The 6 Pillars of Longevity

Search for the pillars of longevity and you will find three, four, five, six, seven or nine, depending on the source. That is not because one source is right. It is because each source answers a different question.

German statutory health insurers group prevention into roughly four fields: physical activity, nutrition, stress management and substance use. That split comes from the prevention framework under section 20 of the German social code ^[19]. The Kneipp tradition names five elements: water, movement, nutrition, medicinal plants and a balanced daily routine. That is a nineteenth-century system of thought, not a study result. Dan Buettner describes nine behaviours for the Blue Zones, the so-called Power 9. That is an observation of regions with many very old residents, not an intervention trial.

Side by side, the best-known lists look like this:

Framework	Count	Pillars	Basis
Peter Attia, Outlive	4	exercise, nutrition, sleep, emotional health	a doctor's book
German statutory insurers (GKV)	4	activity, nutrition, stress, substance use	prevention funding rules
Common five-pillar lists	5	movement, nutrition, sleep, stress reduction, social connection	health media summaries
Kneipp	5	water, movement, nutrition, medicinal plants, daily routine	19th-century system
Lifestyle medicine (ACLM)	6	physical activity, plant-predominant eating, sleep, stress, social connection, avoiding risky substances	professional society
Blue Zones Power 9	9	behaviours seen in long-lived regions	observation
This guide	6	movement, nutrition, sleep, social connection, stress, prevention	mortality data for five, statutory screening for the sixth

Our split answers a narrower question. Which lifestyle domains have large prospective cohorts or meta-analyses with all-cause mortality as the endpoint? Five clear this bar: cardiorespiratory fitness, dietary pattern, sleep, social integration and chronic stress ^{[4, 10, 13, 22, 23]}.

How much of your lifespan is in your genes? It depends on the method:

• Classic twin studies: around 20 to 30 percent ^[1]
• A large family-tree analysis: under 10 percent, after correcting for assortative mating. People tend to pick partners like themselves, so a shared lifestyle can look like shared genes ^[2]
• An analysis that leaves out external causes of death such as accidents and infections: roughly 50 percent ^[3]

The spread is wide. In all three estimates a large share is not explained by genes, and part of that share is lifestyle.

The sixth pillar is prevention and screening. It earns its place for a practical reason, not because of mortality data. In Germany early detection is a real lever you already pay for through your contributions. The general health check from age 35 and cancer screening are standard statutory benefits, defined in the directives of the Federal Joint Committee ^{[17, 18]}. A risk that becomes visible early can be managed differently from one that surfaces in an emergency.

The takeaway: the number is a way of sorting the evidence, not a law of nature. Sources that name four pillars usually leave out social connection, and some also drop sleep. Sources that name nine are counting behaviours instead of domains. What the term itself means is summarised in the longevity glossary and unpacked at length in the guide What is longevity?.

Of the six pillars, movement has the hardest numbers. One study looked back at 122,007 people who had done a treadmill exercise test ^[4]. Higher cardiorespiratory fitness, meaning how well your heart and lungs deliver oxygen during hard effort, went along with a lower risk of dying from any cause. The researchers observed no upper threshold: even at the top end, more fitness still went along with lower risk. The least fit group had about five times the risk of death of the elite group (adjusted hazard ratio 5.04). Smoking and diabetes each came in at about 1.4 in the same data.

Strength is its own factor, not a by-product of endurance work. A meta-analysis of prospective cohorts (studies that follow people forward in time) found that muscle-strengthening activity went along with a 10 to 17 percent lower risk ^[5]. That held across several outcomes: all-cause mortality, cardiovascular disease, cancer and diabetes. The strongest link sat at around 30 to 60 minutes per week. Beyond that the curve flattens, and in some analyses it turns slightly upward at very high volumes.

The WHO physical activity recommendation for adults names 150 to 300 minutes of moderate or 75 to 150 minutes of vigorous aerobic activity per week, plus muscle-strengthening work on at least two days ^[6]. That is a population-level recommendation rather than an individual prescription. What fits you depends on existing conditions, joints and your starting point.

Two markers make your current state visible. VO2 max describes aerobic capacity. Grip strength is a rough stand-in for overall muscle strength. You can estimate both here at no cost: the VO2 max calculator and the grip strength comparison.

For more depth, see Exercise for longevity and Zone 2 and VO2 max training. Local chapters run running groups and shared training sessions you can join without a fee.

