What declines first after 40: muscle, VO2 max, or testosterone?

Patients ask me this in almost exactly those words. They are usually forty-two, they have noticed something, and they want to know which thing to go after.

Short answer: muscle strength falls two to five times faster than muscle mass, which declines under half a percent per year in cross-sectional studies and under one percent even at seventy-five. Bioavailable testosterone falls 2 to 3 percent per year in men aged 40 to 70. Aerobic capacity is the one that accelerates, running 3 to 6 percent per decade in the 20s and above 20 percent by the 70s. Muscle mass, the one most people watch, is the slowest of the three in these cohorts.

Most people asking it assume the three decline at roughly the same speed, and the job is picking the one that starts earliest. They do not, and the gap is large.

Muscle loss after 40 is slower than you think

This surprises people, because muscle is the one you can see.

Mitchell and colleagues published a quantitative review in Frontiers in Physiology in 2012 pulling together the studies on muscle size and strength across the lifespan. Across cross-sectional studies, the median rate of muscle loss was 0.47 percent per year in men and 0.37 percent per year in women. In longitudinal studies of people at age 75, muscle was lost at 0.64 to 0.70 percent per year in women and 0.80 to 0.98 percent per year in men.

Under one percent a year. At the cohort level, a forty-five-year-old is losing muscle mass too slowly to notice month to month. Individuals vary, and illness, injury or long periods of inactivity change that picture.

So when a forty-five-year-old tells me he is losing muscle, he is describing something real. It is usually not mass. The timing of all of this is its own subject, which we covered in how the body changes by decade.

Sarcopenia and dynapenia: why strength falls faster than mass

The same review makes the distinction that matters most here. Muscle mass and muscle strength decline on different trajectories, and strength declines considerably faster.

At age 75, Mitchell reports strength lost at 3 to 4 percent per year in men and 2.5 to 3 percent per year in women, against mass loss of well under 1 percent. In the studies that measured both in the same sample, strength was lost two to five times faster than mass.

The field has separate words for it now. Sarcopenia is the loss of muscle mass. Dynapenia is the loss of strength. They do not move together, which means the ratio between them changes as you age. You are not only getting smaller. You are getting weaker per unit of muscle you still have.

The review's own bottom line is blunter than most summaries of it: loss of strength is a more consistent risk for disability and death than loss of muscle mass.

That is why a scale is close to useless here, and why body composition alone is not much better. A DEXA can tell you your lean mass held steady. It cannot tell you the force that lean mass produces has dropped. That takes a strength measurement, and in my experience most people over forty have never had one recorded.

VO2 max decline by age: the one that accelerates

This is what I most often find missing from a chart, and its rate changes more across the lifespan than anything else here.

Fleg and colleagues published the longitudinal reference in Circulation in 2005, from the Baltimore Longitudinal Study of Aging. Serial treadmill peak oxygen consumption tests on 375 women and 435 men aged 21 to 87, following a community-dwelling cohort free of clinical heart disease for a median of 7.9 years. Rates from a cohort screened that way are a best case. This design matters because earlier estimates came largely from cross-sectional studies, which the authors describe as potentially misleading and overly optimistic about aging.

They found decline in every decade from the 20s through the 70s, in both sexes. The rate was not constant. It ran 3 to 6 percent per ten years in the 20s and 30s and exceeded 20 percent per ten years by the 70s and beyond.

Two details deserve more attention than they get.

From the 40s onward the rate of decline was larger in men than in women. Forty is roughly where the sexes separate. Worth noting that this is not the same curve as metabolic rate, which does not slow when people think it does.

And the same pattern held across all quartiles of self-reported leisure-time physical activity. This is an activity questionnaire in an observational cohort, not a training intervention, and the decade rates were themselves adjusted for that activity. It cannot tell us what structured training does to the slope. What it does show is that being generally more active did not exempt anyone from decline.

