Healthspan is behavioral. The uncomfortable truth for the supplement industry is that the levers with the largest effect sizes cost very little and require the discipline of repetition, not the thrill of novelty. This stack ranks interventions by evidence strength and effect size, and tells you in what order to implement them.
What are the non-negotiable Tier 1 longevity levers?
These are the levers with large-cohort or RCT-level evidence and effect sizes no pharmaceutical class approaches.
Cardiorespiratory fitness
Mandsager 2018 (n=122,007, JAMA Network Open) found each 1-MET higher CRF associated with ~11% lower all-cause mortality ( Mandsager et al. 2018, n=122007 ). The jump from "below average" to "above average" CRF cut mortality hazard ~50%. Dose-response continued into elite fitness.
Target: VO2 max in the 75th-90th percentile for your age + sex band. For a 40-year-old man, that's roughly 42-48 ml/kg/min.
- 3-4 Zone-2 sessions/week, 45-60 minutes, 65-75% HRmax.
- 1-2 VO2 max sessions/week, 4x4 intervals or equivalent.
- See Zone-2 and VO2 Max for the prescription.
Resistance training
Sarcopenia and osteopenia are the path through which most people lose function. Schoenfeld 2017 (meta, 34 RCTs) showed dose-response up to ~10 sets per muscle per week in untrained-to-intermediate lifters ( Schoenfeld et al. 2016 ). Paoli 2024 review established the minimum effective dose: 30-60 min/week of properly programmed resistance can preserve most strength and lean mass in the general population ( Paoli et al. 2024 ).
Target: 2-3 full-body sessions/week of 30-45 min each. Compound lifts (squat/hinge/push/pull) across the week.
Protocol: See Resistance Training Minimum Effective Dose.
Protein intake
Morton 2018 meta-analysis (n=1,863) plateaus at ~1.62 g/kg/day for active adults ( Morton et al. 2017, n=1863 ); Phillips 2016 argues higher (2.0-2.2) for older adults and anyone in a deficit ( Phillips et al. 2016 ).
Target: 1.6-2.2 g/kg/day, split across 3-4 meals. See Protein Targets for Longevity.
Sleep
7-9 hours on consistent timing. Beyond duration, variability in timing independently predicts cardiometabolic disease. Bedroom 16-19°C; caffeine cutoff 6-8 hours pre-bed; no alcohol within 4 hours of bed. See Sleep Architecture + Sleep Hygiene Ranked.
Cardiometabolic management
| Phase | Dose | Notes |
|---|---|---|
| ApoB | < 80 mg/dL, ideally < 60 | LDL-C is a fine proxy but ApoB is better; test at least annually. |
| Blood pressure | < 130/80 mmHg | Home cuff, 3-day morning average. Lower floor debated; 115-120/75 is aspirational. |
| HbA1c | < 5.6% | Fasting glucose + insulin more sensitive early; HbA1c captures 3-month average. |
| Fasting insulin | < 8 uIU/mL | Marker of early insulin resistance before glucose moves. |
| hsCRP | < 1.0 mg/L | Chronic inflammation proxy; elevated values warrant a pattern check, not a reflex statin. |
| Omega-3 Index | 8-12% | See the omega-3 article; supplementation only to correct a measured gap. |
Add statin or ezetimibe if lifestyle alone can't hit ApoB target; most adults 40+ eventually need one. Not medical advice; clinician required.
Each Tier 1 lever, mapped to its strongest single trial
| Study | N | Duration | Design | Outcome | Finding |
|---|---|---|---|---|---|
| Mandsager 2018 (CRF) cite | 122,007 | median 8.4 yr | Retrospective cohort, treadmill VO2 max | All-cause mortality | Each 1-MET higher CRF: ~11% lower all-cause mortality; bottom quartile fitness stronger predictor than smoking |
| Schoenfeld 2017 (volume) cite | pooled across 34 RCTs | trial range 6-12 wk | meta-analysis of weekly per-muscle volume | Hypertrophy | Dose-response up to ~10 sets/muscle/week; gains continue but diminish past that |
| Morton 2018 (protein dose) cite | 1,863 across 49 RCTs | trial range 6-52 wk | meta-analysis of training plus protein RCTs | Fat-free mass and 1RM | Lean-mass plateau at ~1.62 g/kg/day; older adults plateau slightly higher |
| Laukkanen 2015 (sauna) cite | 2,315 | 21 yr follow-up | Finnish KIHD prospective cohort | All-cause and CV mortality | 4-7 sessions/wk associated with ~40% lower all-cause and ~63% lower sudden cardiac death vs 1 session/wk; observational |
| Mannick 2018 (rapamycin) cite | 264 | 12 wk dosing plus 6 wk follow-up | double-blind RCT (TORC1 inhibitor in older adults) | Vaccine response and respiratory infections | RAD001 dosing improved post-vaccination antibody titers and reduced infections in healthy elderly |
| STEP 1 (Wilding 2021) cite | 1,961 | 68 wk | double-blind RCT (semaglutide 2.4 mg/wk) | Mean body weight change | 14.9% mean loss vs 2.4% placebo; cardiovascular benefit confirmed in SELECT |
Synthesis The intervention with the largest mortality effect at population scale is cardiorespiratory fitness, by a wide margin. Tier 2 and Tier 3 candidates each have a defensible single trial, but none approach the magnitude of the CRF and sleep cohort signals on hard outcomes.
