Methodology

Overview

Longevity Path is an evidence-based estimator of biological age, life expectancy and modifiable longevity potential. It is built on three pillars: actuarial life tables from the World Health Organization, peer-reviewed risk equations published in the cardiovascular and aging-biology literature, and age- and sex-adjusted reference distributions for objective health markers (VO2 Max, heart-rate variability, resting heart rate, body composition, blood biomarkers). The model is not a clinical diagnostic — it is a structured way to translate well-established mortality and morbidity findings into a personal number you can act on.

Every score produced by the platform — your longevity score (0 to 10), your biological age in years, your life expectancy in years and your "years within reach" delta — is computed on the backend in a single deterministic pipeline. The frontend does not run any scoring logic locally; it only displays what the backend calculates, which means results are reproducible across the web, iOS and Android clients and across re-renders.

The five-dimension model

We organise the inputs into five health dimensions, each weighted by the strength of its association with all-cause mortality in the published literature. The weights below sum to 100 percent of the longevity score:

• Cardiovascular — 30%. Cardiorespiratory fitness (VO2 Max), resting heart rate, heart rate variability, blood pressure and exercise volume. VO2 Max is the single strongest known predictor of all-cause mortality in healthy adults (Mandsager et al., JAMA Network Open, 2018).
• Metabolic — 25%. Body composition (BMI and body-fat percentage), fasting glucose and HbA1c, lipid profile (LDL, HDL, triglycerides, ApoB) and dietary pattern. Metabolic syndrome roughly doubles cardiovascular mortality risk.
• Sleep — 20%. Total sleep duration, sleep regularity, deep and REM proportions and morning HRV as a recovery marker. Both short (under 6 hours) and long (over 9 hours) sleep are independently associated with elevated all-cause mortality (Cappuccio et al., Sleep, 2010).
• Cognitive & Emotional — 15%. Chronic stress, social connection density, sense of purpose, and substance use (smoking, alcohol). Social isolation has been estimated to carry mortality risk comparable to smoking 15 cigarettes per day (Holt-Lunstad et al., PLoS Medicine, 2010).
• Genetics & History — 10%. Family history of cardiovascular disease, cancer and diabetes, plus the user's own chronic-condition history. Genetics account for roughly 20 to 30 percent of lifespan variance; we weight this dimension lower than the modifiable ones to reflect that individual habits dominate outcomes.

Within each dimension, inputs are organised into 14 sub-categories (Exercise, Functional Fitness, Nutrition, Body Composition, Biomarkers, Sleep Quality, Stress Management, Social Bonds, Cognitive Health, Substance Use, Mental Health, Family Genetics, and Chronic Disease History) so that a strong score in one area cannot fully compensate for a weakness in another.

Inputs

How biological age is computed

Biological age is the modelled functional age of the user's body, derived from how each input compares to age- and sex-adjusted reference distributions. The reference tables we use cover VO2 Max, heart-rate variability, resting heart rate, body-mass index, body-fat percentage and key blood biomarkers, calibrated to published population data (cardiopulmonary fitness norms from the Fitness Registry and the Importance of Exercise National Database; HRV norms from Nunan et al., Pacing and Clinical Electrophysiology, 2010; lipid and glucose reference ranges from the National Cholesterol Education Program and ADA criteria).

For each input, the user's value is converted to a z-score against the population norm for their age and sex. The z-scores are weighted by the strength of the input's association with mortality, summed within each dimension, and combined into a single biological-age estimate. A user whose cardiorespiratory fitness and metabolic markers consistently outperform their chronological-age peers will receive a biological age below their chronological age, with the magnitude of the gap proportional to how far above average they are across dimensions.

How life expectancy and years within reach are computed

Life expectancy is computed in two steps. First, we start with the baseline life expectancy from the WHO Global Health Estimates life table for the user's country and sex. Second, we apply a series of additive adjustments derived from established risk equations:

• Smoking status uses the Doll et al. cohort findings (BMJ, 2004) — current smokers lose roughly 10 years of life expectancy versus never-smokers; former smokers recover most of that gain depending on quit duration.
• Body-mass index adjustments follow the Global BMI Mortality Collaboration meta-analysis (The Lancet, 2016), which quantifies hazard ratios for each BMI band.
• Physical activity and fitness draw on the Mandsager et al. cohort (JAMA Network Open, 2018) — moving from "below average" to "elite" cardiorespiratory fitness is associated with a hazard ratio of roughly 0.2 for all-cause mortality.
• Diet quality follows the Mediterranean-diet trials (PREDIMED, NEJM 2018) and the lifestyle-factor analysis by Li et al. (Circulation, 2018), which showed that adopting five healthy habits added up to 14 years of life expectancy at age 50.
• Sleep duration follows the U-shaped curve from Cappuccio et al. (Sleep, 2010), with optimum durations of 7–9 hours.
• Alcohol intake applies the Wood et al. dose-response analysis (The Lancet, 2018) — the lowest mortality risk is at zero to modest intake, with risk rising progressively above 100 grams of pure alcohol per week.
• Social connection applies the Holt-Lunstad et al. meta-analysis (PLoS Medicine, 2010).
• Family and chronic history apply established prevention-guideline adjustments (ACC/AHA 2019 primary-prevention guidelines for cardiovascular disease; ADA 2024 standards for diabetes).

