Strength Training

What it is

Strength training, also called resistance training or weight training, means working muscles against a load such as machine weights, bars or dumbbells [1]. Programmes differ in load, number of sets, repetitions and how often people train, and these details are what the trials below compare [2].

What the research says

Muscle and strength have the strongest evidence. A network meta-analysis of 178 randomised trials with 5,097 adults found that every strength programme gained more strength than no exercise, and heavier loads gave the largest gains [2]. In a review of 22 trials in adults aged 75 and over, strength also improved, and few adverse events were reported [3].

For blood sugar, a 2025 meta-analysis of 43 trials with 2,012 adults aged 50 and over who had type 2 diabetes found that resistance training lowered HbA1c, a marker of average blood sugar, by 0.55 percentage points [4]. A 2025 review of 19 trials in older adults with type 2 diabetes found a similar drop in HbA1c of 0.51 and lower fasting glucose [5]. In a review of 24 trials, higher-intensity sessions lowered HbA1c more than low-to-moderate sessions (-0.61 vs -0.23) [6].

For mood, 33 randomised trials with 1,877 people found that resistance training reduced depressive symptoms by a moderate amount (effect size 0.66), and the number needed to treat was 4 [7]. In 21 studies of 1,610 adults aged 60 and over, depression scores improved (SMD -1.13) [8]. In a trial of 55 young adults, eight weeks of twice-weekly sessions produced large drops in depressive symptoms compared with a wait-list [9].

For chronic low back pain, 10 trials with 434 adults found that resistance training reduced pain intensity, with a large effect (SMD -1.15), and also reduced disability [10]. The reviewers rated the evidence as moderate certainty and noted that only the pain result reached a clinically meaningful level [10].

For blood pressure, 14 trials with 253 adults with hypertension found that strength training lowered both systolic and diastolic pressure [11]. Across 24 trials in older adults with prehypertension or hypertension, systolic pressure fell by about 7 mmHg on average [12]. In healthy adults the drop was smaller: about 2 mmHg systolic and 3 mmHg diastolic for dynamic resistance training [13]. In a network meta-analysis of 270 trials with 15,827 people, isometric training (holding a contraction, such as a handgrip squeeze or wall sit) lowered systolic pressure by 8.2 mmHg and diastolic by 4.0 mmHg, the largest drops of five exercise modes [14].

Observational studies link resistance training to a lower risk of early death. A meta-analysis found 15% lower all-cause mortality (relative risk 0.85) for any resistance training compared with none, with the largest reduction at about 60 minutes a week [15]. A review of 16 cohort studies found 10-17% lower risk of all-cause mortality, cardiovascular disease, total cancer, diabetes and lung cancer with muscle-strengthening activity [16]. In a US cohort of 479,856 adults, muscle-strengthening alone had a hazard ratio of 0.89 for all-cause death compared with people who did not meet the guidelines, and combined with aerobic activity it had a hazard ratio of 0.60 [17].

For bones, a review of 14 trials in older adults found that resistance training raised hip and femur bone density by 2.77% on average, but did not significantly change lumbar spine density [18]. A meta-analysis of 17 trials in postmenopausal women found gains at the lumbar spine and femoral neck, but heterogeneity between trials was high [19].

For fat, a meta-analysis of 114 trials with 4,184 adults with overweight or obesity found that resistance training alone was associated with about 1.0 kg less fat mass and 1.6% less body fat than no training, and about 0.8 kg more lean mass [20]. A separate review of 19 trials in older adults with type 2 diabetes found no change in body weight or fat mass [5].

For falls, a Cochrane review of 108 trials with 23,407 older people found that exercise overall reduced the rate of falls by 23% [21]. Programmes that combined balance and functional exercise with resistance exercise probably reduced falls by 34%, while resistance training alone had uncertain effects on falls [21].

Doses used in studies

The highest-ranked strength programmes used loads above 80% of one-repetition maximum (1RM), several sets and three sessions a week [2]. In 15 trials of young adults, training to muscle failure gave no significant difference in strength or muscle size gains, though in trained people it gave slightly more muscle growth [22]. For blood pressure, the largest drops came from programmes above 60% of 1RM, at least twice a week, for at least 8 weeks [11]. In older adults with prehypertension or hypertension, moderate loads of 60-80% of 1RM lowered systolic pressure by about 7 mmHg [12]. For bone density, subgroup analyses favoured loads of at least 70% of 1RM, three sessions a week and programmes of 48 weeks or longer, but not for every site [19]. Low-load training with blood-flow restriction produced smaller strength gains than heavy-load training [23]. In the depression trial, participants trained twice a week for eight weeks [9]. In a meta-analysis of trials, total training volume did not change the depression benefit [7]. In the mortality meta-analysis, the largest risk reduction was at about 60 minutes a week, and benefit levelled off at higher volumes [15].

Safety and interactions

Injuries are the main documented risk. In a review of 28 studies, injury rates ranged from 0.21 to 18.9 per 1,000 training hours, and most studies looked at CrossFit-style training. Traditional strength training was rated the safest method and strongman the least safe, but the review says the studies are of limited quality [24]. US emergency department visits for weightlifting-related shoulder injuries rose from national estimates of 8,073 in 2000 to 14,612 in 2017, and 65% were sprains, strains or muscle tears [25]. In exercise trials in older adults, adverse events were mostly musculoskeletal and non-serious, but one trial reported a pelvic stress fracture [21]. In a review of resistance and aerobic training for type 2 diabetes, adverse events did not differ significantly between the two (relative risk 1.17, 95% CI 0.77 to 1.79) [26].

Blood pressure rises sharply during sets. In 20 healthy young adults doing leg-extension protocols, systolic pressure rose by 57 to 87 mmHg during the session, depending on rest and repetitions [27]. In 15 older adults with hypertension, systolic pressure rose during each set, and moderate-speed sets produced more readings of 150 mmHg or higher than high-speed sets [28]. Check with a doctor first if you have heart disease or uncontrolled high blood pressure.

If you are pregnant, check with a doctor first. A review of 9 trials with 1,581 pregnant women reported benefits for quality of life and low back pain, but the authors call the findings promising and say the field needs standardised methods and detailed reporting [29].

Upper limits: no upper limit or toxic dose was found in these studies, and the dose data suggest diminishing returns at higher volumes [15]. Medicine interactions: no data were found on interactions with medicines. Breastfeeding, and people with kidney, liver or bleeding disorders: no data were found in this search.

Caveats and open questions

Most of the mortality and injury evidence comes from observational studies, which show links rather than cause. A 2020 US cohort found lower mortality below 2 hours a week of muscle-strengthening activity and no further reduction at 2 or more hours [30].

The bone trials measured bone density, not fractures, and lumbar spine results were inconsistent [18]. The mood reviews had high variation between trials, and trials with blinded allocation or assessment showed smaller benefits [7]. Blood pressure trials were small. In a 2013 review of 93 exercise trials, blood pressure falls after endurance training were larger in people with hypertension than in people with normal blood pressure [13]. The back pain evidence rests on 10 trials with 434 people [10].

Fat loss results conflict: one large meta-analysis found modest fat loss, while a smaller review in older adults with type 2 diabetes found no change in body weight or fat mass [20] [5].

Funding: one author of the strength network meta-analysis reported grants from dairy and nutrition companies, including the US National Dairy Council, during the conduct of the study, and personal fees from a nutrition company [2]. Several authors of the falls review also ran trials included in it [21].

Not covered here: children, people with heart failure or kidney disease, and long-term effects beyond the trial periods.