This paper argues for an alternative methodology to estimate the value of risk to life. By relaxing the assumption of additive separability, we introduce risk aversion with respect to the length of life and show that the extended model better fits available data. This is crucial for the extrapolation stage that the evaluation of life-saving programs systematically requires. Current practice, we show, puts too little weight on the young. Our correction surpasses in magnitude that introduced by the switch from the notion of number of lives saved to the notion of years of life saved.