The benefits of strength training after 40: what the evidence actually shows

Applied Physiology #2 - A reference page on what resistance work changes in a body past forty, with the eight studies it rests on and the limits of each one.

Second in the applied physiology series. Part one set out recovery capacity as an account you learn to read; this page asks which demand on that account pays back hardest, and part three will take apart the difference between how long you sleep and how your night is built.

Put two muscle biopsies side by side, one from a man of seventy and one from a man of twenty, and the shortfall is not distributed the way intuition suggests. Fibre counts hold up. What has gone is size, and almost all of it has gone from a single class of fibre, the class a comfortable week never calls on.

That is the hinge for everything below, because it means the session most people cancel first is aimed precisely at the thing that erodes. Skeletal muscle also happens to be the heaviest organ a person carries, and virtually none of its work is legible in a mirror.

Hour for hour, nothing else you can do moves as many downstream systems as loading a muscle hard enough to matter. That is the claim, and it is deliberately bounded. Aerobic and mobility training keep their jobs, and lifting cannot do either of them.

This is an educational and strategic perspective, not personal medical advice.

The short version

• Ageing empties one compartment, the type II fibres, and leaves fibre numbers broadly alone.

• Under insulin stimulation, skeletal muscle absorbs about four fifths of the glucose leaving your bloodstream, which is why it turns up in bloodwork.

• Bone responds to a narrow specification, high strain delivered fast, and little outside resistance and impact work meets it.

• Grip strength forecasts mortality better than systolic blood pressure. An association, not a lever anyone has pulled in a trial.

• Running interferes with strength gains; cycling largely does not.

• Every study below is observational or short-term. None followed anyone for a lifetime.

Muscle turned out to be an endocrine organ

For most of the twentieth century muscle was filed as plumbing: it moved the skeleton, it burned fuel, end of description. The reclassification happened quietly and most people training never heard.

Pedersen and Febbraio gathered the case that skeletal muscle manufactures and releases several hundred signalling peptides, now grouped as myokines. Some act on the muscle that made them. Others enter the blood and reach liver, pancreas, adipose tissue, bone and brain (Pedersen and Febbraio, 2012). A tissue doing that is in standing communication with nearly every system that fails by degrees.

Then the storage number, which is the one worth carrying around. When insulin is doing its job, skeletal muscle accounts for something close to 80 percent of the glucose cleared from your blood, and it is on exactly that basis that DeFronzo and Tripathy nominate muscle insulin resistance as the first thing to break in type 2 diabetes, years ahead of any reading that would worry a doctor (DeFronzo and Tripathy, 2009).

Which reframes what muscle is for. It is much of the room your body has to put a meal, and it explains why lifting shows up in panels that look unrelated to a gym.

Only one kind of fibre disappears

Nilwik and colleagues measured cross-sectional area in leg muscle across young and elderly men, then went looking in the biopsies for the missing tissue. They found fibre counts broadly preserved and type II fibres substantially smaller (Nilwik et al., 2013). Sarcopenia is the clinical name for muscle wasting with age, and underneath it looks like this: one compartment quietly emptying rather than an even fade.

Type II fibres handle anything fast or heavy, and they are recruited last, which is a sensible design given how rarely ordinary life needs them. A chair does not. A pavement does not. Nor does a comfortable hour on a bicycle, which is the part that tends to start an argument.

A tissue nobody requisitions is a tissue the body stops funding.

What that costs is not bulk so much as the rate at which you can produce force. It arrives as a specific kind of moment: a stair lower than you read it, and a leg taking the whole correction in a fraction of a second. Neither that nor a deep sofa asking for an arm is athletic. Both test a reserve without warning.

The control system fails before the tissue does

A second decline runs one level above the muscle, in the wiring that drives it.

Each fibre reports to a motor neuron. That neuron plus every fibre under it makes a motor unit, and a motor unit is the smallest thing a nervous system can switch on or leave alone. Neurons die off across a life, and every one that goes leaves its fibres unclaimed.

