Lifestyle — the inputs that compound.

Sleep — the foundation everything sits on

Sleep is the lifestyle variable with the worst signal-to-noise ratio in popular health discourse and one of the highest signal-to-cost ratios in the trial literature. The healthy-adult target — seven to nine hours per night, with consistent timing — sits behind almost every meaningful health endpoint we measure, from glucose control to cardiovascular events to neurodegenerative risk.

The Sleep Heart Health Study and follow-up work established that habitual short sleep (< 6 hours) carries elevated cardiovascular and all-cause mortality risk after adjustment for confounders [Cappuccio 2010]. Acute restriction trials reliably impair insulin sensitivity, raise blood pressure, and shift appetite hormones (ghrelin up, leptin down) within days.

What works to improve sleep, in rough order of effect size: a consistent wake time (more important than bedtime), morning light exposure, restricted late-day caffeine, alcohol avoidance within three to four hours of bed, and a dark, cool bedroom. CBT-I (cognitive-behavioral therapy for insomnia) outperforms most pharmacologic sleep aids for chronic insomnia and is the first-line recommendation in clinical guidelines.

Pharmacologic sleep aids — Z-drugs, antihistamines, melatonin at the consumer doses sold — have a worse risk-benefit profile than the marketing suggests. Low-dose melatonin (0.3–0.5 mg, taken hours before sleep as a chronobiotic rather than a sedative) has the cleanest small-dose use case. The 5–10 mg tablets common in retail are pharmacologic doses sold as wellness products.

Cardiorespiratory fitness — the single biggest mortality lever

VO2max (maximal oxygen uptake) is the strongest single predictor of all-cause mortality in long-running cohorts of healthy adults. Mandsager and colleagues followed 122,000 patients through exercise treadmill testing and found a roughly five-fold hazard ratio for all-cause mortality between elite and low-fitness groups [Mandsager 2018]. That effect size exceeds smoking, diabetes, and coronary artery disease in that cohort.

The training protocol with the strongest evidence is a mix of zone-2 aerobic work (conversational pace, 3–4 hours weekly) and one to two weekly high-intensity sessions (4x4 intervals, sprint-interval training, or sport-specific high-intensity work). Zone-2 builds mitochondrial density and fat oxidation; high intensity raises the VO2max ceiling. Both pathways matter and train differently.

The dose-response is non-linear. Moving from sedentary to roughly 150 minutes/week of moderate activity captures most of the mortality benefit. Additional volume continues to add benefit at a diminishing rate, with the steepest segment of the curve at the sedentary end. The implication: the marginal hour of training does far more for an inactive person than for an already-fit one.

No supplement or pharmaceutical in the longevity literature comes close to the effect size of cardiorespiratory fitness in human outcome data. Anyone selling a longevity pill that doesn't put cardio at the center of the protocol is selling a partial story.

Cardiorespiratory fitness is the most-studied modifiable variable in all-cause mortality. No drug, peptide, or supplement on the market has produced comparable effect sizes in humans. The marginal hour of training will out-perform the marginal capsule every time.

Resistance training — the muscle-as-organ argument

Skeletal muscle is among the largest endocrine organs in the body and the only longevity-relevant tissue that responds directly to mechanical load. Muscle mass and strength independently predict all-cause mortality after adjustment for cardiorespiratory fitness — they're additive levers, not redundant ones [García-Hermoso 2018].

Sarcopenia — age-related muscle loss — is one of the few aging phenotypes with a clear and modifiable intervention: progressive resistance training. The literature supports two sessions per week as a minimum effective dose for general health, with three to four sessions per week producing larger strength and hypertrophy gains. Volume and proximity to failure matter more than specific exercise selection.

For older adults, the case is even stronger. Resistance training in adults over 65 improves functional outcomes (chair stand, gait speed, balance) and reduces fall risk, with downstream effects on independent living. These are not aesthetic outcomes; they are among the strongest predictors of whether someone stays out of long-term care.

