes of daily sitting is associated with a 7% lower early-death risk. But snacking is graded onto the bottom of the curve. It rescues the sedentary; it does not deliver the cardiovascular reduction that ten hours of MVPA is associated with.
A minimal dose that never increases in intensity, variation, or time is a floor you've decided to live on.
So read the week this way. The three-second and four-minute studies are true, and they are answering the question "what protects functional independence in someone doing almost nothing?" The 560-minute figure is also true, and it answers "what substantially lowers cardiovascular risk across a population?"
Those are different questions on different bodies with different endpoints. The honest answer to "how much movement protects an aging body" is that it depends entirely on which protection you mean—and that for the outcome most people actually fear, the number is far larger than any headline was willing to print.
Use the small dose to get moving. Then keep climbing, because the evidence says the curve does too.