From the paper (DEEX - "Depressed mood and exhaustion"): "Nonetheless, adding DEEX to a risk score based on classical risk factors resulted in only nonsignific

From the paper (DEEX - "Depressed mood and exhaustion"): "Nonetheless, adding DEEX to a risk score based on classical risk factors resulted in only nonsignificant improvement of mortality risk prediction." Regarding the causal relationship: "Although not a proof of causality, in general, depressed subjects are more likely to cluster self-harming lifestyle behaviors and may be less likely to adhere to prescribed medication." To put things in context it is also worth noting that, except in old age, a big increase in the risk of death at any given point of time translates into a very small reduction in life expectancy. See for example http://joshmitteldorf.scienceblog.com/2012/11/10/mortality-and-life-expectancy/ http://joshmitteldorf.scienceblog.com/2012/11/10/mortality-a...