Which interventions actually save lives and reduce suffering?
Causal inference on thousands of country-years of policy, budget, and outcome data to identify what maximizes healthy life years and minimizes preventable death and disease.
The Two Metrics That Matter
Every policy, budget, and intervention is scored against two measures of human wellbeing — not GDP, not stock prices, not averages that hide suffering.
Median Healthy Life Years (HALE)
Not just life expectancy — healthy years free of disability and disease. A country where people live to 80 but spend the last 15 years suffering scores lower than one where people live to 78 in good health.
Why median? Because mean life expectancy is dragged down by infant mortality, masking how well a society actually prevents suffering across the lifespan.
Median Real After-Tax Income
Poverty kills. This measures what a typical person can actually afford after taxes and inflation — not GDP, which hides inequality. A country with 10 billionaires and mass poverty looks great on GDP but fails its people.
Why median? Because mean income is skewed by billionaires. The median tells you whether ordinary people can afford healthcare, nutrition, and shelter.
The Data
Across healthcare, drug policy, criminal justice, climate, education, and infrastructure — sourced from OECD, World Bank, WHO, and peer-reviewed studies.
What the Data Reveals
Striking findings from real outcome data across jurisdictions.
83.9 yrs life expectancy at 4.1% GDP
vs US: 77.5 yrs at 17.3% GDP
Singapore's 3M system (Medisave, MediShield, Medifund) achieves world-leading health outcomes at a fraction of US spending through universal coverage with market competition.
View Analysis →Drug deaths fell 70% in Portugal
HIV among users fell 74%
Portugal decriminalized personal possession of all drugs in 2001 and shifted resources to treatment. Drug-induced deaths dropped from 52 to 3 per million population.
View Analysis →Norway recidivism: 20%
vs US: 76%
Norway's rehabilitative prison system — education, vocational training, max 21-year sentences — cut re-offending from 35% to 20% while keeping incarceration rates low.
View Analysis →Rwanda: life expectancy 48 → 69 yrs
Under-5 mortality fell 82%
Rwanda deployed 45,000 community health workers in 2005. Under-5 mortality dropped from 196 to 35 per 1,000 live births over 15 years.
View Analysis →Natural Experiments
When a jurisdiction changes a policy, it creates a natural experiment. We track outcome metrics before and after the intervention with interrupted time-series analysis. This is not theory — it's observed data.
How It Works
Collect
Outcome data from OECD, World Bank, and WHO across jurisdictions and decades
Align
Policy changes paired with outcome trajectories using temporal analysis and onset delays
Score
Causal strength via Bradford Hill criteria — strength, consistency, temporality, gradient
Identify
Optimal funding levels and policy configurations with confidence intervals
Recommend
Evidence-based policies ranked by Predictor Impact Score with effect size estimates
Stop Guessing. Start Saving Lives.
Every analysis is backed by real outcome data from official sources. Whether you're a nonprofit allocating grants, a government setting policy, or a researcher studying what works — the data is open and the methodology is transparent.
Same Engine, Every Scale
The causal inference engine is domain-agnostic. Feed any two time series and it answers: does changing X cause Y to change? By how much?
Nonprofits & NGOs
Find which interventions reduce the most suffering per dollar. Compare approaches across countries and decades of outcome data.
Governments
Optimal policies and budget allocation across jurisdictions. Any city, county, state, or country.
Individuals
Import your health data from wearables, supplements, and habits. Find what works for you via local causal analysis.