Resistance training is the best-supported way to stay strong, keep muscle and bone, and stay independent as you age. Most of the benefit arrives at a strikingly low dose. In trained men, one hard set two or three times a week added roughly 12 kg across the big lifts in a couple of months. People who lift also outlive those who avoid exercise, a pattern seen in large population studies that no trial has proven. Older adults gain the most: the frailest nursing-home residents in their late eighties more than doubled their strength in ten weeks.
It lowers long-term blood sugar in type 2 diabetes, eases depression and anxiety, and builds bone after menopause. Bodyweight alone is enough for most people to start, at no cost and with no equipment. Two habits keep it safe and working: add load gradually, and breathe out as you push through each rep. Most people can start today with a squat to a chair and a push-up against the kitchen counter.
Findings & Outcomes
What It Is
Resistance training means working your muscles against something they must overcome: your own bodyweight, a band, a dumbbell, or a loaded barbell. You load a muscle hard enough that it must adapt, then keep raising the load as it does. Bodyweight costs nothing and needs no equipment. Bands, dumbbells and a gym are steps you grow into as bodyweight gets easy.
Few things you can do carry as much evidence or change as many separate things at once. Resistance training:
- builds strength and muscle
- keeps bone
- lowers long-term blood sugar
- eases low mood
- tracks with living longer and staying independent
Most people get the dose wrong. The amount that produces measurable gains in trials is far smaller than gym culture suggests. It comes down to two choices: how hard you load a muscle, and how often.
What It Does
Two kinds of evidence sit behind these findings. Controlled trials show cause and effect; cohort studies follow large groups and reveal patterns but cannot prove cause. The trials start with how little it takes. In men who already lift, one hard set of six to twelve reps added about 26 lb (12 kg) across the main lifts. The schedule was two or three sessions a week for eight to twelve weeks. Muscle grows faster trained twice a week than once. For size, light loads grow muscle as much as heavy ones when each set is taken close to failure. Heavier loads build more raw strength, so the weight you pick follows your goal.
The same training reaches well past the muscle. In type 2 diabetes it lowers HbA1c, the long-term blood-sugar marker, by about 0.4 percentage points. That is close to what clinicians expect from adding a second glucose-lowering drug on top of the first. Type 2 diabetes is largely driven by modifiable factors (body weight, diet and inactivity) so training addresses a root cause the medication only manages. Metformin, the usual first drug, lowers HbA1c two to three times more, so training earns its place beside the medication. The people who got strongest saw the biggest drops. It also eases depression: for every four people who train, about one more improves meaningfully than would have without it. It lowers anxiety by a smaller but reliable amount, and resting blood pressure by a few points, most in people whose pressure is borderline high.
Older adults gain the most of anyone, the reverse of the usual advice to train them gently. Frail nursing-home residents averaging 87 years old raised their muscle strength 113% in ten weeks. They walked faster and climbed stairs with more power, while a nutrition supplement alone changed nothing. Postmenopausal women with thinning bones, the group most often told to avoid heavy lifting, gained spine bone density over eight months of supervised heavy training. The whole trial produced one minor injury.
People who lift have about 21% lower death rates from any cause than people who do no exercise. That comes from cohort data, the weakest design here. The figure rises to about 40% in those who also do some cardio. And every 11 lb (5 kg) of grip strength lost carries about a 16% higher death rate. These come from watching large groups, so they show a consistent pattern no trial has proven. People who lift differ from people who do not in diet, wealth and health, and no study can separate the training from the person.
Resistance training does relatively little for cardiorespiratory fitness, one of the strongest predictors of a long life. So it should not replace cardio; the biggest drop in death rates came in people who did both. It does not replace balance training or daily movement, and it does not fix a poor diet.
The Research & Studies
Everything here is based on the research we have collected and checked, sorted into groups and ordered with the strongest evidence first. Click any claim to open the studies behind it.
Muscle And Strength
One hard set, 2 to 3 times a week, added about 26 lb (12 kg) across the main lifts
One hard set of 6 to 12 reps at a challenging weight, done two to three times a week for a couple of months, added about 26 lb (12 kg) across the main lifts in men who already trained. That is a small time cost for a worthwhile gain, though the researchers treat it as a floor, not a target.
Meta-analysis of 5 of 6 studies in resistance-trained men found that a single set of 6 to 12 reps at roughly 70 to 85% of a one-rep maximum, performed 2 to 3 times a week with high effort for 8 to 12 weeks, produced a pooled 1RM increase of 26.7 lb (12.09 kg) (95% CI 8.16 to 16.03), made up of 38.5 lb (17.48 kg) on the squat and 18.2 lb (8.25 kg) on the bench press. The base is small, the participants already trained, and the authors describe the gains as suboptimal, not as a goal. There is no data for the deadlift, and none in trained women or highly trained strength athletes, so a woman using this protocol is applying a number produced entirely in men.
Who this may not transfer to:The review states explicitly that it is unclear whether these gains are achievable in trained women, because no eligible study measured them. A woman applying this protocol is using a number produced entirely in men, and the direction and size of any difference are unmeasured, not known to be small.
The study · 1
Androulakis-Korakakis, Fisher and Steele, the minimum effective training dose required to increase 1RM strength in resistance-trained men · Sports Med 2020;50(4):751-765
Training a muscle twice a week grows it more than once a week
Training a muscle twice a week grew it more than once a week when the total amount of work was the same. For building maximum strength, three times a week came out on top.