There is a repair route. Surviving neurons nearby send out sprouts, take the orphaned fibres onto their books, and the remaining units get bigger, which is how a muscle keeps working with fewer commanders. Piasecki and colleagues went after the point where that repair runs out, using intramuscular electromyography on two leg muscles across four groups: young men, then older men separated by whether sarcopenia was absent, developing, or established. Without it, units had enlarged as predicted. With it, that enlargement had not happened, and the units measured smaller than in older men who were not sarcopenic (Piasecki et al., 2018).

Read that as a map of where the system gives way, not as a case for anything. It is a snapshot across groups, all male, and nobody was trained and retested.

So strength is really two separate questions: how much tissue is there, and how much of it can still be called up. Loading heavily and recovering properly works on the second one. That is why a beginner's numbers climb long before anything about them looks different, and why the early progress is neural rather than structural.

Protein stops working as well as it did

Feed an older body the same protein and you get a smaller answer. The reason is specific, and it is not a vague slowing of metabolism.

Cuthbertson and colleagues gave graded amounts of essential amino acids to forty-four healthy men, young and old, matched on build, and measured muscle protein synthesis directly rather than inferring it. At rest the two groups were indistinguishable, which kills off the idea that old muscle simply idles lower. What differed was what happened after feeding. Older tissue was both harder to trigger and weaker in its reply, and the intracellular sensing apparatus, eIF2B, eIF4BP-1, p70 S6 kinase and mTOR among them, showed less expression and weaker activation, with NF-kappaB rising sharply alongside (Cuthbertson et al., 2005).

This has a name, anabolic resistance. The message has to be delivered louder before older tissue accepts delivery.

The reading that follows directly concerns protein. Both the sensitivity and the ceiling drop, so a given meal has to carry more to clear the same bar, and a graded-dose trial is exactly the instrument built to establish that.

The training reading is a step beyond what this work tested. Anabolic resistance is a fair argument for load heavy enough to tax the tissue, as against the token weight handed round a general class, because anything under threshold buys almost none of what anything over it does. But amino acids were the intervention here, not barbells, so treat that as a defensible extension and not as a finding. It is the kind of inference that gets laundered into a result within about two citations.

Bone is listening for something very specific

General movement does not maintain a skeleton, which is why so much sincerely given advice on bone density goes nowhere. The pathway is mechanotransduction, where physical strain gets converted into a signal a cell can act on, and the strain has to be both high in magnitude and fast in delivery before anything registers as worth building for.

The LIFTMOR trial went at that specification directly. A hundred and one women past menopause with already low bone mass were randomised across eight months, one arm to supervised heavy resistance work combined with impact loading, twice weekly, thirty minutes, five sets of five at better than 85 percent of a single-repetition maximum, the other to a gentle home routine. Lumbar spine density gained 2.9 percent under training. The control arm lost 1.2 percent over the same period. Cortical thickness and femoral neck density both shifted the way you would want. Functional measures improved across the board. In eight months of heavy loading in a population long assumed too fragile for it, the safety record ran to one entry, a mild back spasm (Watson et al., 2018).

The participants are not the reader of this page, and the percentages do not transfer. What transfers is the specification: bone answers to a narrow quality of load, and outside impact work there is essentially nowhere to find it except resistance training. Note that LIFTMOR delivered both, which is a detail usually dropped when this trial is cited as an argument for lifting alone.

What grip strength does and does not tell you

Grip looks trivial to measure until you see its record as a predictor. The PURE study attached a hazard ratio of 1.16 for death from any cause to every five kilograms of grip lost, and found grip out-predicting systolic blood pressure for both all-cause and cardiovascular death (Leong et al., 2015).

Now for the part that falls off whenever this result gets repeated. It is an association. No trial has raised anyone's grip and measured what happened to their risk, which would be an entirely different and much harder study to run. Grip is standing in for a great deal at once: how well someone eats, how much they have moved across decades, and whether something is already wrong that nobody has named yet. A person can be weak because illness has started. Observational data cannot untangle that, and no amount of statistical adjustment makes a cohort into a trial.