Hollenberg and colleagues, in the Journals of Gerontology in 2006, followed 339 women and 253 men, relatively healthy older adults, over six years. Aerobic capacity fell 18 percent per decade in women and 24 percent per decade in men. They then adjusted for two variables: forced expiratory volume in one second, and maximal exercise heart rate. That dropped the decline to 9.7 and 10.4 percent per decade. Most of the measured decline was no longer apparent once those two were accounted for.

Testosterone decline per year: 1.6 percent total, 2 to 3 percent bioavailable

Feldman and colleagues published the reference longitudinal numbers in the Journal of Clinical Endocrinology and Metabolism in 2002, from the Massachusetts Male Aging Study. The cohort enrolled 1,709 men aged 40 to 70 and followed 1,156 of them for 7 to 10 years.

Within individual men over time, total testosterone declined 1.6 percent per year and bioavailable testosterone declined 2 to 3 percent per year.

In the same cohort, sex hormone binding globulin rose 1.6 percent per year on the cross-sectional measure. Rising SHBG binds more of the testosterone that is present, which is the standard explanation for why the bioavailable figure falls faster than the total, though this study measured the trends rather than testing that mechanism.

One methodological point here changes how you read most testosterone charts. In the same study the cross-sectional trend for total testosterone was 0.8 percent per year, while the longitudinal within-subject trend was 1.6 percent. Following the same men over time showed decline running at twice the rate that comparing different men of different ages suggested. Most population charts patients bring me are built on the gentler method.

The numbers on one scale

Rates below are converted to percent per decade so they can be compared, multiplied out linearly rather than compounded, with the population each figure was measured in. Three of these come from different age ranges and they are not interchangeable.

Two things fall out of that table.

Strength carries the largest per-decade number in these sources. Muscle mass carries the smallest. Be careful with that gap though: it looks like an order of magnitude only because it sets strength measured at seventy-five against mass from a young-versus-old comparison. Measured in the same people, the honest figure is the two to five times above.

And aerobic capacity does not decline at a fixed rate. Fleg does not publish a midlife decade rate. What that study establishes for the middle of life is a direction rather than a number: the rate accelerates with every successive decade instead of holding steady.

Aerobic capacity and aging: what else moves the number

Travison and colleagues published a follow-on analysis of the same Massachusetts cohort in the Journal of Clinical Endocrinology and Metabolism in 2007, enrolling 1,667 men aged 40 to 70. Of those, 947 returned for the second visit and 584 for the third, across roughly fifteen years. They separated the contribution of chronological aging from health and lifestyle changes.

An increase in body mass index of 4 to 5 kg/m2, or the loss of a spouse, was each associated with a testosterone decline comparable to about ten years of aging. Body composition sits upstream of a lot of this, which is also what actually happens to metabolism with age.

This is an observational association, not a demonstrated effect of intervention. The authors put it as a possibility, that age-related hormone decline may be decelerated through management of health and lifestyle factors. Note also that the direction is not settled. Falling testosterone contributing to weight gain is compatible with the same data.

It is a reason to have the measurement in hand early enough for any of that to be a conversation.

One more thing worth saying plainly: of the three, strength and aerobic capacity are the two that respond most to training. That is not a claim about reversing the curve, and none of the studies here tested it. It is the reason those two are worth knowing your own numbers for.

If you would rather have a baseline than an estimate, the ten-tier longevity protocol shows the order our physicians work in and where each of these measurements sits.

What we measure, and when

At Rebel Health Alliance the first panel runs more than 100 biomarkers through Quest, including total, free and bioavailable testosterone, IGF-1, and a full thyroid panel of TSH, free T3, free T4, total T3 and reverse T3, with SHBG added on the men's panel. It runs about $450 at our negotiated rate, about $600 in New York State. Those prices are current as of September 2026 and subject to change, and labs are billed separately from membership.

Beyond that first panel, members can order from more than 3,000 diagnostic tests. Most of the markers that matter here are ones not typically included in a standard annual panel.

