There is little excuse for “as reported” in a graph – they can easily just graph when the deaths occurred.
I can understand on the news when they need to say numbers “300 new deaths reported over the last two weeks”, but it is pointless in a graph. It is misleading, as in this case.
But wouldn’t such a graph presumably show a declining death rate over, say, the last 2 weeks because those deaths had not yet been reported into the system. And it might mislead people into thinking things were getting better when they were in fact getting worse.
I believe florida does this with its covid tracker. It reports cases by when the infection occurred. Reporting delays means the case count always looks like it’s decreasing.
Exactly. That’s why i always look at the report-date graphs when deciding stuff like, “how comfortable do i feel doing X today?”
When looking at a longer-term trend, like comparing the death rate and infection rate of Omicron to earlier waves, i look at occurrence dates. But i find myself looking at report dates more often.
They need somebody who can work out a factor to multiply by, to estimate how many haven’t yet been reported.
Estimating loss triangles… if only there were a profession where people knew about that sort of thing…
Still thinking about those Costco raviolis eh? PM me your address and I’ll ship you some. 
yes, because I’m using this shit for setting IBNR.
ffs
From the Academy:
With the celerity known of the organization, explaining the distinction between cohort and period life expectancy a year after I kept bitching at people about this. Though I also bitched about this back in 2019, before the pandemic.
Because it’s something that comes up every year in the media.
Well, at least they’ve gotten around to it now.
detecting COVID death undercounts – in countries that will absolutely not surprise you
Of course he’s not bothering with China or the like
Ron DeSantis has been getting around, I see. 
I’m guessing some of the assumptions this author makes may break down for a few of the countries that popped out of this, but the analysis of Russia is really interesting. Shame China wasn’t included.
It’s pointless to include China. They just say 0.
U.S. ‘excess deaths’ during pandemic surpassed 1 million, with covid killing most but other diseases adding to the toll, CDC says
https://www.washingtonpost.com/health/2022/02/15/1-million-excess-deaths-in-pandemic/
I was surprised to see that Arizona and Mississippi had more excess mortality than New York, per capita.
Arizona had some bad COVID mortality.
I will have to do an update on that in a bit. I just pulled the update of numbers yesterday for a graph at work. Maybe I’ll refresh graphs in here this weekend by age, etc.
MS is not too surprising due to higher poverty levels and (presumably) higher comorbidities. Due to the steep age related slope of COVID mortality, I’d love to see normalized population measures, but that is beyond specificity of the pre-digested summaries I follow.
How much AZ is driven by the challenges facing the native american population is also a question.
I havn’t really kept up with the content of this thread. Maybe this topic has been bandied about in my absence.
NY Times writer David Leonhardt wrote today about a subject that I find interesting that has been alluded to but this is the first that I have seen data as a smoking gun. Covid is killing republicans faster than democrats.
He cleverly calls it red covid vs blue covid. Funny ha-ha. Republicans don’t want a vaccine? Guess what, you backward thinking nitwits are gonna die at a higher rate. Hey republicans, you are 1 & 7 in the popular vote in the last 8 presidential elections. How are you going to turn the tide when you stubbornly would rather die than get a vaccine?
I saw this a while ago, probably on Reddit. Is there any explanation other than vaccination rates? Republicans tend to skew rural, but I don’t think population density has been a significant factor (citation needed) and if anything it would help red states. Republicans skew older but the data I found said it was by only a couple of years - might explain a bit of the difference but not all, right?
It would seem that it’s vaccines, but I saw it and was thinking about how to poke holes in the argument.
High population density is correlated with more deaths. They get hit sooner and faster, when there are fewer treatments, with numbers more likely to overwhelm hospitals. That’s why the blue counties were so far ahead at the start of the pandemic. The lines don’t cross until vaccines became widely available.
Maybe bc I live in a whole red state, but lots of every man for himself thinking here. Fewer vaccinated but also shorter or no shutdowns, less adherence to any mitigation strategy whatsoever, more reluctance to believe anything coming from the government.
So yeah, vaccines, but before that a general unwillingness to actually follow practices that might help.
Fair points. I live in Kansas (I am in a little blue island in a sea of red), so I am familiar with these, lol.