In the evidence, nutrition works through patterns rather than single foods or capsules. The clearest example is PREDIMED. It randomised 7,447 people at high cardiovascular risk, meaning chance decided who got which diet: a Mediterranean diet with extra-virgin olive oil, one with nuts, or a reduced-fat control diet. In the 2018 analysis, the rate of major cardiovascular events was lower in both Mediterranean groups than in the control group ^[7]. It remains one of the few randomised diet trials with hard endpoints, meaning real events like heart attacks and strokes rather than lab values.

One detail matters when you read it. The original 2013 analysis was retracted in 2018 because some sites had not randomised correctly. The 2018 paper is the corrected republication, and it no longer assumes that every participant was randomised.

On all-cause mortality specifically, a meta-analysis of eight prospective cohorts with over 500,000 people found that each 2-point rise on a Mediterranean diet adherence score went along with 9 percent lower mortality ^[23].

How big could the lever be in theory? A modelling study estimated it from meta-analyses and Global Burden of Disease data, for a US population ^[8]. It modelled a lasting switch from a typical Western diet to an optimised one:

• Starting at 20: a projected 10.7 extra years for women and 13.0 for men
• Starting at 60: a projected 8.0 extra years for women and 8.8 for men

The largest single gains came from more legumes, more whole grains and more nuts. A model is not a measurement, but it does frame the order of magnitude.

One point grows more important with age: protein. Muscle mass and strength decline from midlife onward. An international expert position paper lists protein, alongside strength training, among the factors discussed for keeping muscle as you age ^[9]. What that means for you depends on body weight, activity and kidney function. The protein calculator puts the general figures into your own numbers.

The patterns themselves, from Mediterranean to plant-forward to the eating habits of the Blue Zones, are unpacked in The longevity diet.

Sleep duration and mortality follow a U shape: risk is higher at both ends and lowest in the middle. A meta-analysis pooled 16 prospective studies with 27 cohorts and 1,382,999 participants ^[10]. Compared with the middle group, short sleepers had a 12 percent higher risk of death and long sleepers a 30 percent higher risk. Long sleep is very likely a marker of existing illness rather than its cause. That is the classic weakness of observational data.

A second finding is more useful day to day. In an analysis of wrist-worn accelerometer data (actigraphy) from around 60,000 UK Biobank participants, the regularity of sleep predicted all-cause mortality more strongly than sleep duration did ^[11]. Regularity here means how similar your sleep and wake times are from day to day, not how long you sleep.

Together these shift the focus. Rather than aiming at a target number of hours, it is worth looking at how stable your rhythm is across the full week, weekends included. Wearables estimate sleep stages only roughly. The times you go to bed and get up are easier to capture ^[12]. So tracking regularity does not need an expensive device.

One caveat: all of these findings are observational. Trials assigning people to years of different sleep patterns do not exist. The link is well documented, but which way cause runs remains partly open.

How sleep works physiologically, from glymphatic clearance to hormonal rhythm, is covered in Sleep and longevity. Deep sleep specifically is covered in Deep sleep.

The pillar most often missing from popular lists has one of the clearest evidence bases. A meta-analysis of 148 studies covering 308,849 participants found that people with strong social relationships had roughly a 50 percent higher likelihood of survival across follow-up ^[13]. The authors placed the effect in the same range as established lifestyle risk factors.

The opposite direction was analysed separately in 2015. Across 70 studies, loneliness, objective social isolation and living alone were each associated with increased mortality, in the range of roughly 26 to 32 percent ^[14]. The key detail: subjective loneliness (feeling alone) and objective isolation (having few contacts) were pooled separately, and both tracked with risk to a similar degree.

That makes this the one pillar where the intervention itself costs nothing and needs no equipment. It is also the reason this community exists at all. Regular meetings with people who care about the same topics are not a side note in this framing, they are a pillar.

What is happening and where is listed under Events. Which cities are active is shown on the Chapters page.

Chronic stress is the hardest pillar to measure, and the effects found so far are moderate. A review in Nature Reviews Cardiology sums up the cohort data ^[15]. Sustained psychosocial strain, meaning pressure that does not let up, such as long-term work stress or chronic exhaustion, goes along with a higher risk of coronary heart disease (narrowed heart arteries) and stroke. The review names plausible routes: blood pressure, inflammation and changes in behaviour. The effect sizes are smaller than those for smoking or high blood pressure.

For all-cause mortality, a pooled analysis of 10 English cohorts with 68,222 adults found that risk rose step by step with psychological distress, even at mild levels. At the highest distress scores it was almost twice as high ^[22].