Protein intake — the right floor for adults

The RDA (recommended dietary allowance) for protein in adults is 0.8 g/kg body weight per day — a floor set to prevent deficiency, not to optimize muscle mass or healthy aging. Contemporary evidence supports a substantially higher intake for adults engaged in resistance training, for older adults, and for those pursuing body-composition goals.

Morton and colleagues' meta-analysis identified roughly 1.6 g/kg as the upper bound of meaningful added benefit for strength and lean mass adaptations to resistance training in young adults [Morton 2018]. The PROT-AGE working group recommended 1.0–1.2 g/kg as a floor for healthy older adults and 1.2–1.5 g/kg in the context of acute or chronic illness [Bauer 2013].

Distribution matters: 25–40 g of protein per meal across three to four meals maximizes muscle protein synthesis on a given total intake. Leucine content is the key amino-acid driver. Animal sources reach the leucine threshold at lower total doses; plant sources require larger total doses but reach the same end.

Stress — chronic vs acute, and what actually helps

Stress is a poorly bounded concept in popular health discourse. Acute stress — short, intense, time-limited — is the substrate of most adaptive responses (training, learning, recovery). Chronic stress — sustained activation of the HPA (hypothalamic-pituitary- adrenal) axis without recovery — is what drives the cardiovascular, metabolic, and immune costs that show up in cohort data.

Interventions with the strongest trial support: aerobic exercise (which dose-dependently reduces depressive and anxiety symptoms), cognitive-behavioral therapy, mindfulness-based stress reduction (Kabat-Zinn protocol and its successors), and adequate sleep. Goyal and colleagues' meta-analysis is the cleanest summary of the meditation literature on anxiety and depression [Goyal 2014].

Where the evidence is weaker: most adaptogenic supplements (ashwagandha has the best trial base, but effect sizes are modest and replication is mixed), most "stress" branded wellness products, and most HRV-biofeedback consumer apps. These can be useful adjuncts; they don't replace the high-evidence interventions.

Social connection — the underrated longevity variable

The mortality effect of loneliness in long-running cohorts is comparable to that of smoking 15 cigarettes a day. Holt-Lunstad and colleagues' meta-analytic work is the most-cited summary, with effect sizes that rival traditional cardiovascular risk factors [Holt-Lunstad 2010]. Social isolation is harder to operationalize as a protocol than sleep or exercise, but it is impossible to ignore as a finding.

The mechanisms are plausible and partly mapped: chronic loneliness elevates inflammatory markers, disrupts sleep, and shifts health behaviors. The effect is independent of objective social network size — felt loneliness, not headcount, is the active variable.

What seems to help: regular high-quality contact with a small number of people, shared physical activity, intergenerational relationships, and structured community (faith, sport, civic groups). What doesn't help much: passive social media, parasocial relationships, or substituting digital contact for in-person contact. The substitution effect is one of the more concerning signals in adolescent and young-adult cohort data.

The Pareto stack — what 80% of the result comes from

If you wanted to build the highest-evidence lifestyle stack and ignore everything else, it would look approximately like this:

Move toward seven to nine hours of sleep with consistent timing. Hit 150–300 minutes per week of cardiovascular work — zone-2 plus one or two high-intensity sessions. Lift twice a week, progressing in load and proximity to failure. Eat 1.2–1.6 g/kg of protein spread across three to four meals. Get 30+ grams of mixed dietary fiber. Drink less alcohol than you currently do. Maintain regular in-person contact with people who matter. Get outside in the morning.

That stack — none of which can be sold at a premium — accounts for the great majority of modifiable mortality risk in human outcome data. Every supplement and pharmaceutical in the longevity marketplace is a rounding error on whether someone is doing those things. The honest hierarchy is the unsexy one.

Critically, none of this requires precision tracking. Approximate adherence over years beats precise adherence over weeks. The lifestyle interventions reward consistency and tolerate imperfection in a way that pharmacology and supplementation usually don't.