Function as a biological-age marker: VO2max, grip strength and gait speed

TL;DR

• The best-validated ageing markers are not molecular - a stopwatch, a dynamometer and a chair, validated against mortality and disability directly rather than against a proxy.
• Gait speed carries the information content of a workup - a pooled analysis of 9 cohorts and 34,485 adults found age, sex and gait speed predicted survival about as well as a full clinical battery.
• Grip strength is not about the hand - PURE found lower grip associated with higher all-cause and cardiovascular mortality across countries at every income level.
• Midlife fitness holds over 46 years of follow-up, which partly answers the objection that low fitness is just a marker of existing illness.
• Repeatability is the practical edge - methylation clocks can move between two samples from the same blood draw; a performance test's variability is visible to you.
• Improving a score is not changing an outcome - the cohorts describe differences between people, not what one person gains by moving up the curve.
• Working on the markers is still worthwhile. Training, protein and sleep are the core; a new section maps the peptides and medicines people discuss with a physician to each marker, and is plain that only the GLP-1 medicines have established human effects here, on body weight and metabolic markers.

Educational content, not medical advice. No doses or protocols appear.

---

Two different claims: "how old are your cells" and "how well does your body work"

The biological-age market sells one claim and the functional literature makes another, and they are routinely spoken about as though they were the same claim in different clothes.

The molecular claim is about state. An epigenetic clock inspects a molecular layer - methylation at selected CpG sites - and returns a number expressed in years. The implicit proposition is that the body carries an internal age distinct from the calendar, that this age is written into the tissue, and that a laboratory can read it. Whether that proposition survives contact with the evidence is the subject of what biological age tests actually measure and epigenetic clocks; the short version is that these are statistical models, not measurements of a hidden quantity, and the unit "years" is a presentation choice.

The functional claim is about capacity, and it is much more modest. It says: here is a task, performed under standard conditions, and performance on that task has been observed to track future health outcomes in large populations. There is no assertion that walking speed is your age. There is only the observation that people who walk more slowly, on average and across very large groups, go on to experience worse outcomes than people who walk faster - after adjustment for the obvious confounders, and consistently enough across independent cohorts that the finding is hard to dismiss.

That is a smaller claim, and it is better supported. It is also why this article names no product. The subject matter here cannot be sold to you: the equipment is a stopwatch, a hand dynamometer and a chair.

VO2max: what the cohort evidence shows, and the gap between association and modifiability

Evidence tier: large prospective observational cohorts, some with multi-decade follow-up. Human data throughout. No randomised trial has assigned people to a fitness level and measured lifespan, and none realistically could.

Cardiorespiratory fitness - the body's capacity to take in, transport and use oxygen during sustained effort, measured as peak or maximal oxygen uptake - is one of the most consistently replicated predictors of all-cause mortality in the epidemiological literature. The finding is old, it has survived repeated re-examination, and it appears in cohorts assembled on different continents with different measurement methods.

Three lines of evidence are worth separating.

Cross-sectional fitness predicts subsequent mortality. In a large retrospective cohort of adults undergoing exercise treadmill testing, higher measured fitness was associated with lower long-term mortality, with the relationship persisting across the range of observed fitness rather than plateauing at a modest level (Mandsager et al., JAMA Network Open, 2018). Similar gradients appear in cohorts stratified by diabetes status and other comorbidities.

The association is durable over very long follow-up. Midlife fitness measurements have been linked to mortality across follow-up periods extending more than four decades, which is important because it partially addresses the most obvious objection - that low fitness is simply a marker of existing illness (Clausen et al., Journal of the American College of Cardiology, 2018). Illness present at baseline should wash out over forty years; the association does not.

Change in fitness tracks change in risk. Population follow-up studies that measured fitness twice have reported that improvement between measurements is associated with lower subsequent mortality, and decline with higher (Laukkanen et al., Mayo Clinic Proceedings, 2016). This is the closest the literature comes to a modifiability argument.