At matched weekly volume, training a muscle group twice a week produced larger hypertrophy effect sizes than once a week (0.49 versus 0.30) across 10 direct comparisons. A Bayesian network meta-analysis of 178 strength trials and 119 hypertrophy trials (5,097 participants for strength, 3,364 for hypertrophy) ranked higher-load multiset thrice-weekly training best for strength and higher-load multiset twice-weekly training best for size. The samples ran 45% women for strength and 47% for hypertrophy, which makes this one of the better sex-balanced findings on the page.
Who this may not transfer to:Women were 45% of the strength sample and 47% of the hypertrophy sample.
The studies · 2
Schoenfeld, Ogborn and Krieger, effects of resistance training frequency on measures of muscle hypertrophy · Sports Med 2016;46(11):1689-1697
Currier et al., resistance training prescription for muscle strength and hypertrophy in healthy adults, a Bayesian network meta-analysis · Br J Sports Med 2023;57(18):1211-1220
Each added weekly set grew muscle about 0.4% more
Doing more sets per week grew muscle a little more, roughly 0.4% extra growth for each added weekly set. The effect is small but consistent, and the study cannot say at what point adding more sets stops paying off.
Across 34 treatment groups from 15 studies, each additional weekly set per muscle group was associated with an effect-size increase of 0.023, about 0.37% more growth, and comparing the higher- against the lower-volume arm within each study gave an effect-size difference of 0.241, about 3.9% more growth. The slope was fitted over the range the studies happened to use, so it says nothing about where the curve turns down. The three-band comparison (under 5, 5 to 9, and 10 or more weekly sets) reached only a trend (p = 0.074), weaker than the way this finding is usually quoted. No pooled sex composition is reported across the 15 studies.
Who this may not transfer to:No pooled sex composition is reported across the 15 studies. Volume tolerance is one of the variables most plausibly affected by recovery differences, and it has not been compared between the sexes at this scale.
The study · 1
Schoenfeld, Ogborn and Krieger, dose-response relationship between weekly resistance training volume and increases in muscle mass · J Sports Sci 2017;35(11):1073-1082
Light loads build as much muscle as heavy ones when sets go to failure
For building muscle size, light weights grew muscle about as well as heavy ones, as long as each set was taken to the point where another rep is impossible. Heavy weights still built more maximal strength.
Across 21 studies comparing low-load and high-load training with sets taken to momentary failure, changes in muscle size were similar between loads at or below 60% of a one-rep maximum and loads above it. One-rep-max strength gains were significantly greater with heavy loads; isometric strength showed no significant difference. The size equivalence holds only because effort was matched to failure: a light set stopped well short of failure is not equivalent to a heavy one. No pooled sex composition is reported, and women tend to sustain more reps at a given percentage of their maximum than men, so how load equivalence behaves by sex is unmeasured here.
Who this may not transfer to:No pooled sex composition is reported. Women tend to sustain more repetitions at a given percentage of their maximum than men do, so how load equivalence behaves by sex is a reasonable question this analysis cannot answer.
The weight matters less for muscle size than gym advice implies, so lighter loads work if you push each set close to failure, but light-load-to-failure is considerably more unpleasant and is a compliance problem, not a physiological one. If your goal is raw strength, or you would rather not grind out long taxing sets, heavier loads are the better trade.
The study · 1
Schoenfeld, Grgic, Ogborn and Krieger, strength and hypertrophy adaptations between low- and high-load resistance training · J Strength Cond Res 2017;31(12):3508-3523
Stopping a rep or two short builds as much muscle as going to failure
Grinding every set to the point of total failure did not build more muscle than stopping a rep or two short.
Across 15 studies in healthy adults, training to momentary muscular failure showed no advantage over non-failure training for muscle growth (effect size 0.12, p = 0.343). Training to a looser definition of set failure gave a trivial edge (0.19, p = 0.045), and velocity-loss thresholds showed no significant advantage (0.08, p = 0.529). The authors describe the relationship between proximity to failure and growth as potentially non-linear. The trials split close to evenly between untrained (7 studies, 252 people) and trained lifters (8 studies), so this is not a trained-lifter finding, and judging reps in reserve is a skill untrained lifters are systematically poor at.
Who this may not transfer to:No pooled sex composition is reported across the 15 studies, and this subfield is heavily male. Fatigue and recovery from failure training are plausible places for a sex difference, and it has not been measured at this scale.
Since stopping a rep or two short grew muscle just as well while leaving you far less drained, there is no need to take every set to failure, so save that effort for the occasional set, not the whole session. If you are new, the occasional set to failure is a useful way to learn what your true limit feels like, because untrained lifters judge how many reps they have left poorly.
The study · 1
Refalo et al., influence of resistance training proximity-to-failure on skeletal muscle hypertrophy · Sports Med 2023;53(3):649-665
Frail 87-year-olds gained 113% in strength in ten weeks
Frail nursing-home residents averaging 87 years old raised their muscle strength 113% in ten weeks, walked faster and climbed stairs with more power. Nutrition supplements alone changed nothing. Older adults are the group that gains the most, not the group to train gently.
In a randomized trial of 100 frail nursing-home residents (63 women, 37 men; mean age 87.1, range 72 to 98), ten weeks of progressive resistance training raised muscle strength 113% against 3% in the non-exercising groups, gait velocity 11.8% against a 1.0% decline, and stair-climbing power 28.4% against 3.6%, while multinutrient supplementation alone changed no primary outcome. An earlier uncontrolled study in ten nonagenarians (mean age 90) found 174% strength gains over eight weeks. Ten weeks is short and the setting was supervised institutional care. Thigh muscle area rose only 2.7% and did not reach significance, so most of the strength gain over this window is the nervous system, not new tissue.
