Exercise and Longevity: Which Exercises Add the Most Years to Your Life?

The 60-second version

Physical inactivity kills approximately 5.3 million people per year globally, making it one of the leading modifiable risk factors for premature death. The relationship between exercise and longevity is not a vague correlation. It is one of the most robustly documented cause-and-effect relationships in all of medicine, supported by prospective cohort studies covering tens of millions of person-years of follow-up.

But here is what most people get wrong: not all exercise is equal. The type of exercise you choose, its intensity, its duration, and even whether you do it alone or with others can shift your life expectancy by nearly a decade in either direction. A solo gym session and a tennis match are both "exercise," yet the mortality data separating them is enormous.

This article walks through every major line of evidence connecting exercise to lifespan. Every claim is anchored to a specific study with a specific sample size, published in a peer-reviewed journal. By the end, you will know exactly which exercises add the most years, how much is enough, whether too much is dangerous, and how the Death Clock calculator converts your exercise habits into a personalised life expectancy estimate.

The Copenhagen City Heart Study: sport type and years gained

The single most influential study on exercise type and longevity was published in 2018 in the Mayo Clinic Proceedings by Schnohr et al. It drew on data from the Copenhagen City Heart Study, a prospective cardiovascular cohort that began enrolling participants in 1976. A total of 8,577 individuals were followed for up to 25 years, during which time 4,448 of them died. The researchers compared life expectancy gains associated with different sport and exercise types, adjusting for age, sex, smoking, education, income, diabetes, alcohol intake, and other confounders.

The results were striking. Compared to sedentary individuals, the estimated years of additional life expectancy by sport were as follows:

Why social sports dominate the leaderboard

The most important pattern in this data is not about heart rate zones or caloric expenditure. It is about social interaction. The three sports with the largest life expectancy gains (tennis, badminton, and soccer) are all inherently social. They require at least one other person. Gym activities and jogging, which are frequently performed alone, sit at the bottom of the table despite being perfectly valid forms of cardiovascular exercise.

This finding aligns with a large body of evidence linking social connectedness to mortality. A 2010 meta-analysis by Holt-Lunstad et al. in PLoS Medicine, pooling 148 studies and 308,849 participants, found that strong social relationships increase survival odds by 50% (odds ratio 1.50, 95% CI 1.42 to 1.59). The effect size was comparable to quitting smoking and larger than the mortality effects of obesity or physical inactivity alone. Social sports may therefore deliver a double benefit: the physiological adaptations of exercise combined with the survival advantage of regular, structured social interaction.

The Copenhagen researchers themselves noted that while they could not fully disentangle the social component from the exercise component, the magnitude of the gap between social and solo activities was too large to be explained by differences in exercise intensity or duration alone. Tennis players and gym-goers in the study had similar weekly exercise volumes, yet the life expectancy gap between them was 8.2 years.

What about the confidence intervals?

It is worth noting that the confidence intervals in this study were wide, particularly for tennis (+4.1 to +15.4 years). This reflects the relatively modest sample size of 8,577 participants and the smaller number of individuals within each sport subgroup. The point estimates should be interpreted as the best available evidence of the relative ordering of activities, not as precise predictions for any individual. That said, the pattern is consistent: social, racquet-based sports sit at the top; solo, repetitive activities sit at the bottom.

Subsequent analyses have broadly supported this hierarchy. A 2016 study by Oja et al. in the British Journal of Sports Medicine, using six large population surveys from England and Scotland covering 80,306 adults, found that racquet sports were associated with the largest reduction in all-cause mortality (47% risk reduction, HR 0.53, 95% CI 0.40 to 0.71) and cardiovascular mortality (56% reduction, HR 0.44, 95% CI 0.24 to 0.83). Swimming was associated with a 28% all-cause mortality reduction, aerobics with 27%, and cycling with 15%. The hierarchy mirrored the Copenhagen findings almost exactly.

Practical implications

The Copenhagen data does not mean you should quit running and take up tennis. Any exercise is vastly better than no exercise. A sedentary person who starts jogging will gain approximately 3.2 years of life expectancy, which is an enormous return on investment. But if you are already exercising and want to optimise your mortality risk, the evidence suggests you should strongly consider adding a social sporting activity to your routine. Join a tennis league, play badminton with friends, sign up for a recreational football team. The social interaction component appears to amplify the longevity benefits of physical activity in ways that a solo gym session cannot replicate.

It is also worth considering that the mechanism behind the social sport advantage may extend beyond simple companionship. Team sports and racquet sports involve cognitive engagement (anticipating an opponent's movements, making rapid tactical decisions), emotional regulation (managing competitive stress), and variable-intensity movement (intermittent bursts of maximal effort followed by recovery). These features resemble high-intensity interval training embedded within a social context, which may explain why the mortality benefits are so pronounced.

The dose-response curve: how much exercise is enough?

The question of "how much exercise do I need?" has been answered with remarkable precision by one of the largest studies ever conducted on the topic. In 2015, Arem et al. published a pooled analysis of six prospective cohort studies in JAMA Internal Medicine, encompassing 661,137 men and women with a median follow-up of 14.2 years. During follow-up, 116,686 participants died, giving the study enormous statistical power to detect dose-response relationships.

The researchers categorised participants by their weekly leisure-time physical activity, measured in metabolic equivalent of task hours per week (MET-hours/week). The physical activity guidelines recommend 7.5 to 15 MET-hours per week, which is equivalent to roughly 150 to 300 minutes of moderate-intensity exercise (such as brisk walking) or 75 to 150 minutes of vigorous exercise (such as running).

