> I'd much rather see a calculation of how many years of life are lost, an actually meaningful metric. How do you calculate the number of years lost? Do you su

> I'd much rather see a calculation of how many years of life are lost, an actually meaningful metric. How do you calculate the number of years lost? Do you subtract their age from the national age expectancy or do you run a prediction model to extrapolate the age each individual would have lived to given what data is available for them?