A study has been doing laps around the health press this month, and the headline is always the same: your body peaks at 35, and it is downhill from there. It is an irresistible line, and it is roughly true. It is also the least interesting thing in the paper.
The research behind it is genuinely unusual. Most of what we think we know about physical aging comes from comparing different people of different ages at a single moment — a room of 25-year-olds measured against a room of 60-year-olds. That design has a well-known weakness: the two groups did not grow up in the same world, did not eat the same food or do the same jobs, and the older group has already been filtered by who survived and who was willing to show up. You end up measuring generational difference as much as aging.
The Swedish Physical Activity and Fitness cohort did it the hard way instead. Researchers at Karolinska Institutet took 427 people — 48% of them women, all born in 1958 — and measured their physical capacity over and over as they aged, from 16 all the way to 63.[1] Same people, same tests, forty-seven years. The comparison is not you against a stranger. It is you against your younger self.
When the peak actually arrives
The answer turns out to depend entirely on what you measure, which is the first thing the headlines flatten.
Maximal aerobic capacity — the body’s ceiling for using oxygen, and the quality behind VO2 max — peaked somewhere between ages 26 and 36 in both men and women. Muscular endurance, tested as bench press repetitions, peaked in the same window. So “around 35” is fair, but it is the late edge of a broad plateau rather than a summit you fall off.
Muscle power was a different story. Measured with the Sargent jump — a standing vertical leap — men peaked at 27 and women at 19. Explosive power, the quality that lets you catch yourself when you trip, fades earliest of all. That asymmetry is worth sitting with: the thing most people train least is the thing that starts slipping first.
The decline, in real numbers
What followed the peak was not a cliff. The decline began at roughly 0.3% to 0.6% per year and stayed there for a long stretch before accelerating to about 2.0% to 2.5% per year with advancing age. Across the whole span from peak to 63, total losses ranged from 30% to 48% depending on the measure. Men and women declined at statistically indistinguishable rates.[1]
Half a percent a year is the important number, and it is important precisely because it is undetectable. Nobody notices being half a percent weaker than last spring. You notice it at 55, as a sudden-seeming discovery that stairs are different now — but the slope began two decades earlier and simply compounded in silence. That is the actual lesson of the early peak: not that something dramatic happens at 35, but that something undramatic starts then and never stops.
The number nobody reported
Here is the finding that deserved the headlines. As this cohort aged, they did not decline together. They scattered.
The group’s variance — the statistical spread between the highest and lowest performers — exploded. For relative aerobic capacity, the variance increased roughly twenty-five-fold from adolescence to age 63. Jump height variance grew nearly five-fold, muscular endurance three-fold.[1]
Translate that out of statistics. At 16, these people were broadly similar: a room of teenagers, some fitter than others, none of them very far apart. At 63, the same people were barely comparable. Some were functionally decades younger than others born in the same year, tested on the same equipment, by the same research group.
This is why the average curve is a poor guide to your own life. A mean decline of 0.5% a year is a real description of the cohort and an almost worthless description of any person in it. The honest picture of physical aging is not a single line sloping down. It is a fan opening outward — and which line inside that fan you end up on is not settled at 35.
What separated the upper lines from the lower ones
The researchers looked for what predicted a better trajectory, and two things stood out.[1]
The first was unsurprising: higher leisure-time physical activity at 16 was associated with better performance across every outcome. An athletic adolescence buys you altitude.
The second matters far more to anyone reading this past 40. Becoming active in adulthood was also associated with better performance across all outcomes. The people who started late were not locked out by their teenage selves. In the Karolinska team’s summary of the work, adults who took up activity improved their physical capacity by something in the range of 5 to 10 percent.[2]
A university degree was separately associated with better absolute aerobic capacity and muscular endurance. That is almost certainly not education doing the work — it is a stand-in for income, occupation, health literacy and the other things that travel with it. Worth naming, not worth over-reading.
What this study cannot tell you
This is observational research, so every relationship above is an association. The study cannot prove that taking up running at 45 causes a shallower decline; it can only show that people who did so measured better. The causal arrow could run partly backwards — people with more robust bodies find exercise more pleasant, and do more of it.
There are narrower limits too. This is a single birth cohort, everyone born in 1958, all in Sweden, and 427 people is a serious longitudinal achievement but still a modest sample to split by sex and activity level. Studies that follow people for 47 years inevitably lose participants, and those who remain tend to be healthier and more engaged than those who drop out — which, if anything, probably makes these decline estimates slightly optimistic. And “strength” here means bench press repetitions and a vertical jump: two specific tests, not a full account of what a body can do.
None of that undermines the core result. It does mean the paper is best read as a high-quality map of what typically happens, not a verdict on what must happen to you.
The useful way to read this
If your physical capacity is going to fall by something like a third to a half between your early thirties and your early sixties no matter what, the obvious strategic question is not how to flatten the slope — it is how high the line sits when the slope begins, and whether you are drifting toward the top or the bottom of that widening fan.
That reframing matters because the level, not the rate, is what the mortality evidence tracks. Cardiorespiratory fitness is one of the strongest predictors of all-cause mortality we have: a large meta-analysis found each one-MET increment in fitness associated with roughly 13% lower all-cause mortality,[3] and in a study of over 122,000 patients undergoing treadmill testing, the most fit group had dramatically lower long-term mortality than the least fit — with no observed ceiling of benefit.[4] Separately, people who were inactive in early adulthood but became active later still showed substantially lower mortality than those who stayed sedentary.[5]
In practice that points at unglamorous things this magazine keeps arriving at from different directions: build the aerobic base, since that is the measure with the steepest mortality gradient and the one that responds to steady low-intensity work plus occasional hard intervals; lift, because structured resistance training beats merely staying busy for preserving muscle; train something explosive, because power peaked first and almost nobody trains jumps; and eat enough protein to support the muscle you are asking your body to keep. If you want a rough sense of where you currently sit in the fan, our fitness age calculator turns a few inputs into a number, and the broader daily habits piece covers the rest of the ground.
The study’s own authors landed on a conclusion more interesting than the headline it generated: decline in physical capacity is measurable well before 40, long before it becomes clinically relevant, and it matters most in people who are sedentary.[1] Thirty-five is not a deadline. It is roughly when the fan starts to open, and the evidence says you get a vote in which direction you go.
Common questions
At what age does physical fitness start to decline?
In the Swedish SPAF cohort, maximal aerobic capacity and muscular endurance peaked between ages 26 and 36 in both men and women, then declined gradually. Muscle power, measured by jump height, peaked earlier - around age 27 in men and 19 in women. So there is no single peak age: it depends on which quality you measure, and the "peak" is better described as a broad plateau through the late twenties and early thirties than a sharp summit at 35.
How fast do strength and fitness actually decline with age?
The decline started slowly, at roughly 0.3% to 0.6% per year, and accelerated with age to about 2.0% to 2.5% per year. Across the full span from peak to age 63, total losses ranged from 30% to 48% depending on the measure. The early rate is small enough that it is essentially invisible from one year to the next, which is part of why it goes unnoticed until it compounds.
Is it too late to start exercising in your forties or fifties?
The study found that people who became physically active in adulthood - not only those who were active as teenagers - performed better across every outcome measured. That is an association rather than proof of cause, but it points the same way as the broader evidence on physical activity and mortality: taking up activity later in life is linked with meaningfully better outcomes than remaining sedentary.