The mortality reduction curve

Several features of this curve are remarkable. First, the biggest jump in mortality reduction happens between zero exercise and sub-guideline exercise. Simply doing something, even if it falls well short of the official 150-minute recommendation, reduces your mortality risk by 20%. This is the single most important finding in exercise epidemiology: the transition from inactive to slightly active is the largest risk reduction step you can take.

Second, the curve exhibits diminishing returns. Moving from zero to 7.5 MET-hours per week buys you a 20% mortality reduction. Doubling that to 15 MET-hours adds another 11 percentage points. Tripling it to 22.5 adds only 6 more. Beyond about 22.5 MET-hours per week (roughly 450 minutes of moderate exercise, or about an hour per day), the curve essentially flattens. You can still exercise more, but the additional mortality benefit is negligible.

Third, and this is crucial, there was no increase in mortality at any level of physical activity studied, including those exercising at 10 times the recommended amount. The highest-volume exercisers had a 31% mortality reduction, which was slightly lower than the peak of 39% but not statistically significantly different, and certainly not elevated above baseline. The feared "U-curve" of exercise and mortality (where extreme exercisers die sooner) did not materialise in this enormous dataset.

Vigorous vs moderate: intensity matters

Within the Arem data, vigorous-intensity activity showed slightly greater mortality reduction per unit of time compared to moderate-intensity activity. This is consistent with findings from the 2012 Copenhagen City Heart Study analysis by Schnohr et al. in the American Journal of Epidemiology, which followed 5,106 healthy joggers and 1,878 non-joggers for up to 35 years. The lowest mortality was observed among joggers who ran at a slow to moderate pace (HR 0.51), two to three times per week (HR 0.68), for 1 to 2.4 hours per week total (HR 0.71). Both non-joggers and high-intensity, high-frequency joggers had higher mortality rates than this moderate group, though the high-intensity finding was based on a very small number of strenuous joggers (n=36) and should be interpreted cautiously.

Converting exercise to years of life

A 2012 analysis by Moore et al. in PLoS Medicine, using pooled data from 650,386 adults with a median follow-up of 10 years, estimated the years of life gained by physical activity level. Compared to inactive individuals, those meeting the minimum guidelines (7.5 MET-hours/week) gained 3.4 years of life expectancy. Those at the highest activity level (22.5+ MET-hours/week) gained 4.5 years. Among individuals with a BMI above 35 who were also inactive, the loss of life expectancy was 7.2 years compared to active, normal-weight individuals, demonstrating that physical inactivity and obesity are additive risk factors.

The J-curve debate: can too much exercise kill you?

Few topics in exercise science generate more controversy than the question of whether extreme exercise is harmful. Media headlines regularly feature stories of marathon runners dying mid-race, ultra-endurance athletes developing heart arrhythmias, and retired professional athletes with cardiac scarring. The narrative is appealing: if some exercise is good, too much must be bad. But what does the data actually show?

Sudden cardiac death during exercise

Sudden cardiac death during or immediately after vigorous exercise does occur, but it is exceedingly rare. A 2012 study by Kim et al. in the New England Journal of Medicine analysed 10.9 million marathon and half-marathon participants in the United States from 2000 to 2010. They identified 59 cardiac arrests, of which 42 were fatal. This translates to an incidence of 0.54 cardiac arrests per 100,000 participants and 0.39 deaths per 100,000 participants. Among male marathoners specifically, the rate was 1.01 per 100,000. The most common cause was hypertrophic cardiomyopathy in younger runners and coronary artery disease in runners over 40.

To put this in context, the background rate of sudden cardiac death in the general population is approximately 1 to 2 per 1,000 per year. The risk during a marathon, while real, is a tiny fraction of the daily background risk that sedentary individuals face. As multiple researchers have pointed out, the overwhelming net effect of regular vigorous exercise is cardioprotective, even accounting for the small transient risk spike during exertion.

Ultra-endurance and cardiac remodelling

A more nuanced concern relates to chronic structural changes in the hearts of extreme endurance athletes. A 2012 review by O'Keefe et al. in the Mayo Clinic Proceedings proposed that prolonged intense exercise (such as marathon running, ultra-marathons, and Ironman-distance triathlon) could cause chronic structural cardiac remodelling, including right ventricular dilation, myocardial fibrosis, and atrial fibrillation. The paper generated significant media attention and introduced the concept of a "U-shaped" curve for exercise and cardiac risk.

However, subsequent large-scale population studies have not supported the U-curve hypothesis. The Arem et al. 2015 study described above, with 661,137 participants, found no excess mortality at any exercise volume, including the very highest. A 2015 analysis by Lee et al. in the Journal of the American College of Cardiology, following 55,137 adults for 15 years, found that runners had a 30% lower all-cause mortality and 45% lower cardiovascular mortality than non-runners, with no evidence of harm at higher running volumes or faster paces. And a 2018 study by Merghani et al. in Circulation, which performed cardiac MRI on 152 masters-level endurance athletes (average age 54, average 31 years of competitive exercise), found that while 17% had myocardial fibrosis, their overall mortality remained far below population averages.

Atrial fibrillation: the one genuine concern

The one area where extreme endurance exercise does appear to carry a measurable risk is atrial fibrillation (AF). A 2019 meta-analysis by Elliott et al. in the British Journal of Sports Medicine found that endurance athletes had approximately 2.5 times the risk of AF compared to non-athletes (pooled OR 2.46, 95% CI 1.73 to 3.51, across 13 studies and 70,478 participants). The mechanism is thought to involve atrial dilation and fibrosis from years of sustained high cardiac output.