Who this may not transfer to:63% women, which is representative of who actually lives in nursing homes and unusually well balanced for a strength training trial. The result is reported pooled, so a sex-specific effect size cannot be read off it.
The studies · 2
Fiatarone et al., exercise training and nutritional supplementation for physical frailty in very elderly people · N Engl J Med 1994;330(25):1769-75
Fiatarone et al., high-intensity strength training in nonagenarians · JAMA 1990;263(22):3029-34
Extra protein added about 0.7 lb (0.3 kg) of muscle, plateauing near 1.6 g/kg a day
Taking extra protein while training added a little more muscle, about 0.7 lb (0.3 kg) over a training block, and the benefit stopped climbing at around 1.6 grams of protein per kilogram of bodyweight a day.
Across 49 studies (1,863 participants), protein supplementation during prolonged resistance training added about 0.7 lb (0.30 kg) of fat-free mass and increased one-rep-max strength beyond training alone. The dose-response plateaued at approximately 1.62 g of protein per kilogram of bodyweight per day, with wide confidence intervals at the top end, and the size of the benefit fell as participant age rose. The effect is measured against control diets that were often already adequate. The senior author has run trials funded by a dairy-industry body that are themselves inside this meta-analysis, and the constituent trials are predominantly young men, so the 1.62 g/kg figure is a number derived largely from men.
Who this may not transfer to:The review does not report a pooled sex breakdown, and the constituent trials are predominantly young men. Combined training-plus-protein interventions have rarely been analyzed by sex at all, so the 1.62 g/kg plateau should be read as a number derived largely from men.
Around 1.6 g of protein per kilogram of bodyweight a day is where the added muscle stops rising in this data, so eating past that is unlikely to add more. The gain itself is small and measured against diets that were often already sufficient, so protein helps most when your current intake is low.
The studies · 2
Morton et al., systematic review, meta-analysis and meta-regression of the effect of protein supplementation on resistance training-induced gains in muscle mass and strength · Br J Sports Med 2018;52(6):376-384
Bauer et al., evidence-based recommendations for optimal dietary protein intake in older people, the PROT-AGE position paper · J Am Med Dir Assoc 2013;14(8):542-59
Men and women over 50 adapt similarly to the same program
Men and women over 50 following the same program adapted similarly. Women gained a little more lower-body strength relative to where they started; men gained more in absolute pounds lifted. Which sex 'gained more' depends entirely on whether you measure the change in relative or absolute terms.
Across 30 studies (41 comparison groups, 1,410 participants: 651 men, 759 women, all over 50) on identical programs, women gained more relative lower-body strength (g = -0.21, 95% CI -0.33 to -0.10). Relative upper-body strength showed no significant difference (g = -0.29, p = 0.08) and relative muscle size no difference (g = 0.10, p = 0.16). Men gained more in absolute terms for upper-body strength (g = 0.48), lower-body strength (g = 0.33) and size (g = 0.45). Mean study quality was 14.7 of 29 on a modified Downs and Black checklist, and the analysis covers adults over 50 only, so it does not answer the same question for younger trainees.
Who this may not transfer to:651 men and 759 women in trials that ran both on the same protocol, which is the design that can actually answer this and is rare. The equivalent analysis in younger adults does not exist at this scale.
The study · 1
Jones et al., sex differences in adaptations in muscle strength and size following resistance training in older adults · Sports Med 2021;51(3):503-517
Cold water right after lifting cut strength and muscle gains over 12 weeks
Sitting in cold water right after lifting built less strength and muscle over 12 weeks than an easy warm-down did. The cold appears to switch off some of the muscle-building signals the workout has just turned on.
Across 12 weeks of twice-weekly strength training, the group doing active recovery gained more strength and muscle mass than the group doing 10 minutes of cold-water immersion afterwards. A parallel mechanistic study in 9 men found suppressed anabolic signaling and fewer satellite cells (NCAM-positive and Pax7-positive) after cold immersion than after active recovery. The training study was 21 physically active men, small for a training outcome, and it compared cold against active recovery, not against doing nothing. The finding applies to cold taken immediately after lifting; it says nothing about cold on a separate day or several hours later. Both arms were men only.
Who this may not transfer to:Both arms were run entirely in men. Women have different resting muscle temperature and subcutaneous fat distribution, which changes how deeply a ten-minute immersion cools the muscle, so the size of the blunting effect in women has not been measured.
Cold water eases how sore you feel but blunts the muscle and strength you just trained for, so if growth is the goal keep the plunge away from the few hours right after lifting. Use it on rest days, after cardio, or when being recovered for tomorrow matters more to you than long-term gains.
The study · 1
Roberts et al., post-exercise cold water immersion attenuates acute anabolic signalling and long-term adaptations in muscle to strength training · J Physiol 2015;593(18):4285-301
Loaded stretching between sets added at most about 0.7 mm of muscle thickness
Holding a loaded stretch between sets may add a small amount of muscle, but the studies are tiny and their results include the real possibility of no effect at all.
Over 8 weeks of twice-weekly calf raises with one leg adding a 20-second loaded inter-set stretch, the soleus showed a possible 0.7 mm thickness benefit (90% CI 0 to 1.6 mm), the lateral gastrocnemius was ambiguous (0.4 mm, -0.4 to 1.3) and the medial gastrocnemius equivocal (0 mm, -0.6 to 0.7). A separate 5-week within-participant flywheel-squat trial adding inter-set static stretching found quadriceps thickness up 7.3% without and 8.0% with stretching, a non-significant difference. Both were small contralateral-limb studies in untrained men, with confidence intervals that include zero on the primary measures, and a published bench-press study found no pectoral difference, so the direct evidence does not yet support the practice.
