COVID-19 from a different angle
post by jmh · 2020-05-04T17:58:02.100Z · LW · GW · 6 commentsContents
6 comments
Just trying to put things into a larger context for my own thinking here. Not much (if anything) new here. Purely USA centered.
Needless to say SARS-CoV-2 is THE event of 2020 and may well prove to be the event of the century (but a lot can happen in 100 years). At least that is probably what everyone would think at this point.
Probably a lot of wide CI bands here but we'll learn going forward.
What I wanted to look at was some guess at what we might see looking back on 2020 regarding mortality.
Currently the USA seems to have a mortality rate of about 8.34 per 1000.
Statista provide a graphic show daily deaths (7,969.7/day) for 2019 and a few days for COVID-19. I very crude calculation (basically Feb, Mar, Apr + 10 days) gets us about 690 COVID-19 deaths per day -- likely an under estimate as we're probably not peak death now.
So what happens to the expect mortality rate if we assume all COVID-19 deaths would not have occurred in the absence of the virus -- in other words, the person with a cause of death identified as COVIV-19 would have lived rather they having died during the year from some other cause.
Well, using the 690/day guess mortality would increase about 7.9%. Using the number for April 29 provided by Statista that goes up to about 12.6%. These mean the (very crude) expected mortality rate for 2020 would be between 9 and 9.3.
In one sense that doesn't seem like a lot. However, it would also represent the rates for the USA in the early 1970s; basically a 45 year set back. That does seem rather significant.
However, that is the worst case scenario as it assumed non of those COVID-19 deaths would have occurred due to some other cause. This seems a highly questionable assumption given the most at risk population.
I'm not sure how to really assess that question. I do think that some insights here are probably very useful in the context of reopening/relaxing the existing restrictions on social and economic activity.
Has anyone seen studies on funeral data that might offer some insight to allow calculation of the net new deaths resulting from the virus? I would be surprised the data is not there but suspect it is not in any consolidated DB.
So a prediction option for those who perhaps want to offer their thought. How much will the current mortality rate, 8.34 is what I've found but not sure if that is 2018 or 2019, increase due to SARS-CoV-2?
6 comments
Comments sorted by top scores.
comment by Jay Molstad (jay-molstad) · 2020-05-04T21:57:25.940Z · LW(p) · GW(p)
Your baseline mortality rate implies an average life expectancy of 120 years. I'd double-check that source.
Also, COVID-19 can cause permanent lung damage, and possibly damage to other organs, even if people are otherwise asymptomatic. The possibility that many people, now young and with sufficient lung capacity to ignore the damage, may become disabled in 20 years or so is what worries me most.
Replies from: Davidmanheim↑ comment by Davidmanheim · 2020-05-05T08:45:51.221Z · LW(p) · GW(p)
The death rate of per thousand is at least approximately correct, and doesn't imply people live to 120. You can't infer time to death by just dividing, because population is not evenly distributed across ages, partly because of birth cohort sizes, partially because people die as they age, so younger people are always over-represented.
comment by Davidmanheim · 2020-05-04T20:22:21.255Z · LW(p) · GW(p)
You'll probably be interested in Good Judgement COVID-19 Dashboard, which asks "How many people will die in the U.S. in 2020 relative to 2019, according to the Centers for Disease Control and Prevention (CDC)?"
(Especially see the comments.)
Replies from: jmh↑ comment by jmh · 2020-05-04T21:53:58.951Z · LW(p) · GW(p)
Thanks, very helpful.
The CDC data is interesting, unless the numbers are really lagged looks like the number of expected deaths dropped of a cliff. Will be interesting to what that over the next couple of weeks.
Replies from: Davidmanheim↑ comment by Davidmanheim · 2020-05-05T08:40:22.942Z · LW(p) · GW(p)
The last several weeks of data isn't entered yet - it takes time for them to get and enter death certificates.
" *Data during this period are incomplete because of the lag in time between when the death occurred and when the death certificate is completed, submitted to NCHS and processed for reporting purposes. This delay can range from 1 week to 8 weeks or more, depending on the jurisdiction, age, and cause of death. "
Replies from: jmh