However, AF alone does not necessarily translate to increased mortality. Many athletes with AF live normal or above-normal lifespans, particularly when the condition is identified and managed. The net cardiovascular mortality among endurance athletes remains significantly below that of sedentary populations. The presence of AF is a genuine medical consideration for extreme exercisers but does not support the broader claim that too much exercise shortens life.

The bottom line on extreme exercise

There is currently no population-level evidence that any volume of exercise increases all-cause mortality in healthy individuals. The transient risks during acute exercise bouts are real but tiny compared to the background risk of cardiovascular events in sedentary people. Structural cardiac changes (fibrosis, AF) do occur in a subset of extreme endurance athletes, and these individuals should receive appropriate cardiac screening. But the overwhelming balance of evidence indicates that the mortality benefits of exercise increase with dose up to a plateau, beyond which they level off rather than reverse. The J-curve remains a hypothesis without adequate empirical support from large population studies.

Strength training vs cardio: separate pathways to survival

For decades, public health messaging focused almost exclusively on aerobic exercise: walk more, run more, cycle more, swim more. Resistance training (strength training, weight lifting) was seen as relevant to athletes and bodybuilders but largely irrelevant to longevity. That view has been comprehensively overturned by epidemiological evidence from the past decade.

The Stamatakis data: strength training and mortality

In 2018, Stamatakis et al. published a landmark study in the American Journal of Epidemiology using pooled data from the Health Survey for England and the Scottish Health Survey. The study included 80,306 adults aged 30 and over, with a mean follow-up of 9.2 years, during which 14,013 deaths occurred. Participants who reported performing strength-promoting exercise had significant reductions in mortality compared to those who did not:

• All-cause mortality: 23% reduction (HR 0.77, 95% CI 0.68 to 0.88)
• Cancer mortality: 31% reduction (HR 0.69, 95% CI 0.53 to 0.90)

These reductions were independent of aerobic physical activity. In other words, strength training provided mortality benefits above and beyond whatever cardio the participants were doing. When the researchers looked at participants who met both the aerobic guidelines and the strength training guidelines, the combined mortality reduction was greater than either alone.

How much strength training is enough?

A 2022 systematic review and meta-analysis by Momma et al. in the British Journal of Sports Medicine pooled 16 studies covering 479,856 participants and found a non-linear dose-response relationship between strength training volume and mortality. The maximum mortality reduction (10 to 17% for all-cause mortality, 14% for CVD mortality, and 9% for cancer mortality) was observed at approximately 30 to 60 minutes per week. Beyond 130 to 140 minutes of weekly strength training, there was no additional mortality benefit, and some analyses suggested a slight attenuation of benefit, though no increase in risk above baseline.

This means that two to three 20-minute sessions per week, or two 30-minute sessions, are sufficient to capture the bulk of strength training's mortality benefits. You do not need to spend hours in the gym to reduce your death risk from resistance exercise. The minimal effective dose is remarkably small.

The combination effect

A 2022 study by Zhao et al. in the British Journal of Sports Medicine, using data from the National Health Interview Survey and the National Death Index covering 416,420 US adults, found that the combination of aerobic exercise and strength training was associated with a 40% lower all-cause mortality risk compared to no exercise (HR 0.60, 95% CI 0.54 to 0.67). The combination outperformed aerobic exercise alone (32% reduction) and strength training alone (14% reduction). For cardiovascular mortality, the combined effect was even more pronounced: a 46% reduction.

Why strength training reduces cancer mortality

The 31% cancer mortality reduction from the Stamatakis study deserves special attention. The proposed mechanisms include improved immune surveillance (strength training increases natural killer cell activity), reduced chronic inflammation (lower IL-6 and TNF-alpha), improved insulin regulation (reducing the cancer-promoting effects of hyperinsulinaemia), and favourable changes in sex hormone binding globulin (SHBG) levels that may reduce exposure to estrogen and testosterone in hormone-sensitive cancers. A 2019 study by Cormie et al. in Epidemiologic Reviews reviewed the evidence and concluded that there is strong biological plausibility for a causal relationship between resistance exercise and reduced cancer progression.

Muscle mass and mortality: grip strength as a crystal ball

If you could choose only one physical measurement to predict how long someone will live, you might be surprised by the answer. It is not blood pressure, not cholesterol, not BMI, not resting heart rate. Increasingly, the evidence points to grip strength as one of the single strongest predictors of all-cause mortality, and by extension, to muscle mass and muscular function as critical determinants of longevity.

The PURE study: 140,000 participants across 17 countries

The Prospective Urban Rural Epidemiology (PURE) study, led by Leong et al. and published in The Lancet in 2015, measured handgrip strength in 139,691 adults aged 35 to 70 from 17 countries across five continents. Over a median follow-up of 4 years, they recorded 3,379 deaths, 2,292 cardiovascular events, 1,117 strokes, and 1,290 myocardial infarctions.

The findings were definitive. Each 5 kg reduction in grip strength was associated with:

• All-cause mortality: 16% increase (HR 1.16, 95% CI 1.13 to 1.20)
• Cardiovascular mortality: 17% increase (HR 1.17, 95% CI 1.11 to 1.24)
• Non-cardiovascular mortality: 17% increase (HR 1.17, 95% CI 1.12 to 1.21)
• Myocardial infarction: 7% increase (HR 1.07, 95% CI 1.02 to 1.11)
• Stroke: 9% increase (HR 1.09, 95% CI 1.05 to 1.15)

Critically, grip strength was a stronger predictor of cardiovascular death than systolic blood pressure (HR 1.09 per 10 mmHg increase). This result startled the cardiology community. A simple, inexpensive, 30-second hand dynamometer test outperformed one of the most established cardiovascular risk factors as a mortality predictor.

