2

ELW comments on Open Thread #39 - Effective Altruism Forum

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (14)

You are viewing a single comment's thread.

Comment author: ELW 26 October 2017 11:05:18AM 1 point [-]

I have a query regarding DALYs which I've been unable to find an answer too, but suspect there is literature on were I more familiar with econ/global health:

By my understanding one of the main advantages of DALYs is that they capture the intuition action in cases like "You may extend either person As life by 5 healthy years or extend person Bs life by 5 mediocre years (lets say they go blind due to the treatment)."

However, there seems to be no way of distinguishing the case where person A and B start of as perfectly healthy and we may help the former more and the cases where B is already blind and we may add "five years at their current state of well-being". This seems to not be ideal.

Is there any talk or use of "marginal DALYs" for want of a better term, where the intervention is considered relative to the previous level of wellbeing? Alternatively, is it simply common practise to use QALYs in the kind of case I am concerned with?

Comment author: Peter_Hurford  (EA Profile) 26 October 2017 02:05:25PM 0 points [-]

However, there seems to be no way of distinguishing the case where person A and B start of as perfectly healthy and we may help the former more

If you improve the number of years lived for a healthy person, that is "straightforward" on the DALY view -- it's +1 DALY for every extra year of life added.

The question of improving the quality of their life is a harder one -- I think the suggestion from the DALY framework is that if the person has perfect health, there isn't any way to improve the quality of their life (because it's already perfect). ...However, we know that's not actually true, because there is no DALY weight for getting tickets to go see Hamilton, while I think that would improve nearly anyone's life. That's just an area where DALY metrics are incomplete, but you could extend the DALY framework that way, by asking people questions like "If you could choose between an free Hamilton tickets but had a 1% chance of death, would you take the tickets?" (I'd probably take the tickets at a 0.005% chance of death.)

-

and the cases where B is already blind and we may add "five years at their current state of well-being". This seems to not be ideal.

This one is also "straightforward" in the DALY view -- you're adding more years at their current disability weight. If I recall correctly, an extra year of life that would otherwise not have been lived, but lived with blindness is worth +0.8 DALY. Thus adding "five years at their current state of well-being" (that is, blindness but no other issues), would be +4 DALY.