It is still not proof of modifiability, and the distinction matters more here than anywhere else on this page. Association is not causation. Fitness is not randomly distributed. People with higher measured fitness differ systematically from people with lower measured fitness in ways that no statistical adjustment fully captures: underlying cardiac and pulmonary reserve, skeletal muscle biology, freedom from undiagnosed disease, socioeconomic position, sleep, and the simple fact that being well enough to train is itself a health signal. A substantial part of measured VO2max is heritable, and a substantial part of the response to training is heritable too. When someone raises their fitness and their measured value rises, it does not follow that their risk has moved by the amount the cohort curve implies - because the curve was drawn across people who differed for many reasons, not across one person changing one variable.

The honest formulation: cardiorespiratory fitness is an excellent prognostic marker with strong biological plausibility as a causal factor, and the causal magnitude is not established by the observational data that establish the association.

One practical caveat attaches to all of it: the cohort evidence rests on measured fitness, and a wrist-estimated value is a much softer quantity - why a wearable's "fitness age" is not a biological-age test sets out how that number is constructed.

Grip strength: a whole-body proxy, and why a hand predicts things a hand should not

Evidence tier: large multinational prospective cohorts and pooled analyses. Human data.

The strangest result in this literature is that squeezing a device with one hand predicts outcomes that have nothing to do with hands.

The Prospective Urban Rural Epidemiology (PURE) study followed a large multinational cohort and reported that lower grip strength was associated with higher all-cause mortality and higher cardiovascular mortality, with the association holding across countries at different income levels (Leong et al., The Lancet, 2015). Pooled and single-cohort analyses since - including work across 28 countries examining all-cause and cancer mortality, and NHANES-based analyses in adults from early adulthood upward - have reproduced the direction of the finding. A Lancet commentary published alongside the PURE results put the question in the title that the field has used ever since: is grip strength a biomarker of ageing?

The mechanistic reading is that grip is not really measuring the hand. It is a cheap, low-noise, highly standardisable proxy for total skeletal muscle function and neuromuscular integrity, which in turn reflect nutritional status, systemic inflammation, physical activity history, neurological function and the presence of occult disease. A hand dynamometer is a crude window onto whole-body physiology, and its predictive value comes from what it correlates with rather than from anything about grip itself.

Two caveats keep the finding proportionate. First, the same objection applies as to fitness: low grip strength may be a consequence of undiagnosed illness rather than an independent cause of anything, and reverse causation is a live concern in cohorts with short follow-up. Second, when grip strength is added to models that already contain conventional risk factors, its incremental predictive contribution is often described as modest - it predicts well on its own, and adds less once age, sex, comorbidity and the usual clinical variables are already in the model. That is a normal fate for a good marker, and it is not usually mentioned by anyone promoting it.

Gait speed, chair-stand and balance: the measures geriatric medicine already trusted

Evidence tier: pooled analyses of individual participant data from multiple prospective cohorts; decades of clinical use. Human data.

Long before biological age became a consumer product, geriatric medicine had settled on a small set of physical performance tests because they predicted the outcomes clinicians cared about - hospitalisation, disability, falls, institutionalisation and death.

Usual walking speed is the most studied. A pooled analysis of individual participant data from nine cohorts, covering 34,485 community-dwelling adults aged 65 and over with follow-up ranging from six to twenty-one years, found gait speed associated with survival; strikingly, prediction based on age, sex and gait speed alone was about as accurate as prediction based on age, sex and a full battery of chronic conditions, smoking history, blood pressure, body mass index and prior hospitalisation (Studenski et al., JAMA, 2011). A stopwatch and a marked corridor carried roughly the information content of a clinical workup.

Chair-stand performance - rising repeatedly from a seated position without using the arms - indexes lower-limb power and is a component of the Short Physical Performance Battery, the composite that also includes gait speed and standing balance (Guralnik et al., 1994). Balance, held in progressively more demanding stances, contributes independently. A broader systematic review and meta-analysis of objectively measured physical capability - grip strength, walking speed, chair rise and standing balance - found each associated with subsequent mortality (Cooper et al., BMJ, 2010).