Why grip strength predicts death

Grip strength is not important in itself. Dying of a weak grip is not a recognised cause of death. Rather, grip strength serves as a biomarker for overall muscular health, nutritional status, neural function, and physiological reserve. Low grip strength reflects sarcopenia (age-related muscle loss), chronic inflammation, malnutrition, hormonal decline, and physical frailty. It captures systemic health in a single number, which is why it predicts not just cardiovascular death but death from all causes including cancer, respiratory disease, and infection.

A 2018 study by Celis-Morales et al. in the BMJ, using UK Biobank data from 502,293 participants followed for a median of 7 years, confirmed and extended the PURE findings. They found that grip strength was inversely associated with all-cause mortality (HR 0.69 per standard deviation increase, 95% CI 0.67 to 0.72 in men), cardiovascular mortality (HR 0.60), cancer mortality (HR 0.76), and respiratory mortality (HR 0.59). The associations remained significant after adjustment for physical activity, diet, smoking, and other confounders, suggesting that muscular strength reflects something beyond exercise habits alone.

Sarcopenia: the silent epidemic

Sarcopenia, the progressive loss of skeletal muscle mass and function that accelerates after age 50, is increasingly recognised as a major cause of disability, hospitalisation, and death in older adults. A 2017 meta-analysis by Beaudart et al. in the Journal of Cachexia, Sarcopenia and Muscle, pooling 17 studies and 35,287 community-dwelling older adults, found that sarcopenia was associated with a 2-fold increase in all-cause mortality (pooled HR 2.00, 95% CI 1.71 to 2.34) over follow-up periods ranging from 2 to 14 years.

The practical message is clear: preserving and building muscle mass through resistance training is not a cosmetic pursuit. It is a survival strategy. Every kilogram of lean mass you maintain past age 50 is a buffer against frailty, falls, infection, and the metabolic cascades that kill older adults. Grip strength is a convenient proxy for this underlying muscular capital, and it should be measured at every medical check-up just as blood pressure is.

VO2 max: the single strongest predictor of death

If grip strength is the best simple test of muscular health, VO2 max (maximal oxygen uptake, measured during a graded exercise test on a treadmill or cycle ergometer) is the best single measure of cardiorespiratory fitness. And in 2018, a study emerged from the Cleveland Clinic that elevated VO2 max from an exercise physiology metric to arguably the most powerful mortality predictor in medicine.

The Cleveland Clinic study: Mandsager et al. 2018

Mandsager et al. published their findings in JAMA Network Open. They retrospectively analysed 122,007 patients who underwent symptom-limited treadmill exercise testing at the Cleveland Clinic between 1991 and 2014. After a median follow-up of 8.4 years, 13,637 deaths occurred. Patients were stratified into five cardiorespiratory fitness categories based on their age- and sex-adjusted peak metabolic equivalents (METs, a proxy for VO2 max): low, below average, above average, high, and elite.

The results were extraordinary:

The most striking finding was the comparison between low fitness and extreme fitness. Being in the lowest fitness quintile carried a mortality risk comparable to coronary artery disease, diabetes, and smoking. The difference between low fitness and elite fitness was associated with a 5-fold mortality risk differential (HR 0.20 for elite vs low). No other single variable in the study, including age, smoking, hypertension, diabetes, or coronary disease, carried as large a relative risk differential.

Furthermore, and this is the point that generated the most discussion, there was no plateau or upper limit to the fitness-mortality benefit. Elite fitness (the top 2.4%) was associated with lower mortality than high fitness (the 75th to 97.6th percentile), which in turn was lower than above average fitness. The benefit continued to increase linearly all the way to the most extreme levels of cardiorespiratory performance, with no evidence of diminishing returns or harm.

What determines VO2 max?

VO2 max is determined by a combination of genetics (approximately 50% heritable according to the HERITAGE Family Study by Bouchard et al., 1999), training status, age, and sex. It declines by approximately 10% per decade after age 30 in sedentary individuals, but regular endurance training can substantially slow this decline. A 2008 study by Trappe et al. in the Journal of Applied Physiology found that lifelong exercisers in their seventies had VO2 max values comparable to sedentary individuals 30 years younger.

The practical implication is that VO2 max is modifiable. While you cannot change your genetic ceiling, most people are operating far below their genetic potential. Structured endurance training, particularly high-intensity interval training (HIIT), has been shown to improve VO2 max by 15 to 30% in previously sedentary individuals over 8 to 12 weeks. This magnitude of improvement, based on the Mandsager data, would be expected to shift a person from low fitness to above-average fitness and cut their mortality risk roughly in half.

Should VO2 max be a vital sign?

Multiple authors have argued, on the basis of the Mandsager data and similar findings, that cardiorespiratory fitness should be measured routinely in clinical practice and treated as a vital sign. A 2016 scientific statement from the American Heart Association, published in Circulation by Ross et al., formally recommended that cardiorespiratory fitness be assessed clinically and reported alongside traditional risk factors. The statement noted that low cardiorespiratory fitness is a stronger predictor of mortality than established risk factors and that improving fitness reduces risk independently of changes in weight, blood pressure, or lipids.