Who this may not transfer to:Both trials were run entirely in untrained young men. Women have greater baseline joint range of motion and different passive tissue stiffness, both of which plausibly change what a loaded inter-set stretch does, and neither has been measured.
The studies · 2
Van Every et al., loaded inter-set stretch may selectively enhance muscular adaptations of the plantar flexors · PLoS One 2022;17(9):e0273451
Nakamura et al., effects of adding inter-set static stretching to flywheel resistance training on flexibility, muscular strength and regional hypertrophy in young men · Int J Environ Res Public Health 2021;18(7):3770
Static stretching between sets weakened the following sets
Doing static stretching between your sets made the following sets weaker. The exact size could not be pinned down from the published figures, but the direction was clear.
In 12 resistance-trained men (mean age 29) doing isokinetic knee extension across three counterbalanced conditions, static stretching between sets reduced performance in the subsequent sets. The magnitude cannot be confirmed: the figures appear only in a chart, and the formula printed in the paper makes their stated direction self-contradictory, so we give the direction and not the size. Measured in men only, and fatigue resistance at a given relative load differs between the sexes, so the size of any penalty in women is an extrapolation.
Who this may not transfer to:Measured in twelve resistance-trained men. Fatigue resistance at a given relative load differs between the sexes, so the size of the volume penalty in women is an extrapolation, not a measurement.
Because a static stretch between sets cost strength on the next set, keep static stretching out of your rest periods and save it for after training, where it improves flexibility without costing you reps.
The studies · 2
Padilha et al., could inter-set stretching increase acute neuromuscular and metabolic responses during resistance exercise? · Eur J Transl Myol 2019;29(4):8579
Chaabene et al., acute effects of static stretching on muscle strength and power, an attempt to clarify previous caveats · Front Physiol 2019;10:1468
Blood Sugar
Resistance training lowers HbA1c about 0.4 points in type 2 diabetes
Lifting weights lowers long-term blood sugar in people with type 2 diabetes. Across 20 trials and 1,172 people, resistance training cut HbA1c by about 0.4 percentage points compared with people who did not train, close to what you would expect from adding a standard diabetes drug (on the HbA1c scale, 6.5% and above is the diabetes threshold). People who got noticeably stronger saw the biggest drops.
Working muscle draws glucose out of the blood, and repeated resistance training enlarges muscle and increases the amount and activity of glucose transporters, so muscle keeps clearing sugar more efficiently between sessions. Building strength also improves how muscle responds to insulin, which is why the trials that produced the largest strength gains also produced the largest HbA1c reductions. The pooled 0.39 percentage-point fall is clinically meaningful because reductions of this size track with lower risk of diabetes complications.
Who this may not transfer to:Trials enrolled both men and women with type 2 diabetes across a range of ages; sex did not emerge as a major moderator of the glycemic effect.
The study · 1
Jansson AK, et al. Resistance training and HbA1c in type 2 diabetes: systematic review and meta-analysis · BMJ Open Diabetes Res Care 2022;10(2):e002595
Mood & stress
Resistance training eases depression, about 1 in 4 meaningfully improved
Lifting weights eases depression. Pooling 33 trials and 1,877 people, resistance training produced a moderate improvement in depressive symptoms (an effect size of 0.66, where 0.5 is moderate and 0.8 is large). For roughly every 4 people who trained, 1 more felt meaningfully better than they would have without training. The benefit showed up whether or not someone started with a diagnosed illness.
The improvement held across ages and starting fitness levels and did not depend on people actually getting stronger, which points to effects beyond the muscle itself. Proposed pathways include exercise-driven changes in mood-regulating brain chemistry and inflammation, along with the sense of mastery and structure that a training routine provides. Because the effect size of 0.66 sits in the moderate range and the number needed to treat is just 4, resistance training is a substantive option alongside standard care.
Who this may not transfer to:Trials enrolled both men and women.
The study · 1
Gordon BR, et al. Resistance exercise training and depressive symptoms: meta-analysis and meta-regression of RCTs · JAMA Psychiatry 2018;75(6):566-576
Longevity And Mortality
Lifters have about 21% lower death rates, 40% with added cardio
People who lift weights have about a 21% lower chance of dying over the study period than people who do no exercise, and about 40% lower when they also do some cardio. Both figures are measured against doing nothing, so the 40% is not added on top of the 21%.
Pooling 11 studies and 370,256 people (ten prospective cohorts and one randomized trial), resistance training carried a hazard ratio of 0.79 for all-cause mortality against no exercise, and 0.60 when combined with aerobic activity, both against the same no-exercise comparison. Cardiovascular mortality did not reach significance and there was no cancer signal, so this holds for death from any cause, not for any single disease. Ten of the eleven studies are observational, so healthy-adherer bias applies: people who lift differ from people who do not in many ways at once, and the design cannot separate the training from the person doing it.
Who this may not transfer to:The review does not report the sex composition of its pooled cohorts.
The study · 1
Saeidifard et al., the association of resistance training with mortality, systematic review and meta-analysis · Eur J Prev Cardiol 2019;26(15):1647-1665
The fittest have about 80% lower death rates than the least fit
Among people given a treadmill test, the fittest had about 80% lower death rates than the least fit over the following years, and there was no point where being fitter stopped helping. Lifting does little for this kind of fitness, so cardio still belongs alongside it.
