Strength Training for Older Adults: The Evidence and the Dose

Lokh, founder of Forma. 31 July 2026.

Your father started using the handrail on the stairs this year. He would say it is nothing. It is not nothing, and it is one of the most reversible things that will happen to him this decade.

Grip going on jar lids, a pause before standing up from a low chair, a walk that has quietly got slower. These are not the inevitable cost of a birthday. They are the visible surface of muscle and power that has been draining away for thirty years, and resistance training is the only intervention with good evidence of putting some of it back.

This page is written for three people at once: someone in their sixties or seventies deciding whether to start, a son or daughter trying to work out what to suggest, and a trainer who wants the dosing straight. The evidence is the same for all three.

What the research actually supports

The claims worth making are narrower than the ones usually made, and they are strong enough without inflation.

Strength and muscle mass both improve, at any age studied. Progressive resistance training produces measurable gains in adults in their seventies, eighties and nineties, including in frail nursing home residents. There is no age at which the response disappears.

Function improves alongside it. Gait speed, chair-rise time and stair climbing all respond. These matter more than a number on a barbell, because they are what independence is actually made of.

Falls become less likely, particularly when balance work is included alongside the strength training. Given that a hip fracture is one of the sharpest turning points in an older person's life, this is the outcome that justifies the whole enterprise.

Where to be careful: much of this research uses supervised programs, and adherence in a study is better than adherence at home. Effect sizes for muscle mass are more modest than the strength numbers suggest, because a lot of early strength gain is neural rather than new tissue. That is not a problem, it is just worth knowing so the first eight weeks are not oversold.

The dose

Two sessions a week is the widely recommended minimum and it is a real threshold rather than a round number. Three is better if recovery allows.

Each session: five or six movements covering the major patterns, two to three sets of eight to twelve reps, stopping two or three reps short of the point where form degrades. Thirty to forty minutes including a warm-up.

• Sit-to-stand or a squat variation. The single most transferable movement in the whole list, because it is the one being lost.
• A hinge. Hip thrust, or a light Romanian deadlift once the pattern is solid.
• A push and a pull. Machine chest press and seated row are perfectly good starting points and remove the balance demand while the strength is built.
• A loaded carry. Walking with weight in one or both hands. Trains grip, trunk and gait at once, and grip strength is one of the better predictors of how the next decade goes.
• Balance work. Single-leg stands, tandem stance, progressing to eyes closed with support nearby. Five minutes, every session.

What heavy means here

The most common mistake in this population is training too light for too long, and it usually comes from good intentions.

Heavy is relative to current capacity, and that is the only definition that means anything. For a deconditioned 78-year-old, standing up from a chair without using the arms may be a genuine near-maximal effort, and that counts. For a 65-year-old who has lifted for years, it may be a loaded barbell. The principle does not change with the numbers: the last repetition of a set should be difficult, and the weight should increase over time.

Two or three sets of a weight that could have been lifted twenty times produces very little. It is comfortable, it feels productive, and it is most of why some people train for a year without changing anything.

Starting safely

Safely does not mean lightly. It means competently, and in the right order.

Spend the first two to three weeks establishing movement patterns with minimal load, then add weight steadily. Machines before free weights if balance or confidence is a concern, since they remove one variable while the strength arrives. Full range of motion at a lighter weight always beats a partial range at a heavier one.

Anyone with uncontrolled hypertension, recent cardiac events, significant osteoporosis or a joint replacement should get individual clearance and a program adjusted for it, rather than a page on the internet. That is a real caveat, not a legal formality, and it applies to a meaningful share of this age group.

Progress by adding a small amount of weight when every set reaches the top of the rep range with good form and reps in reserve. Slowly. There is no deadline and nothing is gained by rushing the first three months.

Protein, briefly

Training without enough protein produces a fraction of the result, and under-eating protein is close to universal in this age group. Appetite drops, meals get smaller, and breakfast in particular tends to be almost entirely carbohydrate.

Around 1.6 grams per kilogram of bodyweight a day, with 30 to 40 grams at each main meal. The per-meal figure matters because the muscle-building response to a small dose is blunted with age, so three solid servings beat the same total dribbled across the day. For most people the fix is adding protein to breakfast and nothing else changes.

Common questions

How heavy is heavy enough?

The last of eight reps should be genuinely difficult while leaving two in reserve. What that corresponds to in kilograms is entirely individual and completely beside the point.

Is this safe at 70 or 80?

Supervised progressive resistance training has a good safety record in this age group, including among people with multiple chronic conditions. The honest comparison is not against doing nothing safely, it is against the falls and lost independence that follow continued muscle loss.

How long until something changes?

Strength shifts noticeably in four to six weeks, mostly through neural adaptation before much tissue is added. Chair-rise and balance often improve on a similar timeline, and for most people that is the change that actually registers.

Where a wearable helps, and where it does not

If you are starting from a deconditioned baseline, none of this needs technology. Two sessions a week, done properly, is the whole intervention, and a watch adds nothing to it.

Where it becomes useful is later, once you are training regularly and also walking, cycling or swimming, and the question becomes whether today is a good day for a hard session. Recovery windows are longer at this age and less predictable week to week, and a bad night genuinely changes what you should do.

Forma reads overnight HRV, sleep and training load from a Garmin watch and returns one session sized to that. Recovery, not effort. It is a tool for the maintenance phase rather than the starting phase, and it is worth saying plainly that the first six months matter far more than anything it does.

For someone in their fifties who wants to train hard rather than maintain, building muscle after 50 covers the same ground more aggressively. If endurance work is also in the picture, strength training for runners has the placement rules. Everything is on the training guides hub.