Despite this recommendation, VO2 max testing remains uncommon in routine clinical practice, largely due to the need for specialised equipment and trained personnel. Submaximal estimates from step tests, 6-minute walk tests, or wearable devices that estimate VO2 max from heart rate data are increasingly available and may eventually make population-level fitness assessment feasible.

Walking: the minimum effective dose

For people who dislike structured exercise, who cannot afford a gym, or who have physical limitations that prevent vigorous activity, the single most important question in exercise epidemiology is: how much benefit can you get from simply walking? The answer, based on recent large-scale studies, is: far more than most people realise.

Steps and mortality: the Saint-Maurice data

In 2020, Saint-Maurice et al. published a prospective study in JAMA that followed 4,840 US adults aged 40 and older from the National Health and Nutrition Examination Survey (NHANES) who wore accelerometers for seven consecutive days. Over a median follow-up of 10.1 years, 1,165 participants died. The researchers examined the relationship between daily step count and all-cause mortality.

The most important finding is the enormous mortality reduction from moving out of the lowest step category. Going from fewer than 4,000 steps per day to 4,000 to 8,000 steps cut mortality risk by roughly half. This is a staggering effect size, comparable to the mortality benefit of some pharmaceutical interventions, and it requires nothing more than walking for 20 to 40 additional minutes per day.

Above 8,000 steps, the curve flattened. Walking 12,000 steps provided essentially the same mortality benefit as walking 8,000 steps. Crucially, step intensity (how fast you walked) was not associated with mortality after adjusting for total step volume. The number of steps, not the pace, was what mattered.

The Paluch meta-analysis: an even larger picture

A 2022 meta-analysis by Paluch et al. in The Lancet Public Health pooled 15 studies comprising 47,471 adults. The results refined the Saint-Maurice findings: each additional 1,000 steps per day was associated with a 15% reduction in all-cause mortality in adults aged 60 and over, and a 7% reduction in adults under 60. The optimal step count for adults over 60 was approximately 6,000 to 8,000 steps per day, while for adults under 60 it was 8,000 to 10,000 steps per day, beyond which benefits plateaued.

Notably, neither study found an increase in mortality at any step count. The 10,000-step-per-day target (popularised by a 1960s Japanese marketing campaign for pedometers and not based on any medical evidence at the time) turns out to be a reasonable, if slightly generous, approximation of the optimal range for younger adults.

Walking speed: an independent predictor

While step volume matters more than step intensity for the purposes of mortality reduction from walking, habitual walking speed is itself an independent predictor of survival. A 2019 study by Zaccardi et al. in the Mayo Clinic Proceedings, using UK Biobank data from 474,919 participants, found that self-reported slow walkers had significantly higher mortality than brisk walkers: life expectancy for slow-walking women was 72.4 years compared to 86.7 years for brisk-walking women. For men, the figures were 64.8 years and 85.2 years. The difference of 14.2 years for slow versus brisk walking women and 20.4 years for men was far larger than what exercise alone could plausibly explain, suggesting that walking speed serves as a composite biomarker of underlying health, much like grip strength.

HIIT vs LISS: what the randomised trials say

High-intensity interval training (HIIT) involves repeated bursts of near-maximal effort (typically 85 to 95% of maximum heart rate) alternated with recovery periods. Low-intensity steady-state training (LISS) involves sustained moderate effort (typically 50 to 70% of maximum heart rate) for 30 minutes or more. The debate over which is better for longevity has been one of the most active in exercise physiology over the past decade.

The GENERATION 100 trial

The most important randomised controlled trial addressing this question is the GENERATION 100 study, published by Stensvold et al. in the BMJ in 2020. This Norwegian trial randomised 1,567 adults aged 70 to 77 to three groups: supervised high-intensity interval training (HIIT, 4x4 minute intervals at 90% of peak heart rate, twice per week), moderate-intensity continuous training (MICT, 50 minutes at 70% of peak heart rate, twice per week), or a control group given general physical activity recommendations. Follow-up was 5 years.

The primary outcome was all-cause mortality. The results showed a trend favouring exercise over the control group, but the between-group differences were not statistically significant for the primary outcome (HIIT: 4.6% mortality, MICT: 5.9%, control: 5.7%). In secondary analyses, the HIIT group had significantly better improvements in peak oxygen uptake (VO2 peak) compared to both MICT and the control group. Since VO2 max is the strongest predictor of mortality, this improvement has important long-term implications even if the 5-year mortality endpoint did not reach significance (likely due to the relatively small sample size and short follow-up for a mortality trial in a group already below population-average mortality).

VO2 max improvements: the mechanistic argument for HIIT

A 2017 study by Weston et al. in the British Journal of Sports Medicine conducted a meta-analysis of 10 randomised trials comparing HIIT to moderate-intensity continuous training. HIIT produced a significantly greater improvement in VO2 max (mean difference 3.03 ml/kg/min, 95% CI 2.00 to 4.07), equivalent to approximately a 10% relative improvement over moderate-intensity training.

Given the Mandsager data showing that each increment in cardiorespiratory fitness produces a measurable mortality reduction, the superior VO2 max adaptations from HIIT likely translate to greater long-term survival benefits, even if direct mortality data from randomised trials is limited by practical constraints on sample size and follow-up duration.

