Alexandre,   Statistics are probabalistic statements about populations not specific individuals. It would be a mistake to say that because x% of a certain population is likely to die then this specific person has X% of dying. There is no one-to-one mapping. Also, life expectancie tables are  (loosely defined) "models" with certain assumptions and not "provable" per se. For instance, an assumption of the model might be that a certain percentage of people in the population might die from a certain disease. Perhaps there will be a cure for that particular disease causing the death rate to go down. In this case the model will be less accurate than it might have been and it's estimate will have to be revised or abandoned. Most importantly statistical life expectancy tables do not account for individual experience.  How many of the people that died prematurely could have been helped by a skilled NLP practitioner who was able to change their beliefs (and reality) about living a long healthy life? Perhaps, it would be wiser if we, as a society, focused our efforts on modeling healthy, happy people who live long-productive lives? How strongly would that lengthen the survival curve? I'm sure that there are NLP'ers out there modeling health who will reply and share resources with you. Ryan