In 122,007 patients referred for exercise treadmill testing (41% women, mean age 53, median 8.4 years follow-up), the fittest group had a hazard ratio of 0.20 for mortality against the least fit, and 0.77 against the next-fittest group, with no upper limit of benefit. Fitness was measured directly on the treadmill, not self-reported, which is what gives this weight over most activity research. It is observational and the patients were clinically referred, not a general-population sample, so undetected early illness, which lowers measured fitness and independently raises mortality, explains part of the gradient.
Who this may not transfer to:At 41% women in a directly-measured-fitness cohort, this is unusually well balanced for this literature.
The study · 1
Mandsager et al., cardiorespiratory fitness and long-term mortality among adults undergoing exercise treadmill testing · JAMA Netw Open 2018;1(6):e183605
The longevity benefit flattens above about an hour a week
Any muscle-strengthening work lowered death and disease rates by about 10 to 17%, with most of the benefit reached at roughly 30 to 60 minutes a week. Above an hour a week the extra benefit for living longer flattened out. Diabetes was the exception, where the risk kept falling with more.
Across 16 prospective cohorts of adults 18 and over, muscle-strengthening activity carried 10 to 17% lower risk of all-cause mortality, cardiovascular disease, total cancer, diabetes and lung cancer. For all-cause mortality, cardiovascular disease and cancer the dose-response was J-shaped, with a maximum risk reduction of roughly 10 to 20% at about 30 to 60 minutes a week and the benefit flattening or bending back above that; diabetes fell steadily to 60 minutes a week. The J-shape rests on the subset of cohorts that reported enough dose detail, all using self-reported weekly minutes, and self-report is least reliable in the people reporting the most. The authors conclude the influence of higher volumes is unclear, not that it is harmful.
Who this may not transfer to:The review does not report a pooled sex composition across its 16 cohorts, so whether the J-shape looks the same in women is not answerable from this analysis.
For living longer, about an hour a week of strength work captures most of the benefit, so train beyond that for strength, muscle or blood-sugar control if those matter to you, not on the expectation of a longer life. The flattening above an hour is an absence of extra longevity benefit at higher volumes, not evidence that more training shortens life.
The study · 1
Momma et al., muscle-strengthening activities are associated with lower risk and mortality in major non-communicable diseases · Br J Sports Med 2022;56(13):755-763
Progress Markers
Every 11 lb (5 kg) of grip strength lost tracks with 16% higher death risk
Stronger people tend to live longer: across studies, a stronger grip tracked with roughly a third lower death rate, and every 11 lb (5 kg) of grip strength lost tracked with a 16% higher rate of dying. Grip predicted death better than blood pressure did. This shows that being strong goes with living longer, not that training your grip adds years.
Pooling 38 cohort studies (about 1.9 million participants, 63,087 deaths), higher handgrip strength carried a hazard ratio of 0.69 for all-cause mortality, stronger in women (0.60) than in men (0.69), and higher knee-extension strength carried a 14% lower risk. In the PURE cohort (139,691 people in 17 countries, median 4 years, 3,379 deaths), every 11 lb (5 kg) of grip lost carried a hazard ratio of 1.16 (95% CI 1.13 to 1.20), and grip predicted mortality better than systolic blood pressure. Reverse causation dominates: undiagnosed illness, low-grade inflammation, cancer and neurological disease all reduce grip for months to years before diagnosis and independently raise mortality, and neither dataset asked whether the strong people had trained.
Who this may not transfer to:Both sexes are represented and the pooled analysis reports them separately, which is unusual in this literature. The association was stronger in women, so the common practice of quoting a single pooled hazard ratio understates it for female readers.
The studies · 2
García-Hermoso et al., muscular strength as a predictor of all-cause mortality in an apparently healthy population · Arch Phys Med Rehabil 2018;99(10):2100-2113.e5
Leong et al., prognostic value of grip strength, findings from the PURE study · Lancet 2015;386(9990):266-73
Bone Density
Heavy lifting raised postmenopausal spine bone density 2.9%
Postmenopausal women with thinning bones did eight months of heavy, supervised lifting twice a week and gained bone density in the spine, while a light-exercise group lost it. There was one minor injury, a back spasm, across the whole trial.
In 101 postmenopausal women with osteopenia or osteoporosis (mean age 65), eight months of twice-weekly, 30-minute supervised training at five sets of five reps above 85% of a one-rep maximum raised lumbar-spine bone mineral density 2.9% against a 1.2% fall in the low-intensity control, raised femoral-neck density 0.3% against a 1.9% fall, and improved every functional measure, with one adverse event (a minor back spasm). Every session was supervised and compliance was 92%, so the safety record belongs to that setting, not to unsupervised heavy lifting, and the authors say so. The lead author licenses the commercial program derived from the trial. An erratum on the paper could not be read, and nothing suggests it touches the bone results.
Who this may not transfer to:A women-only trial, which is what makes it useful here: the heavy-loading protocol was tested in the population most often advised to avoid it. Whether older men with low bone mass respond the same way was not measured in this trial.
The study · 1
Watson et al., high-intensity resistance and impact training improves bone mineral density and physical function in postmenopausal women, the LIFTMOR randomized controlled trial · J Bone Miner Res 2018;33(2):211-220
How it works
Only about 39% of sports-science study participants are women
Across major sports-medicine journals, only about 39% of study participants were women, and the typical study ran 35 to 37% female. Much of the research this page rests on was done mostly in men.
Across 1,382 original articles in three major sports and exercise medicine journals over three years (6,076,580 participants), 39% of participants were female and 61% male, with the per-article share of women running 35% to 37% and the under-representation significant in all three journals. The counting mixes study types, so one large balanced cohort can offset dozens of all-male laboratory trials, which is why the per-article figure describes the typical study better than the pooled 39% does. The survey ends in the early 2010s and representation has improved since without closing. The second review cited here documents that combined resistance-training-plus-protein trials have rarely been evaluated by sex at all, which is why sex composition is flagged on every row of this page.