Two limitations should be stated plainly. These measures were developed and validated predominantly in older adults, and their discriminative power is greatest where there is meaningful variation to detect. In healthy midlife populations most people perform them without difficulty, which compresses the range and reduces what an individual result tells you. And because they are performance tests, they are sensitive to effort, footwear, instruction, surface, time of day and practice - the standardisation that makes them reliable in research settings is not automatic at home.

How these compare with epigenetic and organ-specific clocks

The comparison is not "which is more scientific". Both categories are statistical relationships derived from human cohorts. The differences are in what was validated, how reproducible the measurement is, and what a change in the number means.

The reproducibility row deserves emphasis, because it is the practical difference. Work on principal-component-based clocks has documented substantial technical noise in conventional methylation clocks - enough that the same person, sampled twice, can receive materially different results (Higgins-Chen et al., Nature Aging, 2022). A measure whose test–retest variation approaches the size of the effect you are looking for cannot track an individual over months, whatever its population-level performance. Functional measures do not have this problem; their variability comes from the person and the protocol, both of which are visible to you.

None of this makes molecular clocks worthless. They interrogate biology that no performance test can reach, and organ-specific approaches in particular may eventually answer questions functional tests cannot - see organ-specific ageing clocks. The point is narrower: if the purpose of a test is to predict hard outcomes reliably and cheaply, the performance measures are not a poor relation to the molecular ones.

What none of them do

This is the section that should shape any decision made after reading the rest.

No functional measure diagnoses anything. Slow walking, weak grip and low fitness are non-specific. They are associated with dozens of conditions and with none in particular. A poor result is a reason for a clinical conversation, not a finding. Conversely, a good result excludes nothing - asymptomatic disease coexists comfortably with strong physical performance.

Improving a measure is not the same as changing an outcome. This is the central error, and it is the same error that afflicts the molecular clocks. The cohorts established that people who perform better on these tasks experience better outcomes. They did not establish that a person who improves their performance thereby acquires the outcomes of the group they have moved into. Where an improvement reflects a genuine underlying physiological change, benefit is plausible; where it reflects practice, technique, better footwear or a better night's sleep, it reflects nothing about biology at all. Association is not causation, and a moved marker is not a moved outcome.

These are population instruments used on individuals. Every figure in the literature is a group average with wide individual scatter. The gradient that is real and reproducible across 34,485 people says relatively little about any one of them.

No compound or supplement is presented here as a lever on these measures, and that is deliberate. The evidence supporting the markers is observational; extrapolating from "this marker predicts mortality" to "this product improves this marker, therefore it improves mortality" chains two unproven steps onto one observed association. Where a marker is genuinely abnormal for someone's circumstances, the appropriate next step is clinical assessment of why - not a purchase. The section that follows lists options some people discuss with a physician, with the evidence for each stated plainly, and it does not claim any of them improves a functional marker except where that is established.

They do not replace medical assessment. Physical performance testing is an adjunct to clinical evaluation, not a substitute for it. Anyone with unexplained weakness, breathlessness, falls or a decline in walking ability should be assessed by a clinician rather than tracked with a stopwatch.

Where peptides fit: options to discuss with a physician

The markers on this page are worth working on, and the evidence-backed ways to work on them are training, protein, sleep and treating whatever underlying condition a poor result points to. Some people also ask about peptides and medicines alongside those, so the options are mapped below to the markers they are usually discussed for, with evidence and status stated plainly. One rule governs the whole section: apart from the GLP-1 medicines' established effects on body weight and metabolic markers, none of these products has been shown in humans to improve grip strength, gait speed, chair-stand performance or VO2max. Even where a marker does move, the point above still applies: a moved marker is not a moved outcome. Each option is a conversation for a physician who knows your history. Vivre Labs offers access, and we recommend beginning with baseline bloodwork (what to test).

