Major strides in forecasting future health

Around 2008, I asked demography colleagues: “Surely we could better forecast future mortality rates if we first forecast future risk factor trends—like smoking rates?” Their answer then was no—simply using historical mortality rates to forecast future mortality rates was better than any attempt to build in forecast trends for smoking, blood pressure, and other risk factors that are in turn all mathematically linked to likely mortality rates. That was then. Now in The Lancet, Kyle Foreman and colleagues1 forecast mortality rates and years of life lost for 250 causes of death to the year 2040.