Who this may not transfer to:This row exists so the sex composition of every other row on this table can be read in context. The second citation documents the narrower gap that matters most here: combined resistance training plus protein has rarely been evaluated by sex at all.
The studies · 2
Costello, Bieuzen and Bleakley, where are all the female participants in sports and exercise medicine research? · Eur J Sport Sci 2014;14(8):847-51
Noh and Park, effects of resistance training and protein supplementation interventions on muscle volume and muscle function, sex differences in humans · Phys Act Nutr 2023;27(4):15-25
Young men and women build muscle protein at the same rate despite a 45-fold testosterone gap
After a workout with a protein drink, young men and young women built new muscle protein at nearly the same rate, even though men had a 45-fold bigger testosterone spike. The much larger hormone response did not turn into more muscle-building.
In 8 young men (mean age 20) and 8 young women (mean age 22), myofibrillar protein synthesis after resistance exercise with 25 g of whey rose 2.3-fold in men and 2.7-fold in women over the 1-to-5-hour window, with no meaningful difference at 24 to 28 hours. Men showed a 45-fold greater testosterone response and greater acute mTOR phosphorylation, with no corresponding difference in synthesis rate. It is 16 people, single-session, measuring acute synthesis, not muscle actually gained, and short-term synthesis rates have repeatedly failed to predict long-term growth. The women's menstrual-cycle phase was not controlled across the group.
Who this may not transfer to:Deliberately balanced, 8 per sex, which is why it can speak to this question at all. It is also the reason the finding is preliminary: a balanced study of 16 people is the direct evidence on a question the field mostly answered by assumption.
The study · 1
West et al., sex-based comparisons of myofibrillar protein synthesis after resistance exercise in the fed state · J Appl Physiol 2012;112(11):1805-13
Older women's resting muscle-building ran about 30% faster than younger women's
In a small study, older women's muscles ran their resting building machinery about 30% faster than younger women's, while older men's response to a nutrient-and-hormone boost dropped by roughly 60%. The pattern runs opposite to the common assumption that older women have the weaker muscle-building system.
In 8 young men, 10 young women (25 to 45), 10 older men and 10 older women (65 to 85), basal muscle protein synthesis was about 30% higher in older women than younger women (0.060 versus 0.046 %/h), while men's basal rate did not change with age. Men showed roughly a 60% smaller synthesis response to hormonal and nutrient stimulation when older; older women showed no significant increase to the clamp, from an already higher baseline. The trial excluded hormone-therapy and hormonal-contraceptive users, so the sex difference is not an artifact of who was taking what. It is a small mechanistic study measuring synthesis rates, not muscle gained.
Who this may not transfer to:Four sex-and-age groups of 8 to 10, which is what makes the comparison possible and what keeps it preliminary. The direction of the finding runs opposite to the popular assumption that older women have the weaker anabolic machinery.
The study · 1
Smith et al., sexually dimorphic effect of aging on skeletal muscle protein synthesis · Biol Sex Differ 2012;3(1):11
Much of the first weeks' size gain is swelling, not new muscle
Some of the 'muscle growth' you see in the first few weeks of training is swelling from muscle damage, not new tissue. By week 3 much of the size increase was fluid; by week 10 it was real growth and strength had climbed.
In 10 untrained young men over a 10-week program, vastus lateralis cross-sectional area rose about 2.7% by week 3 and about 10.4% by week 10. Ultrasound echo intensity, normalized to the change in area, was significantly raised only at week 3, alongside elevated myoglobin and interleukin-6, indicating that much of the week-3 increase was swelling from muscle damage, not new tissue, and maximal voluntary contraction rose only by week 10. It is ten men, one muscle, ultrasound, not biopsy or MRI, so it shows that early size measurements overstate growth in untrained people without quantifying by how much in an individual. Women show less exercise-induced muscle damage, so the early swelling artifact may be smaller in women, which has not been measured directly.
Who this may not transfer to:Ten untrained men. Women show less exercise-induced muscle damage and lower inflammatory markers after unaccustomed resistance exercise in several comparisons, so the size of the early swelling artefact may well be smaller in women, which has not been measured directly.
The study · 1
Damas et al., early resistance training-induced increases in muscle cross-sectional area are concomitant with edema-induced muscle swelling · Eur J Appl Physiol 2016;116(1):49-56
Heart And Vascular
Resistance training lowers blood pressure most in prehypertension, about 4/4 mmHg
Weeks of resistance training lowered resting blood pressure by a few points, about 1.8 on the top number and 3.2 on the bottom, and most in people whose pressure was borderline high, about 4 points on each. In people whose pressure was already high, the drop was smaller.
Across 93 trials (5,223 participants; 29 groups doing dynamic resistance training, 5 isometric), dynamic resistance training lowered resting blood pressure by about 1.8 mmHg systolic and 3.2 mmHg diastolic overall, with the largest reductions in prehypertensive people, about 4.0/3.8 mmHg, and an explicitly smaller effect in those already hypertensive. The review reports subgroup effects by baseline blood pressure, not by sex, so no sex-specific estimate can be read from it.
Who this may not transfer to:The review reports subgroup effects by baseline blood pressure, not by sex, so no sex-specific estimate can be read from it.
Because it does more to hold a borderline reading down than to bring an established high reading back, resistance training fits best as prevention and as something done alongside blood-pressure treatment, not in place of it.