Because the experience of stress is subjective, many people reach for a physiological proxy: heart rate variability (HRV), the small shifts in timing from one heartbeat to the next. It varies widely between individuals with age, breathing and measurement method, so it works better as a trend line for yourself than as a comparison with other people. What HRV numbers can and cannot tell you is covered in HRV and wearables.

A second research strand looks at brief, dosed strain as a counterpart to chronic stress, usually under the heading of hormesis. Heat and cold exposure belong here. The evidence base is considerably thinner than for movement or sleep and rests largely on small studies plus one Finnish cohort. We have covered both separately: Sauna and longevity and Cold exposure and longevity.

On relaxation techniques, breathing practices or meditation we deliberately name no durations or frequencies. The trials are heterogeneous. What actually lowers strain for you is a personal question with no cohort data behind it.

The sixth pillar is the only one made of appointments rather than behaviour. In Germany the Federal Joint Committee (G-BA, the body that decides what statutory health insurance pays for) sets the early-detection benefits. The main one is the general health check, known as the check-up 35 (health examination directive):

• From 35: every three years. It covers history taking, a physical exam, blood pressure, a lipid panel (blood fats), fasting glucose and a urine test. Once from 35 you can also be tested for hepatitis B and C.
• Between 18 and 34: once. Lab work is done only if your risk profile calls for it.

Men from 65 are also entitled to a one-off ultrasound of the abdominal aorta, the main artery in the belly, to spot a dangerous bulge early. Cancer screening is governed separately and staggered by age (cancer screening directive). The federal health ministry keeps a plain-language overview.

One caveat: a Cochrane review of 17 randomised trials found that general health checks had little or no effect on total mortality ^[20]. A meta-analysis of cancer screening trials found a significant lifetime gain only for sigmoidoscopy, a shorter form of bowel scope ^[21]. Screening pays off for specific, treatable findings. It is not a mortality lever on the scale of the behaviour pillars.

Beyond the statutory catalogue sit markers that come up constantly in longevity discussions: apolipoprotein B, lipoprotein(a), HbA1c and high-sensitivity CRP. What each one indicates is set out in Longevity blood tests and biomarkers. For cardiovascular risk assessment the coronary calcium score is also debated, a CT finding with its own discussion around benefit, radiation dose and cost: CAC score and the power of zero.

Prevention also covers vaccination according to national immunisation recommendations, not smoking and going easy on alcohol. How much lifestyle factors weigh when they come together shows in an analysis of two large US cohorts ^[16]. It looked at five low-risk factors. Vaccination and screening were not among them:

• never smoking
• a healthy body weight
• regular physical activity
• a good dietary pattern
• moderate alcohol intake

At age 50, people who met all five had a projected life expectancy about 14 years longer for women and about 12 years longer for men than people who met none.

Many lists rank the pillars, because a ranking looks tidy. The evidence does not support one. The studies behind the six domains differ in design, population, follow-up and endpoint, so their effect sizes cannot be placed side by side.

Two things can still be said. First, the largest single association in the data cited here sits with cardiorespiratory fitness ^[4]. Anyone at the bottom end there has the most to gain arithmetically. Second, the combination outweighs any individual factor. That is exactly what the five low-risk factor analysis shows ^[16], where projected life expectancy rose with each additional factor met.

A third point gets lost in rankings: the pillars do not stand apart from one another. Poor sleep lowers the appetite for training the next day, chronic stress changes eating behaviour, and for many people movement happens in social groups anyway. Change one pillar and a second usually moves with it.

In practice this means the most important pillar is the one currently weakest for you. Someone already sleeping seven and a half hours on a stable schedule gains little from further sleep optimisation. Someone with no aerobic activity for years is standing on the steep part of the curve.

If you want an order to work in, an honest inventory across all six domains beats any general ranking.

The six pillars stay abstract as long as they are only a list. Here is how they translate into an ordinary month without a budget.

Movement: several cities run running groups and shared training sessions. Dates are listed under Events. You can place your own starting point roughly with the VO2 max calculator.

Nutrition: the protein calculator and the other tools turn general recommendations into your own numbers. Recipes and dietary patterns sit in the guides.

Sleep: nothing to buy here. A stable bed and wake time across the week is the lever you can track without a device ^[11].

Social connection: this is the part a community covers directly. Meetups are free and the chapters are organised locally.

Stress management: what is telling is your own trend, not the comparison with other people. What is measurable is covered in the HRV guide.

Prevention: one phone call to a practice is enough to book the check-up you are entitled to.

None of this requires a purchase or a subscription. That follows from the evidence: the studies behind the six pillars examine behaviour and care, not products.

The graded studies our guides rest on are listed under Research, each with our assessment of how strong it is.