Methodology
A mirror that does the math out loud
Biome runs your inputs against published epidemiological models — all-cause mortality risk modifiers and condition-specific risk factors from peer-reviewed cohort studies — and shows the modeled trajectory of your current environment.
The frame is honest: “Here's how the math runs, given what you've told me.” Not “you will.”
Data sources
- Framingham Heart Study — cardiovascular risk
- UK Biobank — sleep, mortality, lifestyle modifiers
- NHANES — population health benchmarks
- Nurses' Health Study — dietary and lifestyle outcomes
- ARIC — atherosclerosis risk
Actuarial modifiers
Every coefficient used in life expectancy modeling, with source:
| Factor | Modifier | Source |
|---|---|---|
| Current smoking | ~−10 years | Jha et al., NEJM, 2013 |
| Former smoking | ~−3 years | Jha et al., NEJM, 2013 |
| Severe insomnia (<6 hrs) | ~−3 years | Li et al., UK Biobank, 2021 |
| Optimal sleep (7–8 hrs) | ~+1 year | Li et al., UK Biobank, 2021 |
| Regular exercise (5+ days/wk) | ~+3 years | Ekelund et al., BJSM, 2019 |
| Moderate exercise (3–4 days/wk) | ~+2 years | Ekelund et al., BJSM, 2019 |
| Sedentary lifestyle | ~−2 years | Ekelund et al., BJSM, 2019 |
| High perceived stress (PSS >14) | ~−2 years | Cohen et al., 1983 |
| Social isolation | ~−3 years | Rico-Uribe et al., 2018 |
| Strong sense of purpose | ~+2 years | Alimujiang et al., JAMA, 2019 |
System assessment model
Each of the eight systems receives a state estimate (thriving, stable, strained, compromised, or unknown) based on relevant questionnaire inputs. Confidence increases with more completed sections. Cross-system effects are modeled explicitly: stress affects sleep, sleep affects gut, gut affects immune, etc.
Sensitivity sliders apply published modifier coefficients to show how changing a single input affects the modeled trajectory. Uncertainty bands widen over longer time horizons.
If Biome cannot show its math, it does not show a number.