The study · 1
Cornelissen and Smart, exercise training for blood pressure, a systematic review and meta-analysis · J Am Heart Assoc 2013;2(1):e004473
Anxiety And Stress
Resistance training gives a small-to-moderate drop in anxiety
Lifting weights lowers anxiety. Across 16 trials and 922 people, resistance training produced a small-to-moderate improvement in anxiety symptoms (an effect size of 0.31, where 0.2 is small and 0.5 is moderate). Healthy people saw the largest benefit. Training intensity, session length and how often people trained did not change the result much, so a modest, consistent routine is enough, and it can be done with bodyweight or resistance bands at home.
The calming effect appeared across a range of programs and did not hinge on high intensity or long sessions, suggesting that the act of training regularly matters more than pushing to a specific threshold. Candidate mechanisms include a dampened physical stress response, improved sleep, and the confidence that comes from handling a progressively harder physical challenge. The effect is smaller than for depression and drawn from fewer trials, so it is best read as a reliable, modest benefit, not a standalone treatment for an anxiety disorder.
Who this may not transfer to:Samples were mixed-sex and skewed female (about 68% women); sex did not significantly moderate the effect on anxiety.
The study · 1
Gordon BR, et al. Resistance exercise training and anxiety: meta-analysis and meta-regression of RCTs · Sports Med 2017;47(12):2521-2532
How It Works
Resistance training is a hormetic stress, the same family as heat and cold: a controlled load the body adapts to by building more than it had. Muscle cells respond to the tension inside their fibers by raising a growth signal, a process called mechanotransduction. They hold the new size only while that tension keeps exceeding what they have already adapted to.
Progressive overload (adding a little weight, a rep, or a set over time) is what turns exercise into training. Without it a program stops producing gains after a few months, while feeling exactly the same to do.
Anatomy of the Practice
1The first sessions
Strength climbs first. Most of the early rise is the nervous system learning the movement, more motor units recruited, better timing, less tension wasted in opposing muscles. Within two to four weeks an untrained person can lift noticeably more with little visible change in the muscle. Some of the early size is swelling from muscle damage, and new tissue comes later.
2The adaptation
Tension is the signal, and the muscle rebuilds thicker in response. The rebuilding happens mostly between sessions, in the recovery after the set. Raised protein synthesis runs for two to three days afterwards, so rest days are part of the training. A lighter set taken close to failure creates much the same tension as a heavy one, which is why load matters less for size than gym advice implies.
3Over weeks and months
Visible size arrives around two to four months in. Bone and tendon adapt to the same loading more slowly, over months. That is the timescale on which loading strengthens the skeleton.
How to Start
Ways to Do It
The dose that works is small, so aim for a level you can repeat. Technique comes before load: learn each movement at a weight you could do five more reps with before you add any. Two sessions a week reach almost all of the available benefit. Start at the first rung and climb as it suits you. If you are older, unwell, or new to it, begin gently and add load slowly.
A squat to a chair, a push-up at whatever angle is hard, a hip hinge, a row on a towel round a door handle, a calf raise. Two hard sets each, stopping a rep or two short of failure. Those trial gains came from barbell lifts in trained men; bodyweight applies the same low-dose principle at a lighter absolute load. Your legs get strong faster than bodyweight can keep challenging them, so bodyweight work stalls there first. The fix costs nothing: single-leg versions: split squats, step-ups, single-leg hinges.
Books or water bottles in a backpack add 22–44 lb (10–20 kg) to squats, step-ups and push-ups. Bands are the cheapest way to add the rows and pull-downs that bodyweight covers badly, though a band pulls harder the more you stretch it, unlike a fixed weight. Together they cover most of a program for very little, and they keep the progression going once bodyweight alone stops being hard.
Machines let you load the legs and back heavily without a spotter, the one thing a home setup does worst. It is also the easiest route to the loads the strength trials used. Often it is already included in a membership you have.
Equipping yourself to train at any load at home turns lifting into something you can keep up on your own schedule for years. There are two routes: cheap and hands-on, or costly and effortless.
A second-hand barbell, rack and plates. Squat, press, hinge and row at any load, in your own time. Used bars and plates cost a fraction of new and are functionally identical, since iron does not wear out. Cheaper, and a project to source and set up.
New adjustable dumbbells or a complete rack, delivered and ready. One purchase covers most of a program and progresses for years. Several times the cost of a second-hand build, and the easiest to sustain.
Go Deeper
- Protein and muscle: what to eat around training so the muscle you build stays on, and the daily protein target the trials keep pointing to.
- Balance and fall prevention: the other half of staying steady on your feet, and the skill lifting leaves untrained.
- Tai chi and qi gong: the balance practice with the most randomized evidence behind it, gentle as conditioning and a natural pairing with lifting.
- Hormesis: the biology of why a stress that would harm in excess strengthens the body in the right dose.
- Cold exposure: why the timing of a cold plunge matters on the days you lift.
Cautions For This Practice
Everything to be aware of is here, in one place. This practice suits most healthy people; a few situations call for real care.