For the two things bloodwork cannot see, we use body composition by DEXA for lean mass and bone density, and VO2 max testing for aerobic capacity, which is a maximal exercise test done under supervision. Both are cash-pay add-ons, ordered after a physician review and only when your physician thinks the result will change something. Add-on pricing is quoted before anything is ordered.

The sequence matters more than the menu. Everything runs through a sequential, patient-paced ten-tier longevity protocol that starts at metabolic foundation and cancer screening. It does not skip ahead, because a hormone number read without the metabolic picture behind it tends to produce the wrong intervention.

Questions patients ask

At what age does this start?
Aerobic capacity is already declining from the 20s in the Baltimore data, so nothing starts at forty. What changes around forty is the rate, and in Fleg 2005 it is also where men and women separate, with men declining faster from that decade onward. We have written separately on how the body changes by decade.

If I train hard, am I exempt?
Fleg cannot tell you. In the Baltimore cohort the rate of decline was similar across all quartiles of self-reported leisure-time physical activity, and the decade rates were adjusted for that activity. But self-reported leisure activity is not structured training, and the study was observational, so it was never designed to test whether deliberate training alters the rate. What it establishes is that general activity level did not flatten the curve for anyone in that cohort. Absolute fitness still matters enormously for what you can do at any age.

Which of these is most often missing from a chart?
Aerobic capacity. In my practice, most people over forty arrive with a lipid panel and a weight on file and no VO2 max ever recorded, so there is nothing for a later result to be compared against. Whether a VO2 max test makes sense for any particular person, and when, is a conversation with their own physician. It is a maximal exercise test, performed under supervision, and it is not appropriate for everyone.

Is a normal testosterone result reassuring?
It depends what was measured. In the Massachusetts data the bioavailable fraction fell 2 to 3 percent per year within individual men, while SHBG trended upward, so total testosterone can look stable while the usable portion falls. Ask which one was run.

Does this apply to women?
The aerobic and muscle data cover both sexes and the direction is the same, though the rates differ, with women declining more slowly than men from the 40s onward in the Baltimore cohort. The testosterone data cited here is from a male cohort and should not be read across.

Getting a baseline

The reason to read any of this is that a first measurement is worth more the earlier you take it. A VO2 max at fifty tells you less than one at forty and a second at forty-three, because the second one is what turns a number into a direction.

If you want to talk through which of these is worth measuring in your case, book a call.

Sources

1. Fleg JL, et al. Accelerated longitudinal decline of aerobic capacity in healthy older adults. Circulation. 2005. PMID 16043637.
2. Mitchell WK, et al. Sarcopenia, dynapenia, and the impact of advancing age on human skeletal muscle size and strength; a quantitative review. Frontiers in Physiology. 2012. PMID 22934016.
3. Feldman HA, et al. Age trends in the level of serum testosterone and other hormones in middle-aged men: longitudinal results from the Massachusetts Male Aging Study. Journal of Clinical Endocrinology and Metabolism. 2002. PMID 11836290.
4. Travison TG, et al. The relative contributions of aging, health, and lifestyle factors to serum testosterone decline in men. Journal of Clinical Endocrinology and Metabolism. 2007. PMID 17148559.
5. Hollenberg M, et al. Longitudinal changes in aerobic capacity: implications for concepts of aging. Journals of Gerontology Series A. 2006. PMID 16912104.

Dr. Alec Weir is the Chief Medical Officer at Rebel Health Alliance, a physician-led longevity and health optimization membership, 100 percent virtual, with 50-state operational capability. Medical services are provided by independent physician-owned practices. rebelhealthalliance.io

This article is educational and is not medical advice, and reading it does not create a physician-patient relationship. Rates of decline described here are population averages from the cited cohorts and do not predict any individual's trajectory. Talk to your own licensed physician about your results and before beginning any exercise testing. Any testing or treatment depends on an individual clinical evaluation, and availability varies by state and by applicable law. Nothing here is an offer to prescribe. Rebel Health Alliance is a longevity and health optimization membership. It is not health insurance. Diagnostics are cash-pay and billed separately from membership.