The Cell study: HIIT and mitochondrial rejuvenation

A 2017 study by Robinson et al. in Cell Metabolism provided a cellular-level explanation for HIIT's benefits. The researchers randomised 72 sedentary adults (aged 18 to 30 and 65 to 80) to 12 weeks of HIIT (cycling), resistance training, or combined training. HIIT produced the largest improvements in mitochondrial function, with a 49% increase in mitochondrial capacity in older adults and a 69% increase in younger adults. HIIT also enhanced the expression of 274 genes in older adults (compared to 170 for combined training and 33 for resistance training alone), many of which were involved in mitochondrial biogenesis and protein quality control. The authors concluded that HIIT effectively reversed many age-related declines in mitochondrial function.

Exercise and cancer survival

Cancer is the second leading cause of death globally, responsible for approximately 10 million deaths per year. The relationship between physical activity and cancer risk is one of the most actively researched areas in oncology epidemiology, and the evidence is now substantial.

Primary prevention: reducing cancer incidence

A 2016 study by Moore et al. in JAMA Internal Medicine pooled 12 prospective cohort studies comprising 1.44 million participants and analysed the association between leisure-time physical activity and the incidence of 26 different cancer types. Higher physical activity was associated with reduced risk of 13 of 26 cancers, including:

• Oesophageal adenocarcinoma: 42% lower risk (HR 0.58)
• Liver cancer: 27% lower risk (HR 0.73)
• Lung cancer: 26% lower risk (HR 0.74)
• Kidney cancer: 23% lower risk (HR 0.77)
• Endometrial cancer: 21% lower risk (HR 0.79)
• Myeloid leukaemia: 20% lower risk (HR 0.80)
• Colon cancer: 16% lower risk (HR 0.84)
• Breast cancer: 10% lower risk (HR 0.90)

The associations were generally robust to adjustment for BMI, suggesting that the cancer-protective effects of exercise are not entirely mediated by weight management. Direct biological mechanisms include improved immune surveillance, reduced chronic inflammation (lower CRP, IL-6, and TNF-alpha), improved insulin sensitivity (reducing cancer-promoting hyperinsulinaemia), favourable changes in sex hormones, and enhanced DNA repair mechanisms.

Secondary prevention: exercise after diagnosis

The evidence for exercise improving survival after a cancer diagnosis is also substantial. A series of meta-analyses by Friedenreich et al., including a 2020 comprehensive review in the Journal of Clinical Oncology, synthesised the evidence from observational studies and randomised trials. Key findings include:

• Breast cancer: Post-diagnosis physical activity is associated with a 30 to 40% reduction in breast cancer-specific mortality and a 40 to 50% reduction in all-cause mortality (pooled across multiple studies).
• Colorectal cancer: Post-diagnosis physical activity is associated with a 30 to 40% reduction in colorectal cancer-specific mortality.
• Prostate cancer: Vigorous physical activity after diagnosis is associated with a 30% reduction in prostate cancer-specific mortality.

The magnitude of these effects is comparable to many pharmaceutical oncology interventions. The American Society of Clinical Oncology (ASCO), the American Cancer Society (ACS), and the National Comprehensive Cancer Network (NCCN) all now recommend exercise as an integral component of cancer treatment and survivorship care.

The mechanisms: how exercise fights cancer

The biological pathways through which exercise reduces cancer risk and improves cancer survival are numerous and increasingly well understood. A 2019 review by Pedersen et al. in Cell Metabolism identified several key mechanisms. First, exercise increases the mobilisation and tumour infiltration of natural killer (NK) cells and cytotoxic T cells through epinephrine-dependent pathways, enhancing immune surveillance against developing tumours. Second, exercise reduces circulating levels of insulin and insulin-like growth factor 1 (IGF-1), both of which promote cancer cell proliferation. Third, exercise reduces chronic systemic inflammation by lowering levels of pro-inflammatory cytokines and increasing anti-inflammatory mediators such as IL-10. Fourth, exercise improves endothelial function and reduces tumour angiogenesis by modulating vascular endothelial growth factor (VEGF) signalling. Fifth, exercise enhances DNA repair capacity and reduces oxidative DNA damage, potentially reducing the rate at which mutations accumulate.

In animal models, Pedersen et al. demonstrated that voluntary wheel running in mice reduced tumour growth by 60% across multiple tumour types, and this effect was largely mediated by epinephrine-driven NK cell infiltration into tumour tissue. While animal models do not directly translate to human outcomes, they provide mechanistic evidence that supports the epidemiological findings.

Exercise and the brain: neurogenesis, BDNF, and dementia prevention

Dementia, and Alzheimer's disease in particular, is among the most feared consequences of ageing. There is no cure, and pharmaceutical interventions have shown limited efficacy. Exercise, however, has emerged as one of the most promising modifiable risk factors for cognitive decline and dementia prevention.

Hippocampal volume: the Erickson study

A landmark 2011 randomised controlled trial by Erickson et al., published in the Proceedings of the National Academy of Sciences, randomised 120 older adults (mean age 66) to either a moderate-intensity aerobic walking programme (40 minutes per day, three days per week) or a stretching control group for one year. Before and after the intervention, participants underwent brain MRI to measure hippocampal volume.

The results were striking. The aerobic exercise group showed a 2% increase in hippocampal volume, while the stretching control group showed a 1.4% decrease. Given that the hippocampus normally shrinks by 1 to 2% per year in older adults, this effectively reversed age-related hippocampal atrophy by approximately 1 to 2 years. The increase in hippocampal volume was correlated with improvements in spatial memory and with increased serum levels of brain-derived neurotrophic factor (BDNF), a protein that supports the survival and growth of neurons.