Severe exercise muscle breakdown is rare, about 1 case per 100,000 a year
In a population-based medical-records cohort (Olmsted County, Minnesota, 2003 to 2015), the age- and sex-adjusted incidence of exertional rhabdomyolysis was 1.06 per 100,000 person-years (95% CI 0.59 to 1.52), and exertional cases were 4.9% of 431 rhabdomyolysis cases. The 20 exertional triggers were endurance activity (7), manual labor (5) and weight lifting (4); five patients had kidney injury, 16 were hospitalized a median of 2 days, there were no deaths and one recurrence. Only people who sought care and had creatine kinase measured appear, which selects for severity, so milder cases are invisible and the true rate of muscle damage after an unaccustomed session is far higher. 18 of 20 cases were male.Luetmer et al., exertional rhabdomyolysis, a retrospective population-based study
A held breath in heavy lifting spiked arterial pressure to 320/250 mmHg
Direct intra-arterial recording in 5 experienced bodybuilders lifting at 80, 90, 95 and 100% of maximum gave group mean peak pressures of 320/250 mmHg on the double-leg press, with one participant exceeding 480/350 mmHg. Pressures rose during the concentric phase and fell during the lowering phase, and lasted seconds. This is five experienced heavy lifters using a deliberate breath hold at maximal loads, the extreme end of the exposure, so it establishes the mechanism, not a measured rate of harm. The equivalent invasive measurement has not been made in women, and resting and exercising pressures differ by sex, so the absolute numbers should not be read across.MacDougall et al., arterial blood pressure response to heavy resistance exercise
A held breath raised eye pressure about 4.3 mmHg, twice as much as breathing out
In 30 men aged 18 to 40 (mean 26) without glaucoma, intraocular pressure during the fourth rep of a bench press rose an average of 4.3 mmHg with the breath held and 2.2 mmHg when the participant exhaled normally. Pressure rose in 90% of participants with the breath held and 62% without, and exceeded a 5 mmHg rise in 30% and 21% respectively. It measures single reps from a normal baseline in men without glaucoma. The authors raise glaucoma risk as a hypothesis, not a finding, and no study has followed lifters to see whether repeated transient rises damage the optic nerve. Baseline eye pressure differs with age and glaucoma is commoner in older adults than this sample.Vieira et al., intraocular pressure variation during weight lifting
Talk to a doctor before you start if this is you
The conditions that warrant a check first: known heart disease, uncontrolled high blood pressure, an aortic aneurysm, a bicuspid aortic valve, Marfan syndrome or another connective tissue disorder. Also a recent stroke, glaucoma or proliferative retinopathy, an unrepaired or recently repaired hernia, uncontrolled diabetes, or being pregnant and new to training. See a doctor too for chest pain, unexplained breathlessness, or fainting or near-fainting on exertion. None is an automatic no; most just change how you train.
Breathe through the lift
The sharp rises in blood pressure and eye pressure measured during heavy lifting come from holding your breath and straining under load. Breathe out through the hard part of each repetition and keep breathing across the set, and those spikes largely settle. A deliberate breath hold under a maximal load is a technique for competitive lifters under supervision, and none of the movements here need it.
Start with half of what you think you can do
The first month is where injuries happen, and doing too much on day one is the common mistake that ends programs. A very large volume of unfamiliar work, heavy on the lowering phase, is also what triggers the rare cases of severe muscle breakdown. Halve it for the first two weeks and build the movement before the load. Ordinary soreness that eases in a day or two is normal. Severe swelling, pain out of proportion to the session, a limb that will not straighten, or dark tea-colored urine are a same-day medical visit.
Some conditions change which exercises you pick
Blood thinners, osteoporosis and spinal conditions usually change the exercises you pick; they rarely rule training out. Loaded spinal flexion is the movement most often restricted in vertebral fragility. Where heavy training is used with osteoporosis, close supervision is what keeps it safe. Anyone with a known hernia or recent abdominal repair should follow the surgeon's timeline before loading.
On insulin or a sulfonylurea, watch for a low
If you take insulin or a sulfonylurea for type 2 diabetes, the added insulin sensitivity from regular training can push blood sugar low. Check your levels around your sessions in the first weeks, keep something sugary to hand, and ask whoever set the dose to lower it if the numbers run low. As training takes hold, the dose usually comes down, you keep training.
Start slow, be smart, read the research, and consult a professional if you have any concerns. This is here to inform your choice, not make it for you.
Common Questions
Do women need to train differently?
Nothing in the mixed-sex trials supports a different program for women. Women over 50 gained more relative lower-body strength than men, and the same relative muscle size. Direct measurement does not back the shorter post-training protein window often claimed for women.
Is soreness the sign a session worked?
Soreness is a poor scorecard. It tends to peak 24 to 72 hours after unfamiliar work and eases as you repeat the movement, so a session that leaves you sore did not necessarily build more. Judge progress by whether the load you can handle climbs from month to month.
How much protein do I need, and does cold water after lifting help?
Around 1.6 g of protein per kilogram of bodyweight a day is the level where added muscle stops rising in trials. Supplementing on top of a solid diet adds little, and the gain is largest when current intake is low. Sitting in cold water straight after lifting eased soreness but built less strength and muscle over 12 weeks than an easy warm-down did. The cold appears to switch off some of the muscle-building signals the session just turned on. If growth is the goal, keep a plunge clear of the hours right after lifting, and use it on rest days or after cardio.
Explore Related
Other pages this one connects to, by the evidence they share, the outcomes they touch, and the ground they cover.
Pages that lead here: Hormesis: How a Low Dose of Stress Strengthens and a High Dose Harms · Doing Hard Things: What the Research on Discipline Actually Found · Self-Measures: The Body Numbers You Can Track Yourself, and What Each One Predicts · Low Back Pain: why most improves, and what to try first
All 34 sources on this page independently checked and cross-referenced.
Thomas Dehli, Founder & Editor, Sacred Lotus
Sacred Lotus has published Chinese medicine reference material since 2001. Integrative pages are held to the same standard as the herb and formula library: cite the source, grade the claim at its real strength, and say where the research has not looked. This page is educational and it is not medical advice. Last reviewed and updated August 8, 2026.
Evidence strength
How confidently the research supports a claim. Strength describes the evidence, not our endorsement.