These average predictors have always bugged me. While it may not be possible to accurately predict the exact cause of my death, the risks for a wealthy retired couple living a healthy lifestyle in a cottage in the rural Switzerland is going to be different than those for a single retiree who smokes, drinks whiskey, and lives in the centre of Beijing. There is no cause of death that is meaningfully captured by averaging a coarse demographic.<p>The by far the most important determinant (more important than even sex and race) is missing here: is your body fat level healthy? The median American is overweight, and we'll soon reach the tipping point when the median American is obese[1]. The huge % of deaths due to circulatory diseases is almost entirely due to this (as is a fraction of the other causes). If you're healthy, the data is too pessimistic. If you're too fat, the data is too optimistic.<p>[1] <a href="http://stateofobesity.org/obesity-rates-trends-overview/" rel="nofollow">http://stateofobesity.org/obesity-rates-trends-overview/</a>