Body composition and metabolic markers

• Semaglutide and tirzepatide are FDA-approved for type 2 diabetes and for chronic weight management in eligible adults. In large randomised trials they reduce body weight and improve glycaemic and other metabolic markers; that effect is established. Part of the weight lost is lean mass, which matters on a page where grip strength and chair-stand performance are markers: preserving muscle depends on resistance training and adequate protein, planned with your physician (GLP-1 and muscle loss).
• The metabolic and weight protocol pairs tirzepatide with tesamorelin. Tesamorelin is FDA-approved only to reduce excess abdominal fat in HIV-associated lipodystrophy, so its use here is off-label, and the evidence for the pair rests on each component separately rather than on a trial of the combination. Background in tesamorelin.

Muscle, strength and recovery (grip strength, chair-stand, gait)

• The lean mass and sleep quality protocol pairs CJC-1295, a GHRH analogue, with ipamorelin, a ghrelin-receptor agonist. Evidence: small human pharmacology studies show compounds in these classes raise growth hormone and IGF-1; no trial shows better grip strength, gait speed or physical function. Status: both investigational, not FDA-approved. Sermorelin is an older GHRH analogue whose FDA-approved product was discontinued in 2008 for reasons unrelated to safety.
• BPC-157, TB-500 and the GLOW protocol (BPC-157, TB-500 and GHK-Cu) are discussed for recovery from training and injury. Evidence: overwhelmingly animal studies of tendon, ligament and muscle; in humans, one uncontrolled knee-pain chart review with subjective outcomes and no control group. Status: investigational, not FDA-approved. An FDA advisory committee recommended BPC-157 and TB-500 for the 503A compounding list in July 2026; that vote is advisory, and neither has been added.

Mitochondrial energy (VO2max)

• MOTS-c is a peptide encoded in mitochondrial DNA. Evidence: animal studies suggest effects on metabolism and exercise capacity in mice; there is no completed interventional human trial. Status: investigational, not FDA-approved. Background in MOTS-c, the mitochondrial peptide.
• SS-31 (elamipretide) targets the inner mitochondrial membrane. Status: FDA-approved since September 2025 only for Barth syndrome; any other use is off-label. Evidence outside Barth syndrome is limited, and a Phase 3 trial in primary mitochondrial myopathy did not meet its walking-distance primary endpoint. See is SS-31 FDA-approved?
• The cellular energy and longevity protocol combines SS-31, MOTS-c and epithalon, so it carries the evidence and status of each: SS-31 off-label, MOTS-c and epithalon investigational.

Sleep

Sleep affects performance on every test on this page, which is one reason to test at a consistent time after a normal night. DSIP and epithalon are the peptides most often asked about. DSIP's human sleep data are small, old and inconsistent, and epithalon's rest mostly on small Russian studies and animal work. Both are investigational, and the FDA advisory committee voted 6 to 7 against recommending DSIP for the 503A list. See peptides for sleep.

Frequently asked questions

Are functional tests better than an epigenetic test? For predicting mortality and disability, the functional measures have a longer track record, larger pooled cohorts, direct validation against hard outcomes, and far better repeatability at essentially no marginal cost. For interrogating molecular biology, they say nothing at all. "Better" depends entirely on which question is being asked - and if the question is "will this number tell me something reliable about my trajectory that I can check again next quarter", the functional measures currently answer it more honestly.

Is grip strength really a longevity marker? It is a well-replicated predictor in large prospective cohorts, most prominently PURE, and it is best understood as a proxy for whole-body neuromuscular and nutritional status rather than as a property of the hand. It is not a longevity mechanism, its incremental value on top of conventional clinical risk factors is often modest, and reverse causation - illness causing weakness rather than weakness causing illness - has not been excluded.

