I feel a single "deaths" number is not that useful of a metric, and something like "life years lost" would be better. A healthy 20 year old would lose ~60 years

I feel a single "deaths" number is not that useful of a metric, and something like "life years lost" would be better. A healthy 20 year old would lose ~60 years; a diabetic obese 60-year old much less. Of course we should take reasonable efforts to protect those more vulnerable to disease, but with more nuanced data like this you can better decide what "reasonable" is.