The "years within reach" delta is the gap between your current life-expectancy estimate and the estimate you would receive if you optimised every modifiable factor to its best-case value (capped at realistic upper bounds — for example, we do not credit moving cardiorespiratory fitness above the 95th percentile). It is split into a conservative minimum and an aspirational maximum to convey uncertainty.

Calibration

Reference data are stratified by 5-year age band and biological sex. Country is used to select the correct baseline life expectancy from the WHO table and to localise dietary recommendations, but country does not directly modify the biomarker reference distributions, which we treat as global. Where published norms exist only for specific populations (for example, VO2 Max norms from primarily North-American cohorts) we use them as the global default and flag the limitation here for transparency.

Wearable averages use a rolling 90-day window rather than 30, on evidence that shorter windows are dominated by week-to-week variation in training and stress, while 90 days captures the slower physiologic baseline relevant to longevity.

Score interpretation

Longevity score (0–10): 8 or higher is excellent (top quintile of healthy lifestyle and biomarkers); 6.5 to 8 is good with room to optimise; 5 to 6.5 is fair, with several modifiable factors to address; below 5 indicates higher-than-average mortality risk that warrants a structured improvement plan and, where relevant, a clinical consultation.

Biological age delta: a difference of −3 or more years (biological younger than chronological) is meaningfully favourable; ±2 years is roughly on par with your peers; +3 or more years suggests an elevated functional-age burden worth addressing.

Years within reach: this is the potential gain from optimising your lifestyle, not a guarantee. It assumes sustained adherence to the recommended habit changes over years, which is the relevant time scale for the underlying evidence.

What this is not

Longevity Path is not a medical device, not a diagnostic service and not a substitute for licensed clinical evaluation. The model cannot predict any individual's actual lifespan — the inputs available to a consumer tool do not capture the full set of biological, environmental and stochastic factors that determine an individual outcome. What the model can do well is quantify the mortality impact of your modifiable habits relative to peers, in a way that has been validated against population-level cohort data.

If you have a diagnosed condition, abnormal lab values, or any concerning symptom, the right next step is a clinician — not a score. Use the result as a structured starting point for a conversation about your own data, not as an answer in itself.

References

The methodology summarised above draws on the following peer-reviewed sources. Each is publicly accessible via PubMed or the publisher's site.

• World Health Organization. Global Health Estimates: life tables by country, year and sex.
• Li Y, Pan A, Wang DD, et al. Impact of healthy lifestyle factors on life expectancies in the US population. Circulation, 2018;138:345-355.
• Mandsager K, Harb S, Cremer P, et al. Association of cardiorespiratory fitness with long-term mortality among adults undergoing exercise treadmill testing. JAMA Network Open, 2018;1(6):e183605.
• Doll R, Peto R, Boreham J, Sutherland I. Mortality in relation to smoking: 50 years' observations on male British doctors. BMJ, 2004;328:1519.
• Global BMI Mortality Collaboration. Body-mass index and all-cause mortality: individual-participant-data meta-analysis of 239 prospective studies. The Lancet, 2016;388:776-786.
• Cappuccio FP, D'Elia L, Strazzullo P, Miller MA. Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies. Sleep, 2010;33(5):585-592.
• Holt-Lunstad J, Smith TB, Layton JB. Social relationships and mortality risk: a meta-analytic review. PLoS Medicine, 2010;7(7):e1000316.
• Wood AM, Kaptoge S, Butterworth AS, et al. Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599,912 current drinkers in 83 prospective studies. The Lancet, 2018;391:1513-1523.
• Estruch R, Ros E, Salas-Salvadó J, et al. Primary prevention of cardiovascular disease with a Mediterranean diet supplemented with extra-virgin olive oil or nuts. New England Journal of Medicine, 2018;378:e34 (PREDIMED).
• Nunan D, Sandercock GR, Brodie DA. A quantitative systematic review of normal values for short-term heart rate variability in healthy adults. Pacing and Clinical Electrophysiology, 2010;33(11):1407-1417.
• D'Agostino RB, Vasan RS, Pencina MJ, et al. General cardiovascular risk profile for use in primary care: the Framingham Heart Study. Circulation, 2008;117:743-753.
• Arnett DK, Blumenthal RS, Albert MA, et al. 2019 ACC/AHA guideline on the primary prevention of cardiovascular disease. Journal of the American College of Cardiology, 2019;74(10):e177-e232.