Does raising my VO2max lower my mortality risk by the amount the studies show? No study has established that, and the studies are not designed to. The cohort data describe differences between people, and improvement studies describe change within people without random assignment. Higher fitness is plausibly protective and biologically coherent as a causal factor; the specific magnitude implied by cohort gradients should not be read as the effect an individual obtains by training. This article gives no training guidance for the same reason it gives no thresholds: that conversation belongs with a clinician who knows your cardiovascular status.

Why does this article not tell me what a good result is? Because published reference values are population- and protocol-specific, they shift with age, sex, equipment and testing method, and a number presented without that context invites exactly the wrong response - self-diagnosis, or reassurance that is not warranted. Interpretation of a functional result is a clinical task performed against the right reference for the person in front of you.

Related reading

• What a biological age test actually measures
• Epigenetic clocks
• Organ-specific ageing clocks
• Can you lower your epigenetic age?

Part of The Complete Guide to Biological Age

Explore it with verified supply. Every Vivre Labs lot is independently tested and the certificate of analysis is published, so you know exactly what is in the vial.

---

This article is educational and does not constitute medical advice. No doses, dosing schedules, training prescriptions, target values or diagnostic thresholds are provided; compounds and protocols named in the physician-discussion section are linked for information, not prescribed. The evidence described is predominantly observational: it establishes association between functional measures and subsequent outcomes in human cohorts, not causation, and does not establish that improving a measure produces the outcomes observed in better-performing groups. Functional testing does not diagnose disease and does not replace clinical assessment. Anyone with unexplained weakness, breathlessness, falls or declining mobility should consult a physician. Claims for the Malaysian market are subject to MAB requirements.

References

• Studenski S, Perera S, Patel K, et al. "Gait Speed and Survival in Older Adults." JAMA 2011;305(1):50–58. PubMed 21205966. Pooled individual-participant analysis of 9 cohorts; 34,485 community-dwelling adults aged 65+; follow-up 6–21 years.
• Leong DP, Teo KK, Rangarajan S, et al. "Prognostic value of grip strength: findings from the Prospective Urban Rural Epidemiology (PURE) study." The Lancet 2015;386(9990):266–273.
• Sayer AA, Kirkwood TBL. "Grip strength and mortality: a biomarker of ageing?" The Lancet 2015 (commentary accompanying PURE).
• "Associations of handgrip strength with all-cause and cancer mortality in older adults: a prospective cohort study in 28 countries." Age and Ageing 2022;51(5):afac117.
• "Comparison of grip strength measurements for predicting all-cause mortality among adults aged 20+ years from NHANES 2011–2014." Scientific Reports 2024; s41598-024-80487-y.
• Mandsager K, Harb S, Cremer P, et al. "Association of Cardiorespiratory Fitness With Long-term Mortality Among Adults Undergoing Exercise Treadmill Testing." JAMA Network Open 2018;1(6):e183605. PMC6324439.
• Clausen JSR, Marott JL, Holtermann A, et al. "Midlife Cardiorespiratory Fitness and the Long-Term Risk of Mortality: 46 Years of Follow-Up." Journal of the American College of Cardiology 2018.
• Laukkanen JA, Zaccardi F, Khan H, et al. "Long-term Change in Cardiorespiratory Fitness and All-Cause Mortality: A Population-Based Follow-up Study." Mayo Clinic Proceedings 2016. PubMed 27444976.
• Guralnik JM, Simonsick EM, Ferrucci L, et al. "A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission." Journal of Gerontology 1994.
• Cooper R, Kuh D, Hardy R. "Objectively measured physical capability levels and mortality: systematic review and meta-analysis." BMJ 2010;341:c4467.
• "Predictive value of sarcopenia components for all-cause mortality: findings from population-based cohorts." Aging Clinical and Experimental Research 2024. PMC11156728.
• Higgins-Chen AT, Thrush KL, Wang Y, et al. "A computational solution for bolstering reliability of epigenetic clocks: implications for clinical trials and longitudinal tracking." Nature Aging 2022 - documents technical noise in conventional methylation clocks.