Why this works, and what is not proven

1. Cardiorespiratory fitness and mortality — Cleveland Clinic cohort.

Mandsager K et al., JAMA Netw Open 2018;1(6):e183605. PMID 30646252. DOI https://doi.org/10.1001/jamanetworkopen.2018.3605

• 122,007 patients referred for symptom-limited treadmill testing, median follow-up 8.4 years; 13,637 deaths over 1.1 million person-years.
• Risk-adjusted all-cause mortality was inversely proportional to cardiorespiratory fitness; low vs elite fitness adjusted HR 5.04 (95% CI 4.10–6.20); below-average vs above-average HR 1.41 (1.34–1.49).
• The mortality increase associated with reduced fitness was "comparable to or greater than" coronary artery disease (HR 1.29), smoking (HR 1.41) and diabetes (HR 1.40).
• Benefit of elite over high fitness was present in patients 70 years or older (HR 0.71, 0.52–0.98). "No observed upper limit of benefit."

2. Cardiorespiratory fitness and mortality across age, race and sex — 750,302 US veterans.

Kokkinos P et al., J Am Coll Cardiol 2022;80(6):598-609. PMID 35926933. DOI https://doi.org/10.1016/j.jacc.2022.05.031

• 750,302 veterans aged 30–95 (including 110,637 septuagenarians and 26,989 octogenarians), median follow-up 10.2 years, 174,807 deaths.
• Association inverse and graded across the whole age spectrum; least fit (20th percentile) had 4-fold higher risk vs extremely fit (HR 4.09, 3.90–4.20); no increased risk with extreme fitness.
• "Being unfit carried a greater risk than any of the cardiac risk factors examined."

3. Grip strength predicts death — PURE study, 17 countries.

Leong DP et al., Lancet 2015;386(9990):266-73. PMID 25982160. DOI https://doi.org/10.1016/S0140-6736(14)62000-6

• 139,691 participants aged 35–70, median follow-up 4.0 years; grip measured with a Jamar dynamometer.
• Each 5 kg lower grip strength: all-cause mortality HR 1.16 (1.13–1.20); cardiovascular mortality 1.17; myocardial infarction 1.07; stroke 1.09.
• "Grip strength was a stronger predictor of all-cause and cardiovascular mortality than systolic blood pressure."
• No significant association with injury from falls or fracture in this cohort. The authors call for trials testing whether improving strength reduces mortality — i.e., this is prognostic evidence, not proof that grip training extends life.

4. The 10-second one-legged stance and survival.

Araujo CG et al., Br J Sports Med 2022;56(17):975-980. PMID 35728834. DOI https://doi.org/10.1136/bjsports-2021-105360

• 1,702 individuals aged 51–75; 20.4% could not complete a 10-second one-legged stance; median follow-up 7 years.
• Deaths: 4.6% of those who could hold 10 s vs 17.5% of those who could not; adjusted HR for all-cause mortality 1.84 (1.23–2.78) for inability.
• Limitations stated by the authors: uncontrolled variables such as recent falls history and physical activity.

5. Gait speed and survival — 9 pooled cohorts.

Studenski S et al., JAMA 2011;305(1):50-8. PMID 21205966. DOI https://doi.org/10.1001/jama.2010.1923

• 34,485 community-dwelling adults aged 65+, followed 6–21 years; mean gait speed 0.92 m/s.
• Pooled HR per 0.1 m/s faster gait speed 0.88 (0.87–0.90); survival increased across the full range of gait speeds.
• At age 75, predicted 10-year survival ranged from 19% to 87% in men and 35% to 91% in women across the range of gait speeds.

6. Muscle-strengthening activity and mortality — meta-analysis of 16 cohort studies.

Momma H et al., Br J Sports Med 2022;56(13):755-763. PMID 35228201. DOI https://doi.org/10.1136/bjsports-2021-105061

• Muscle-strengthening activities associated with 10–17% lower risk of all-cause mortality, cardiovascular disease, total cancer, diabetes and lung cancer, independent of aerobic activity.
• J-shaped association with maximum risk reduction (about 10–20%) at about 30–60 min/week of muscle-strengthening activity for all-cause mortality, CVD and total cancer.
• Combined muscle-strengthening and aerobic activity (vs none) associated with lower all-cause, CVD and total cancer mortality.

7. Weightlifting plus aerobic activity — PLCO cohort, 99,713 adults, mean age 71.3.

Gorzelitz J et al., Br J Sports Med 2022;56(22):1277-1283. PMID 36167669. DOI https://doi.org/10.1136/bjsports-2021-105315

• Weightlifting associated with 9% lower all-cause mortality (HR 0.91, 0.88–0.94) after adjusting for aerobic activity.
• Adults meeting aerobic recommendations without weightlifting: 32% lower all-cause mortality (HR 0.68); adding weightlifting 1–2 times/week: 41% lower (HR 0.59, 0.54–0.64), both vs adults reporting neither.