BDNF: exercise as a brain fertiliser

Brain-derived neurotrophic factor (BDNF) is central to the relationship between exercise and brain health. BDNF promotes neurogenesis (the creation of new neurons, particularly in the hippocampus), synaptogenesis (the formation of new synaptic connections), and neuronal survival. A 2016 meta-analysis by Szuhany et al. in the Journal of Psychiatric Research, pooling 29 studies, found that a single bout of exercise increases BDNF levels acutely, and that regular exercise training elevates resting BDNF levels chronically. The effect was observed across all age groups, though the magnitude was greater in older adults.

Low BDNF levels are consistently found in patients with Alzheimer's disease, depression, and other neurodegenerative conditions. The exercise-induced increase in BDNF is one of the leading hypotheses for why exercise protects against cognitive decline, and it provides a direct molecular link between physical activity and brain structure.

Exercise and dementia risk: the epidemiological evidence

A 2017 meta-analysis by Hamer and Chida in Psychological Medicine pooled 16 prospective studies and found that high physical activity levels were associated with a 28% reduction in dementia risk (RR 0.72, 95% CI 0.60 to 0.86) and a 45% reduction in Alzheimer's disease risk (RR 0.55, 95% CI 0.36 to 0.84) compared to low physical activity. A 2019 study by Raichlen et al. in Neurology using UK Biobank data from 78,430 participants found that 10 minutes of moderate-to-vigorous physical activity per day was associated with significant reductions in all-cause dementia risk, including vascular dementia and Alzheimer's disease, over a mean follow-up of 6.9 years.

The evidence is consistent across study designs, populations, and follow-up durations: physical activity is one of the strongest protective factors against cognitive decline and dementia.

Exercise and depression: BDNF and beyond

The brain benefits of exercise extend well beyond dementia prevention. A 2023 umbrella meta-analysis by Singh et al. in the British Journal of Sports Medicine, synthesising 97 reviews comprising 1,039 randomised controlled trials and 128,119 participants, found that exercise was 1.5 times more effective than cognitive behavioural therapy (CBT) and pharmacotherapy for reducing symptoms of depression, anxiety, and psychological distress. Higher-intensity exercise produced larger benefits than lower-intensity exercise. Walking, jogging, yoga, and strength training all showed significant antidepressant effects.

Exercise and telomere length

Telomeres are the protective caps at the ends of chromosomes that shorten with each cell division. When telomeres become critically short, cells enter senescence (a state of permanent growth arrest) or undergo apoptosis (programmed death). Telomere length is considered a biomarker of biological ageing: shorter telomeres are associated with increased mortality, cardiovascular disease, cancer, and age-related diseases. The question of whether exercise can slow or reverse telomere shortening has been the subject of intense research interest.

The Puterman studies

Puterman et al. published several influential studies on this topic. A 2010 study in PLoS ONE followed 63 post-menopausal women over one year and found that vigorous physical activity buffered the effect of perceived psychological stress on telomere shortening. Among sedentary women, high stress was associated with significant telomere shortening. Among physically active women, high stress had no effect on telomere length. Exercise appeared to act as a "stress buffer" that protected cellular ageing from the damaging effects of psychological distress.

A 2015 follow-up study by Puterman et al. in Molecular Psychiatry, using data from 1,014 participants in the Health and Retirement Study, found that adults who did not exercise had shorter telomere length and faster telomere attrition over a two-year period compared to those who were physically active, after adjusting for age, sex, BMI, smoking, and chronic health conditions. The magnitude of the difference was equivalent to approximately 2 to 4 years of biological ageing.

The Werner study: exercise type and telomere biology

A 2018 randomised controlled trial by Werner et al. in the European Heart Journal randomised 124 previously sedentary adults aged 30 to 60 to three groups: endurance running, HIIT, or resistance training, for 26 weeks of three sessions per week. The researchers measured telomere length and telomerase activity (the enzyme that rebuilds telomeres) in white blood cells before and after the intervention.

Both endurance running and HIIT significantly increased telomerase activity (by approximately 2-fold) and telomere length compared to baseline. Resistance training did not produce significant changes in either measure. This suggests that the cellular anti-ageing effects of exercise are predominantly driven by aerobic, rather than resistance, training. The mechanism likely involves exercise-induced reductions in oxidative stress and chronic inflammation, both of which accelerate telomere shortening.

Population-level evidence

A 2017 study by Tucker in Preventive Medicine, using NHANES data from 5,823 adults, found a dose-response relationship between physical activity and telomere length. Adults with the highest levels of physical activity had telomere lengths corresponding to a biological age advantage of approximately 9 years compared to sedentary adults. Even moderate physical activity (walking, light cycling) was associated with significantly longer telomeres compared to inactivity.

Sitting: the independent risk that exercise cannot fully erase

One of the most alarming findings in modern epidemiology is that prolonged sitting appears to increase mortality risk independently of exercise. You can be a regular exerciser and still face elevated death risk if you spend the rest of your day sitting. This has led to the widely quoted (and partially accurate) claim that "sitting is the new smoking."

The Ekelund meta-analysis: 1 million participants

The definitive study on this topic was published by Ekelund et al. in The Lancet in 2016. It was a harmonised meta-analysis of 16 prospective cohort studies comprising 1,005,791 participants followed for 2 to 18.1 years, during which 84,609 deaths occurred. The researchers examined the joint effects of sitting time and physical activity on mortality.

The findings were nuanced and important. Among the least active individuals (those in the lowest quartile of physical activity, doing less than 5 minutes per day of moderate-intensity exercise):

• Sitting 8+ hours per day was associated with a 59% increase in all-cause mortality compared to sitting less than 4 hours per day (HR 1.59, 95% CI 1.39 to 1.82).
• Even sitting 4 to 8 hours per day was associated with a significant increase in mortality among the least active.

