How many quality-adjusted life years will "Long Covid" take away from an average adult on the West Coast of the United States of America?

Mini

Ṁ14

Aug 2

ALL

77%Other

1.5%

0.17-0.18 disability-adjusted life years per https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9009167/. Guaranteed to be wrong: Excludes diabetes, includes fatal outcomes (can't have Long Covid if you're dead), Australian population.

20%

0.09 disability-adjusted life years per https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9009167/. Also guaranteed to be wrong for same reasons as 0.17-0.18. This figure accounts for vaccinated status.

1.6%

0.01 - 0.02 disability-adjusted life years per

Provide both a # (or confidence interval + mean) and a citation or explanation for your prediction. The answer with the most convincing evidence will win. If two answers are equally persuasive, I will choose the one with the tighter range. Definitions: - Long Covid: Whatever the Wikipedia page says at the time the question is resolved. If someone meddles with the Wikipedia page at that time, I will use the newest one that has not been screwed up. - Average adult: Younger than ~65 years old. Older than ~18 years old. Exceptional Circumstances: - If a very definitive answer falls out of the sky and no one happens to reference it, I reserve the right to resolve the question based on whoever's closest to that #. - If all the sources are equally muddled or there are reports that we've been thinking about this all wrong, the question may be resolved as N/A. Jun 21, 12:02am: Should've mentioned that I'm interested in outcomes for fully vaccinated people. And by that, I mean "2 shots plus 1 or plus 2, if immunocompromised". In other words, assume that this person has read https://www.lesswrong.com/posts/z8usYeKX7dtTWsEnk/more-dakka. Jun 21, 12:02am: For comparison, https://pubmed.ncbi.nlm.nih.gov/25074692/ says the years lived disabled or injured (YLD) is about 18.4 years for car occupants. Jun 29, 10:30pm: Jack correctly pointed out that this question wasn't very clear. I'm revising this question's definition of "Long COVID" to include all long-lasting impacts of COVID-19. In light of the WHO web page and his question, I would include "Long COVID", "PICS", "Permanent", and "Mortality" from table 5 of https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9009167/ even though "acute" should be at least partially included. I'd have a hard time arguing with someone who spent the last >1 months in an ICU if they say that their "acute" DALY loss was a pretty "long-lasting" chunk of their lifespan if they've only lived for 10 years.

Get

1,000

and

1.00

8 Comments

Sort by:

@ZianChoy When will this resolve? Should it remain open for trading until we know the answer?

@ZianChoy The main point I was trying to clarify was what is the conditional in this question. Is it QALYs lost conditional on being infected by Covid?

@jack Thanks for spotting the problems with the question. I think of Covid as having 3 basic outcomes: you get sick a bit for a while and recover ("acute" in Table 5) or you end up in the hospital and have a bad time ("PICS", "acute", and "fatal"). Then, after you get over it, you might have long-term effects ("Long COVID", "PICS", and "permanent"). You can tell that this is a muddled way to think about it. I'll post a clarification to the question.

@ZianChoy Thanks! So your number is QALYs lost for an average covid case then, right? Want to clarify that because estimating average QALYs lost for someone with long covid, or an average person who may or may not get covid, are other possible interpretations of your question.

https://www.who.int/data/gho/indicator-metadata-registry/imr-details/158 convinced me to also include fatal years lost. Doing so results in a value of 0.09 when using 1 significant figure.

@Duncan I'm afraid your link got eaten. Any chance you could try posting it again? Maybe as a reply to this thread?

@jack First, get an estimated DALY loss for each health state. 1. Copy table 5 into Microsoft Excel. 2. Ignore the existence of "Share of DALY loss for each health state" and the 2020 Actual column. 3. Look at the row labelled "Total DALYs (fatal and non-fatal)" 4. Divide each column's YLD by the # of cases. You will end up with the following numbers: - 0.17 - 0.17 - 0.18 - 0.18 Next, go to the section in the article labelled "Post-acute consequences." You will see an odds ratio of 0.51 for people who have been vaccinated. Multiply 0.18 by 0.51 and chop off a significant figure for no reason at all. You will get 0.09

@ZianChoy Can you explain where in the paper this is from / how you calculated it? I can't find it from a quick skim.

Related questions

Related questions