8. Daily steps and mortality — 15 cohorts, 47,471 adults.

Paluch AE et al., Lancet Public Health 2022;7(3):e219-e228. PMID 35247352. DOI https://doi.org/10.1016/S2468-2667(21)00302-9

• Compared with the lowest quartile (median 3,553 steps/day), HR 0.60 for quartile 2 (median 5,801), 0.55 for quartile 3 (7,842), 0.47 for quartile 4 (10,901).
• Among adults 60 and older, risk decreased progressively until about 6,000–8,000 steps/day (8,000–10,000 in adults under 60). Stepping rate added little once total steps were accounted for.

9. Short bursts of vigorous everyday activity — UK Biobank, 25,241 non-exercisers.

Stamatakis E et al., Nat Med 2022;28(12):2521-2529. PMID 36482104. DOI https://doi.org/10.1038/s41591-022-02100-x

• Mean age 61.8; median 3 bouts/day of 1–2 minutes of vigorous intermittent lifestyle activity (VILPA) associated with 38–40% lower all-cause and cancer mortality and 48–49% lower CVD mortality vs none; the median 4.4 min/day associated with 26–30% lower all-cause mortality.
• Observational; the authors suggest VILPA "may be a suitable physical activity target, especially in people not able or willing to exercise."

10. Aerobic training raises VO2max in sedentary older adults — meta-analysis, 1,257 exercisers / 845 controls, mean age 67.

Huang G et al., Eur J Prev Cardiol 2016;23(5):518-29. PMID 25901000. DOI https://doi.org/10.1177/2047487315582322

• Weighted net VO2max increase 3.78 mL/kg/min (95% CI 3.29–4.27).
• Intensity mattered: gains from 35–50% heart-rate reserve upward; the largest adaptation at a mean intensity of 66–73% HRR; intensities above 75–80% HRR did not add more. Effective/optimal prescription in the meta-analysis: 40–50 min per session, 3–4 days/week, 30–40 weeks.

11. Resistance training raises strength in adults 50+ — meta-analysis of 47 studies, 1,079 participants.

Peterson MD et al., Ageing Res Rev 2010;9(3):226-37. PMID 20385254. DOI https://doi.org/10.1016/j.arr.2010.03.004

• Percent strength changes: leg press +29%, chest press +24%, knee extension +33%, lat pull +25%; higher-intensity training produced greater improvement.

12. NSCA position statement on resistance training for older adults.

Fragala MS et al., J Strength Cond Res 2019;33(8):2019-2052. PMID 31343601. DOI https://doi.org/10.1519/JSC.0000000000003230

• "Countering muscle disuse through resistance training is a powerful intervention to combat the loss of muscle strength and muscle mass, physiological vulnerability, and their debilitating consequences on physical functioning, mobility, independence, chronic disease management, psychological well-being, quality of life, and healthy life expectancy." Provides evidence-based program-design recommendations, including for frailty and sarcopenia.

13. Balance training improves balance in healthy adults 65+ — meta-analysis of 23 RCTs.

Lesinski M et al., Sports Med 2015;45(12):1721-38. PMID 26325622. DOI https://doi.org/10.1007/s40279-015-0375-y

• Balance training improved static steady-state balance (SMD 0.51), dynamic balance (0.44), proactive (1.73) and reactive balance (1.01) and balance test batteries (1.52) vs controls.
• Most effective modalities (modality-specific, indirect comparisons): training period 11–12 weeks, 3 sessions/week, 36–40 total sessions, 31–45 min per session, 91–120 min/week. Study quality limited (mean PEDro 5).

14. Five-year supervised exercise vs guidelines in 70–77-year-olds — Generation 100 RCT, 1,567 participants.

Stensvold D et al., BMJ 2020;371:m3485. PMID 33028588. DOI https://doi.org/10.1136/bmj.m3485

• All-cause mortality did not differ between the control group (following national guidelines) and combined HIIT+MICT. Observed mortality: control 4.7%; HIIT absolute risk reduction 1.7 points (HR 0.63, 0.33–1.20); MICT absolute increase 1.2 points (HR 1.24, 0.73–2.10); HIIT vs MICT HR 0.51 (0.25–1.02).
• Important context: 80% of participants reported medium or high activity at baseline and controls chose to do more of their activity as HIIT — an active, healthy population. Use this to say honestly that a mortality effect of adding structured training on top of an already active life is not proven in a trial.

15. Cochrane review: exercise for preventing falls in community-dwelling older people — 108 RCTs, 23,407 participants (mean age 76).

