I perceive a weakness in GiveWell and EA in general that it seems to prioritise saving lives over improving quality of life. The standard Disability Adjusted Life Year measure often cited is a pretty blunt instrument. To take one example, if you give someone three years of life but you know they will live it in "overwhelming, constant sadness [to the point they] cannot function in daily life" that's still "equivalent" to giving them one more year of full functional life using the DALY. The disability adjustment weight can reduce the value of an added period of life but not to zero (or to a negative value).
If, to use another example, you save the life of a child but leave it dependent on daily intensive care so that its family is reduced to bare subsistence, this is also considered a net benefit. (I note that hunger or lack of shelter, not being diseases, are not factored into DALY calculations).
It seems that there are many key interventions (improving access to pain relief, preventing or curing Trachoma/river blindness, preventing tertiary syphilis) which may not be costly, where not many lives are saved or lengthened but life quality is greatly enhanced.
The idea of a disability adjustment weight seems broadly useful - I assume it's carefully measured using some kind of rigorous evidence - but the "utility" numbers themselves that result from their application are questionable.
Can I suggest it might be useful to attempt to take these adjustment weights and re-interpret them (perhaps allowing for negative values in some cases?), supplementing them with other quality of life metrics and use the resulting numbers as more nuanced but still evidence-based means of prioritising resource allocation? Perhaps give donors and others a means to adjust their own preferences before calculating the best result?
My apologies if this is an issue which has already been grappled with in detail - in that case I would be grateful if someone could point me to the relevant literature on this point.
I agree that this seems important.
If I remember/understand correctly, the normal instruments fail to deliver useful answers for very bad conditions. For example, if you administer a survey asking how many years of healthy life the survey-taker thinks a year where they suffer X is worth, very bad situations generate incredibly broad answers.
Some people say those years are valueless (so just at 0), some say they have huge disvalue (so they'd rather die now than face one year with the condition and then the rest of their life in good health), and some say that it's close to the value of a year of healthy life (not so sure about this one; I think I remember someone saying it, but it was just in conversation).
As far as discussion on it, I found this GiveWell post that glances on it, and health economist Paul Dolan saying
I didn't read around enough to see if he offers a justification for it.
One small note for DALYs as they currently are: if the major impacts of impairment are essentially flow-through effects of economic value (so if I'm bed-bound my whole life, that's bad for me, but the economic slowdown has overall worse effects), then it may be plausible that DALYs shouldn't go below zero. Your post mentions the case of the child, where it seems to go below zero. But it could be a practical point that keeping the measure at zero results in better estimations of the scale of economic impact than it going below zero.
I’m working with Paul Dolan on a report related to this topic, and the need to value states worse than dead (SWD) is our only major point of disagreement. He gives some kind of justification for his views on p26 of this report, but I find it extremely unconvincing.
But to be fair, nobody has proposed a particularly good method for dealing with SWD. Most attempts have used versions of the time trade-off, with limited success – for a slightly outdated review, see Tilling et al., 2012.
In this Facebook post I’ve listed a few options for achieving the related but more modest objective of determining the point in happiness scales that is equivalent to death.
Thanks for the additional readings. I think Paul Dolan is asking the right questions. I am disappointed that after a promising initial discussion eight years ago, Holden doesn't seem to have spoken again on the subject and to the best of my knowledge there is still no way on GiveWell to put different weights on "impact" to give different results.
I don't understand your last paragraph though. DALYs don't seem to measure economic effects on others at all, so if you do start to consider them wouldn't that be a big argument to make some DALYs negative?
Sorry, I didn't communicate what I meant well there.
It might be the case that DALYs somewhat faithfully track both (a) the impact of conditions on subjective wellbeing and (b) the impact of conditions on economic contribution, even if they're not explicitly intended to track (b). It might also be the case that efforts to extend DALYs to more faithfully track (a) for things that are worse than death would mean that they tracked (b) less well in those cases.
Then, it could be the case that it's better to stick with the current way of doing things.
I don't actually think the above is particularly likely (and yet less so after writing it out) and even in the case that it captures something correct for some moral frameworks, it probably looks different under others.