Take the smaller conclusion and it still changes your behaviour. Strength belongs on the same page as lipids and blood pressure: something recorded, watched across years, and looked into when it starts moving the wrong way.

What it costs, and where the hours come from

Two genuinely heavy sessions land between ninety minutes and two hours a week once warm-ups and the rest between sets are counted without flattery. Those hours come out of a week already promised to endurance work most people would rather keep.

Wilson and colleagues put a number on the collision, pooling twenty-one studies and four hundred and twenty-two effect sizes. Training strength on its own returned a hypertrophy effect size of 1.23, against 0.85 for identical work done alongside endurance training. Power showed a similar penalty. Both how often and how long the endurance sessions ran tracked inversely against improvements in size, in strength and in power. The mode of endurance mattered more than most people expect, in that running blunted hypertrophy and strength significantly while cycling did not (Wilson et al., 2012).

A second cost barely appears in training research. Part one of this series described a single autonomic account that work, broken sleep, emotional load and training all draw down together, and heavy lifting draws on that account like everything else does. Add two hard sessions to an unchanged week in a punishing month and you will not get the response a calm month would have handed you.

Then the timescale, which is the honest reason most people give up. Adaptation arrives across months, almost no single session feels like anything, and a strong year buys a marginally heavier bar plus a change in shape that only shows up between photographs taken far apart.

None of which argues for dropping everything else. Aerobic work builds parasympathetic capacity, and lifting will not do that job. Mobility work answers a narrower question of its own. And if you have been chasing a single number as a proxy for fitness, the limits of VO2 max as a predictor are worth reading alongside this. Strength is simply the component most often missing.

What each study actually covered

Every finding on this page comes with a boundary, and the boundaries matter more than the effect sizes. This is what each piece of evidence can carry and what it cannot.

Two limits run across the whole set. The first is that most of this literature was built on men. LIFTMOR is the exception, and the only trial here with women in it. Applying the motor unit or amino acid findings to a woman is extrapolation, and should be named as such.

The second is duration. The longest intervention here ran eight months, so every claim about decades is inference stitched from short trials and cohorts. That is the normal situation in this field rather than a scandal, but it is not the same thing as evidence. What would change my reading is a properly powered resistance trial with hard outcomes over years. Nobody can blind one and nobody will fund one.

Two checks worth running this week

One question tells strength training apart from the activity that looks like it. Do you know what went on the bar for your main lifts three months ago, and does today's number beat it?

If the first half has no answer, then no record exists, and unrecorded progress has a habit of not occurring. If you can answer and nothing has moved, you have been maintaining a habit rather than delivering a stimulus, and the fibres, the units and the skeleton finished responding to that message a while back.

The second check needs thirty seconds and no equipment. Rise five times from a low seat, hands off, and watch whether the fifth attempt has visibly degraded against the first. What comes back is a reading on your present reserve. Character does not enter into it, and this is the sort of measure that improves when you attend to it.

Questions people actually ask

Is it too late to start lifting at 60 or 70?

No, and LIFTMOR is the direct evidence: women past menopause with already low bone mass trained heavily for eight months with one minor adverse event between them. What changes with age is how much signal it takes to get a response, which argues for competent supervision at the start rather than for waiting.

How much protein do I actually need per meal?

The honest answer is that this page cannot give you a number. What the evidence supports is the direction: older muscle answers a given dose less readily and less completely, so whatever sufficed at thirty is unlikely to suffice at sixty. A specific target belongs to someone looking at your intake, your kidney function and your training, not to an article.

Can I get the same benefit from bodyweight work or classes?

Partly, and it depends entirely on whether the load reaches threshold. The type II fibres are recruited last and only for real force, so anything comfortable leaves them unaddressed however long it lasts. Bodyweight work that is genuinely hard for you counts. A class calibrated to be manageable for everyone in the room generally does not.