Sherrington C et al., Cochrane Database Syst Rev 2019;1(1):CD012424. PMID 30703272. DOI https://doi.org/10.1002/14651858.CD012424.pub2

• Exercise (all types) reduces the rate of falls by 23% (rate ratio 0.77, 0.71–0.83; high-certainty) and the number of people who fall by 15% (RR 0.85; high-certainty). Illustrative: 195 fewer falls per 1,000 people per year.
• Balance and functional exercises reduce the rate of falls by 24% (high-certainty); multiple types of exercise (most commonly balance and functional plus resistance exercise) probably reduce falls by 34% (moderate-certainty); Tai Chi may reduce falls by 19%.
• Effect on falls did not differ between trials that selected people at increased risk and those that did not.
• Exercise may reduce fall-related fractures (RR 0.73; low-certainty). Adverse events were predominantly non-serious.
• "We are uncertain of the effects of programmes that are primarily resistance training, or dance or walking programmes on the rate of falls" — the balance component is what the evidence supports for falls.

16. Cochrane review (2026 update): multifactorial and multiple-component interventions for falls — 110 RCTs, 48,919 people.

Swarnalatha C et al., Cochrane Database Syst Rev 2026;9(9):CD012221. PMID 42764178. DOI https://doi.org/10.1002/14651858.CD012221.pub3

• Multifactorial interventions may reduce falls rate (RaR 0.76) but may have little or no effect on the risk of falling; compared with falls advice or education they add no benefit. Multiple-component programs may have little or no effect on falls rate vs usual care. Use this to keep the program's falls claims tied to exercise (ref 15), not to "multi-component" bundles.

17. The Otago home exercise programme — original RCT in women 80+ (233 women, Dunedin GP practices).

Campbell AJ et al., BMJ 1997;315(7115):1065-9. PMID 9366737. DOI https://doi.org/10.1136/bmj.315.7115.1065

• Individually tailored home strength and balance retraining: 88 falls in the exercise group vs 152 in controls over one year; relative hazard for first four falls 0.68 (0.52–0.90); first injurious fall 0.61 (0.39–0.97); balance improved at six months.

18. Home-based strength and balance exercise after a fall — RCT, 345 adults 70+ (mean age 81.6).

Liu-Ambrose T et al., JAMA 2019;321(21):2092-2100. PMID 31162569. DOI https://doi.org/10.1001/jama.2019.5795

• Falls per person-year 1.4 (exercise) vs 2.1 (usual geriatrician care); incidence rate ratio 0.64 (0.46–0.90); no intervention-related adverse events reported.

19. WHO 2020 physical activity guidelines.

Bull FC et al., Br J Sports Med 2020;54(24):1451-1462. PMID 33239350. DOI https://doi.org/10.1136/bjsports-2020-102955

• All adults: 150–300 min/week moderate-intensity or 75–150 min/week vigorous-intensity aerobic activity, or an equivalent combination; regular muscle-strengthening activity for all age groups; reduce sedentary behaviour. "Some physical activity is better than none."
• Gait speed: Perera S et al., J Am Geriatr Soc 2006;54(5):743-9. PMID 16696738. DOI https://doi.org/10.1111/j.1532-5415.2006.00701.x — small meaningful change about 0.05 m/s; substantial change about 0.10 m/s (older adults with mobility disability, stroke survivors, community cohort).
• Chair stand and TUG: Wright AA et al., J Orthop Sports Phys Ther 2011;41(5):319-27. PMID 21335930. DOI https://doi.org/10.2519/jospt.2011.3515 — in 65 patients with hip osteoarthritis after 9 weeks of physiotherapy, major clinically important improvement thresholds: 30-second chair stand +2.0 to +2.6 repetitions; TUG −0.8 to −1.4 seconds; 40-m self-paced walk +0.2 to +0.3 m/s. (Population caveat: hip OA; used here as the best available anchor-based thresholds.)
• Program definition of "improved" (recorded here so the scoreboard is auditable): chair stand ≥ +2 reps; gait speed ≥ +0.10 m/s; TUG ≥ 1.0 s faster; single-leg stance ≥ +5 s (SLS threshold is a program convention, not a published MCID — say so if asked).

21. Senior Fitness Test criterion standards for independence.

Rikli RE, Jones CJ. Gerontologist 2013;53(2):255-67. PMID 22613940. DOI https://doi.org/10.1093/geront/gns071 — criterion-referenced standards for men and women 60–94 associated with remaining physically independent, validated on 2,140 moderate-functioning older adults (reliability/validity indicators 0.79–0.97). The chair-stand cut-offs used by the Movement Lab come from the CDC STEADI norms (already verified in movement-view.js).

• Refs 1–9 are observational: fit, strong, steady people live longer; the trials that prove training changes those measurements are refs 10–13; the trials that prove exercise prevents falls are refs 15, 17, 18. No randomized trial has shown that starting a program at 60 lengthens life (ref 14 is the closest and was null overall).
• Falls evidence is for balance + functional exercise; walking alone or strength alone is uncertain for falls (ref 15).
• Grip strength is a marker; training grip itself has not been shown to change mortality (ref 3).
• Thresholds in ref 20 come from specific populations; they are practical anchors, not laws.