Which Tier 2 longevity additions earn their place?
Sauna
Laukkanen 2015 (Finnish cohort, n=2,315, 21-year follow-up) found 4-7 sauna sessions/week associated with ~40% lower all-cause mortality and ~63% lower sudden cardiac death vs 1 session/week ( Laukkanen et al. 2015, n=2315 ). Observational, not randomized, but the mechanism story (heat-shock proteins, endothelial adaptation) is sound.
Target: 3-4 sessions/week, 20 min at 80-100°C. See Sauna for Cardiovascular Healthspan.
Social connection
Loneliness predicts mortality at roughly the magnitude of moderate smoking in cohort data. Hard to operationalize; real. Regular non-transactional contact with a handful of humans is the floor.
Smoking abstinence and alcohol moderation
Smoking shaves ~10 years off life expectancy. Alcohol beyond ~7 drinks/week associates with worse cardiometabolic and cancer outcomes; the "two glasses of wine" cardioprotective story has mostly collapsed in modern analyses. Low alcohol is safer than moderate; zero is safer than low.
Sun exposure + vitamin D sufficiency
Outdoor light 10+ minutes/morning anchors circadian phase. 25-OH vitamin D 40-60 ng/mL target via ambient sun + diet; supplement only if measured low. See Vitamin D and K2.
Which Tier 3 longevity interventions are worth the bet?
These belong here because evidence is thinner or effect sizes smaller. Worth exploring only after Tier 1 is solid.
Rapamycin
Mannick 2018 (n=264) showed TORC1 inhibition improved immune function in healthy elderly ( Mannick et al. 2018, n=264 ). Long-term healthspan benefit in humans remains under-powered. Typical weekly dosing (5-6 mg/week), cycled, is the emerging off-label protocol. See the rapamycin compound entry.
GLP-1 receptor agonists
Semaglutide (STEP 1, n=1,961) produced ~14.9% weight loss at 68 weeks with cardiovascular benefit in high-risk patients ( Wilding et al. (STEP 1) 2021, n=1961 ). If overweight with cardiometabolic risk, GLP-1 has a case. For healthy-BMI adults, the longevity-specific rationale doesn't exist yet. See GLP-1s Without the Hype.
NAD+ precursors (NMN, NR)
Animal data suggestive; human RCTs mixed, effect sizes small. No intervention has reversibly or durably raised NAD+ levels in ways that map to outcome improvements.
Senolytics (dasatinib + quercetin, fisetin)
Promising preclinically. No positive RCT in humans as of 2026.
Peptides (BPC-157, thymosin, epithalon)
Mechanistic stories are interesting. Human RCT data is thin-to-absent. See BPC-157 entry for the state of play.
How to sequence
- Weeks 1-12: Build Tier 1 habit grid. Do not add Tier 2 or Tier 3. The first 12 weeks are about showing up.
- Weeks 13-24: Add Tier 2 (sauna if accessible, social discipline).
- Month 7+: Consider Tier 3 only if Tier 1 is stable. Rapamycin first if interested in off-label longevity protocols.
- Annual: Full panel + wearable trend audit. Adjust ApoB target, sleep regimen, training periodization based on the data.
Counter-view
Peter Attia is more aggressive on volume (4 Zone-2 + 3 resistance + 1-2 VO2 max sessions/week) and on rapamycin. Matt Kaeberlein is more skeptical of pharmacologic longevity and more emphatic on the basics. Valter Longo advocates periodic fasting and protein cycling as orthogonal levers to the resistance-training protein prescription above. All three are defensible; the overlap across them is exactly Tier 1.