Should I stop running so my lifting works?

Almost certainly not. Running interferes and cycling largely does not, but that meta-analysis compares against training strength in isolation, which is not the life most readers have. If both matter, the levers are separating the sessions, keeping the truly hard efforts few, and watching how the week is built rather than dropping a modality.

Is grip strength worth measuring at home?

As a tracked number over years, yes, and a cheap dynamometer will do. As a one-off reading to panic about, no. The finding concerns populations and trajectories, and one measurement of yours carries almost none of what the cohort carried.

Where Atlas Cove fits

An Atlas Cove week puts the strength work next to the recovery data rather than in its own box, so what somebody loads follows from the state they turned up in and not from a plan drafted before anyone met them. What should outlive the flight home is small: load on purpose, then check whether it landed.

Muscle reads easily as decoration and badly as an organ, which is exactly why it goes first when a diary fills. Read as an organ, the accounting looks entirely different. The work needed to keep it is unremarkable in hours and in feel, which is almost certainly why it goes on being skipped.

Sources

1. Nilwik, R., Snijders, T., Leenders, M., Groen, B. B. L., van Kranenburg, J., Verdijk, L. B., and van Loon, L. J. C. (2013). The decline in skeletal muscle mass with aging is mainly attributed to a reduction in type II muscle fiber size. Experimental Gerontology, 48(5), 492 to 498. DOI: 10.1016/j.exger.2013.02.012

2. Piasecki, M., Ireland, A., Piasecki, J., Stashuk, D. W., Swiecicka, A., Rutter, M. K., Jones, D. A., and McPhee, J. S. (2018). Failure to expand the motor unit size to compensate for declining motor unit numbers distinguishes sarcopenic from non-sarcopenic older men. The Journal of Physiology, 596(9), 1627 to 1637. DOI: 10.1113/JP275520

3. Cuthbertson, D., Smith, K., Babraj, J., Leese, G., Waddell, T., Atherton, P., Wackerhage, H., Taylor, P. M., and Rennie, M. J. (2005). Anabolic signaling deficits underlie amino acid resistance of wasting, aging muscle. The FASEB Journal, 19(3), 422 to 424. DOI: 10.1096/fj.04-2640fje

4. Pedersen, B. K., and Febbraio, M. A. (2012). Muscles, exercise and obesity: skeletal muscle as a secretory organ. Nature Reviews Endocrinology, 8(8), 457 to 465. DOI: 10.1038/nrendo.2012.49

5. DeFronzo, R. A., and Tripathy, D. (2009). Skeletal muscle insulin resistance is the primary defect in type 2 diabetes. Diabetes Care, 32(Suppl. 2), S157 to S163. DOI: 10.2337/dc09-S302

6. Watson, S. L., Weeks, B. K., Weis, L. J., Harding, A. T., Horan, S. A., and Beck, B. R. (2018). High-intensity resistance and impact training improves bone mineral density and physical function in postmenopausal women with osteopenia and osteoporosis: the LIFTMOR randomized controlled trial. Journal of Bone and Mineral Research, 33(2), 211 to 220. DOI: 10.1002/jbmr.3284

7. Leong, D. P., Teo, K. K., Rangarajan, S., Lopez-Jaramillo, P., Avezum, A., Orlandini, A., et al. (2015). Prognostic value of grip strength: findings from the Prospective Urban Rural Epidemiology (PURE) study. The Lancet, 386(9990), 266 to 273. DOI: 10.1016/S0140-6736(14)62000-6

8. Wilson, J. M., Marin, P. J., Rhea, M. R., Wilson, S. M. C., Loenneke, J. P., and Anderson, J. C. (2012). Concurrent training: a meta-analysis examining interference of aerobic and resistance exercises. Journal of Strength and Conditioning Research, 26(8), 2293 to 2307. DOI: 10.1519/JSC.0b013e31823a3e2d

This is an educational and strategic perspective, not personal medical advice.