> True statement but it should be the other way around. Deaths by overdoses are much more impactful in terms of life expectancy decline, since they mostly kill

> True statement but it should be the other way around. Deaths by overdoses are much more impactful in terms of life expectancy decline, since they mostly kill younger people. This, of course, actually depends on the numbers from both (both ages and volumes). The graph has a very clear inflection point from 2019 to 2020, overdoses were increasing for quite a while before that, so do you actually have numbers suggesting that the 2020 and 2021 impact of overdoses and accidents is greater than the completely-new-to-those-years factor?