However, among the most active individuals (those in the highest quartile, doing approximately 60 to 75 minutes per day of moderate-intensity activity):

• The excess mortality risk from sitting 8+ hours per day was completely eliminated.
• High-volume physical activity effectively neutralised the harmful effects of prolonged sitting.

The practical message is twofold. First, if you sit for long periods (as most office workers do), you need more exercise than the standard guidelines suggest to offset the sitting risk. The standard recommendation of 150 minutes per week (roughly 21 minutes per day) is insufficient to fully counteract 8 or more hours of daily sitting. You need closer to 60 to 75 minutes per day of moderate activity, or approximately 420 to 525 minutes per week.

Second, reducing sitting time is independently beneficial, even for exercisers. Breaking up prolonged sitting with standing or walking breaks has been shown to improve metabolic markers. A 2016 study by Dempsey et al. in Diabetes Care randomised 24 adults with type 2 diabetes to different sitting-interruption protocols and found that breaking up sitting with 3-minute bouts of light walking every 30 minutes significantly reduced postprandial glucose, insulin, and triglyceride levels compared to uninterrupted sitting.

TV viewing: the most dangerous form of sitting

Not all sitting is equal. A 2011 study by Wijndaele et al. in the Journal of the American Heart Association found that TV viewing time was more strongly associated with mortality than total sitting time, likely because TV viewing is often accompanied by other unhealthy behaviours (snacking, alcohol consumption, later bedtimes). A 2012 meta-analysis by Grontved and Hu in JAMA estimated that every 2 hours per day of TV viewing was associated with a 20% increase in type 2 diabetes risk, a 15% increase in cardiovascular disease risk, and a 13% increase in all-cause mortality.

Standing desks and active workstations

The rise of standing desks, treadmill desks, and active sitting devices (such as stability balls and wobble chairs) has been driven by the sitting-mortality data. While randomised trial evidence for these interventions is still limited, a 2018 Cochrane review by Shrestha et al. found that sit-stand desks reduce sitting time by 30 minutes to 2 hours per day in the short term. Whether this magnitude of reduction is sufficient to meaningfully affect long-term mortality risk remains unclear, but it is a reasonable precautionary measure for individuals who cannot significantly reduce their daily sitting hours.

How the Death Clock calculator uses exercise data

The Death Clock life expectancy calculator incorporates exercise data through several input variables, each of which is calibrated against the epidemiological evidence described in this article.

Exercise frequency and intensity

The calculator asks users about their weekly exercise frequency and typical intensity level. These inputs are mapped to the dose-response curve established by Arem et al. (2015) and other studies. A user who reports no exercise receives the baseline life expectancy. A user who meets the physical activity guidelines (150 minutes of moderate activity or 75 minutes of vigorous activity per week) receives a life expectancy adjustment reflecting the 31% mortality reduction associated with guideline-level activity. Users who exceed the guidelines receive progressively larger adjustments, up to the plateau at approximately 3 times the recommended amount.

Exercise type and social context

The calculator also considers the type of exercise reported. Based on the Copenhagen City Heart Study data, social and team-based activities (racquet sports, team sports) receive larger life expectancy adjustments than solo activities (gym workouts, solitary running), reflecting the additional mortality benefit associated with socially interactive exercise. This does not penalise solo exercisers; rather, it provides an additional bonus for social exercisers that reflects the observed survival advantage.

Sedentary behaviour

Daily sitting time is a separate input variable. Based on the Ekelund et al. (2016) meta-analysis, users who report high sitting time (8 or more hours per day) receive a negative life expectancy adjustment that is partially offset by their exercise level. Users who report both high sitting time and high exercise volume see the sitting penalty reduced or eliminated, consistent with the study finding that 60 to 75 minutes of daily moderate activity neutralises the sitting risk.

Strength training

The calculator includes a specific question about resistance or strength training. Based on the Stamatakis et al. (2018) and Momma et al. (2022) data, users who report regular strength training receive an independent life expectancy adjustment of +0.5 to +1.5 years, depending on frequency. This adjustment is additive with the aerobic exercise adjustment, reflecting the independent mortality pathways of resistance and aerobic exercise.

Because exercise interacts with other risk factors (BMI, smoking, sleep, diet, social isolation, chronic disease), the calculator models these interactions rather than treating exercise as an isolated variable. A physically active smoker receives a different life expectancy adjustment than a physically active non-smoker, because the absolute risk reduction from exercise depends on the baseline risk level. Similarly, exercise provides larger absolute gains for individuals with elevated metabolic risk (obesity, prediabetes) than for those who are already metabolically healthy.

Full study reference table

Below is a summary of the major studies cited in this article, with their sample sizes, follow-up periods, and key findings.

This table represents only the studies discussed in detail in this article. The broader evidence base for exercise and mortality encompasses thousands of studies across multiple decades. The consistency of findings across different study designs, populations, and time periods provides a high degree of confidence in the overall conclusion: regular physical activity is one of the most powerful and well-evidenced interventions for extending human lifespan.

Note on methodology: Year-impact values used in the Death Clock calculator are derived from the hazard ratios and relative risks reported in these studies, converted to estimated years using standard actuarial life tables. All studies cited are peer-reviewed and published in indexed journals. Population-level statistics may not reflect individual risk, which is influenced by genetics, environment, and interaction effects between multiple lifestyle factors.