Strength Training and Longevity: How Many Times a Week Should You Train to Live Longer?

For most people this is good news: the minimum dose linked with the highest benefit in population analyses is small and achievable even on a tight schedule. You do not need to spend hours in the gym to fall within the range that was linked with lower risk of death in the data. You do need to train regularly, with a sensible choice of exercises and gradually increasing load.

In this article we focus on the question of dose: how many minutes and how many sessions a week, how to read the J curve and how to combine strength training with endurance training. We do not repeat topics covered separately, such as strength training for women over 40 or the effect of strength training on laboratory parameters. A broader discussion is in our entries on strength training, grip strength as a biomarker of ageing and VO2 max.

A second large meta-analysis, covering 10 studies, confirmed a similar picture. Any amount of resistance training compared with none was associated with a 15% lower risk of all-cause death (RR 0.85; 95% CI 0.77-0.93), a 19% lower risk of cardiovascular death (RR 0.81; 0.66-1.00) and a 14% lower risk of death from cancer (RR 0.86; 0.78-0.95). In the dose-response analysis (4 studies) the greatest reduction in all-cause mortality risk, 27%, was seen at about 60 minutes a week (RR 0.74; 0.64-0.86), and at larger volumes the benefit diminished.

Note two things. First, the relative values (10-27%) are real but moderate differences, not a guarantee: they mean that at population level the number of deaths is lower, not that everyone who trains will live a given number of years longer. Second, these are data from observational studies, in which people who do strength exercise also differ from the rest in other characteristics (aerobic activity, diet, smoking, socioeconomic status), so statistical adjustments will not remove all confounding.

A J shape means that risk first falls, reaches a minimum at a moderate amount of exercise, and then stops falling and in some analyses even rises slightly. It is easy to conclude that 'too much gym shortens life', but the data do not prove that. In a study of 28,879 women with a mean age of 62 years, followed for an average of 12 years (3,055 deaths), hazard ratios for successive categories of strength training time (0, 1-19, 20-59, 60-149 and at least 150 minutes a week) were 1.0 (reference), 0.73, 0.71, 0.81 and 1.10 (95% CI 0.77-1.56) respectively. The last value lies within a wide confidence interval that includes 1, so neither benefit nor harm can be concluded.

Why are the data less favourable at larger volumes? There are at least three possible explanations. First, the highest categories are often very small, so estimates are uncertain. Second, training time was declared in a questionnaire, which favours errors, and people who exercise very much may differ from the rest (for example more often having conditions that prompt intensive programmes, or training in a way that burdens joints and the circulation). Third, the effect may be real but have nothing to do with strength training itself, rather with insufficient rest or eating disorders in some people.

Since the data describe total weekly strength exercise time, the question naturally arises whether it is better to train twice for 30 minutes or four times for 15. For muscle gain the experimental data are well studied. A meta-analysis of 25 studies on training frequency found that, with equal training volume, the difference between lower and higher weekly frequency has no significant effect on muscle hypertrophy, and the authors conclude that frequency per muscle group can be chosen according to personal preference.

For longevity there are no such direct comparisons, but in a study of 80,306 adults from England and Scotland (5,763 deaths) participation in strength exercise was linked with lower all-cause mortality (HR 0.77; 95% CI 0.69-0.87), and meeting only the strength exercise guideline (at least 2 sessions a week) with an HR of 0.79 (0.66-0.94). Meeting only the aerobic guideline (equivalent to 150 minutes of moderate effort a week) was linked with an HR of 0.84, and meeting both guidelines with an HR of 0.71 (0.57-0.87). The authors conclude that it is worth promoting adherence to strength exercise guidelines over and above general physical activity targets.

Practical conclusion: two to three sessions a week is a reasonable pattern, consistent both with the guidelines and with the data on strength gains. If your time is limited, regularity is better than rare, very long workouts.

Besides data on strength exercise, a separate line of research concerns muscle strength itself as a predictor of survival. In a meta-analysis of 38 studies covering 1,907,580 participants and 63,087 deaths, higher grip strength was linked with lower risk of death from any cause (HR 0.69; 95% CI 0.64-0.74), somewhat more strongly in women (HR 0.60) than in men (HR 0.69), and higher knee extension strength with 14% lower risk (HR 0.86). The authors stress that strength tests are easy to perform and can help identify people at increased risk.

That last caveat is key. Grip strength is a strong marker of overall health and reserves, but that it is a good predictor does not prove that raising it will extend life. Muscle is also an organ that shrinks with age. A meta-analysis of 49 studies (1,328 participants aged at least 50) showed that resistance training increased lean body mass by an average of 1.1 kg (95% CI 0.9-1.2), with larger gains in higher-volume programmes, and in older people gains were somewhat smaller. The result encourages starting strength training as early as possible, but also shows that at any age there is something to improve.

Epidemiological data do not say how to train, only how much time to devote. For building a plan, principles from training research help: multi-joint exercises (squats or their variants, deadlift or hip hinge, pressing, rowing, pull-ups, lunges), gradual increases in load or repetitions, training each main muscle group at least once a week, and performing sets close to muscular fatigue with correct technique.

An example minimum scheme for a beginner: two sessions a week of 30 minutes, each covering the whole body. In a session, 6-8 exercises, 1-3 sets of 6-12 repetitions, with 1-2 minute rests. For someone with more time, a third session or a longer workout would extend total time to the upper end of the range (60 minutes a week) without entering the area where data are uncertain. This is only an example, not a prescription; the scope and type of exercise should be matched to age, health and starting fitness.

Remember endurance exercise. In cohort studies the combination of strength and aerobic exercise was linked with lower mortality than neither, and in the English-Scottish analysis meeting both guidelines gave the lowest risk (HR 0.71). Strength training does not replace aerobic fitness, which is a separate, strong predictor of survival, as we write in our entry on VO2 max.

The myth of 'more is better' is especially tempting in gym culture. The data do not say that large volumes are harmful, only that there is no proof of additional benefit in terms of survival. Large volumes can make sense for aesthetic, sporting or hypertrophy goals, but those are different goals from minimising the risk of death. Distinguishing goals helps avoid disappointment.

If you had to choose one thing to do about strength, it would be consistency: two or three short, well-planned workouts a week, with progression and without record volumes. That is the range linked with the greatest benefit in population analyses, it is realistic for busy people and needs no expensive equipment. Add endurance exercise and recovery, and you have a plan in line with the best available data.