Live data from Hacker News

Estimation of total mortality due to COVID-19

healthdata.org

21–30 of 387 posts

Re: Estimation of total mortality due to COVID-19

#21
Third, we build a statistical model that predicts the weekly ratio of total COVID-19 deaths to reported COVID-19 deaths based on covariates and spatial effects.

Given how disastrously wrong most modeling/forecasting(1) has been during this pandemic color me a little bit skeptical that their model is somehow better. Models give wide latitude to fiddle with parameters.

(1) https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7447267/

Re: Estimation of total mortality due to COVID-19

#22

This is anecdotal at best, but when my grandfather passed last May from COVID, his death certificate did not record the cause of death as COVID. He tested negative twice after his month long bought with it, but only lived for another few weeks until dying of respiratory failure. We were told that because he was not testing positive at the time of death, his cause of death could not be recorded as COVID. I thought cau…

Death certificates can record both a primary cause and contributing factors. It's often a judgment call on what to consider as the cause and how far back in the chain of causality to go. There's no statistical manipulation there, just a lack of consistent standards.

Re: Estimation of total mortality due to COVID-19

#23

In absolute number terms it's more deaths in total in the USA than the Spanish Flu (675k deaths total since it began), HIV (700k deaths total since it began) or any other pandemic. Population has increased so the US did lose a smaller percentage of population but on the other hand Spanish Flu didn't have a vaccine to stop it one year in. Hopefully enough of the population takes the vaccine now to stop this. Note this…

It was #3 in 2020 even though widespread testing only started in March. Seriously frightening stuff. Imagine if we went from zero heart disease or zero cancer deaths to present numbers in the span of one year... https://www.cdc.gov/mmwr/volumes/70/wr/mm7014e1.htm

Heart disease and cancer aren't even singular diseases.

Hell, even the Flu is a combination of diseases: TypeA, TypeC, etc. etc. Even Flu-TypeA is a combination of different diseases: H3N2 (Bird Flu) and H1N1 (Swine Flu) are both Flu-TypeA.

For a singular disease + minor mutations (COVID19 vs B.1.1.7 vs P.1 extremely similar and covered by the same vaccine) to cause such a death count is pretty remarkable.

Re: Estimation of total mortality due to COVID-19

#24

This is anecdotal at best, but when my grandfather passed last May from COVID, his death certificate did not record the cause of death as COVID. He tested negative twice after his month long bought with it, but only lived for another few weeks until dying of respiratory failure. We were told that because he was not testing positive at the time of death, his cause of death could not be recorded as COVID. I thought cau…

More generally, there is no universally agreed way to record all the factors that contributed to a death at the time of death or certification.

This is why excess deaths is the real metric by which the toll of this pandemic will be measured, at least in countries with reliable records about whether deaths happened at all.

Re: Estimation of total mortality due to COVID-19

#25

It looks like Florida is two shades "better" than California in the diagram, which would put the total Florida death rate as better than California. That would be interesting because Florida did almost no lockdown. How could that be?

WHO has been consistent about being anti-lockdown during all of Covid.

There's limited evidence they work more than a short-term delay tactic and cause tremendous harm.

Re: Estimation of total mortality due to COVID-19

#26

It looks like Florida is two shades "better" than California in the diagram, which would put the total Florida death rate as better than California. That would be interesting because Florida did almost no lockdown. How could that be?

That's what's confusing. It's really not clear what works and what doesn't. Germany and India did well until they didn't. Maybe all hell will break loose in Florida soon.

What I don't get that people who don't want lockdowns also oppose masking and other measures. Why not try to keep things open but also strongly support masking just to be sure?

Re: Estimation of total mortality due to COVID-19

#27
post #24

This is anecdotal at best, but when my grandfather passed last May from COVID, his death certificate did not record the cause of death as COVID. He tested negative twice after his month long bought with it, but only lived for another few weeks until dying of respiratory failure. We were told that because he was not testing positive at the time of death, his cause of death could not be recorded as COVID. I thought cau…

More generally, there is no universally agreed way to record all the factors that contributed to a death at the time of death or certification. This is why excess deaths is the real metric by which the toll of this pandemic will be measured, at least in countries with reliable records about whether deaths happened at all.

Excess deaths doesn't work either. Gun violence has increased everywhere. So have suicides in my region of the world. I don't know who or why this comment got flagged, it doesn't change reality

Re: Estimation of total mortality due to COVID-19

#28

It looks like Florida is two shades "better" than California in the diagram, which would put the total Florida death rate as better than California. That would be interesting because Florida did almost no lockdown. How could that be?

One of the people I know who died, did so after traveling to Florida where they did catch covid, return home and died 40 some days later on a ventilator.

Maybe travel in and out of Florida is fluid enough to explain it to some degree.

Re: Estimation of total mortality due to COVID-19

#29

In absolute number terms it's more deaths in total in the USA than the Spanish Flu (675k deaths total since it began), HIV (700k deaths total since it began) or any other pandemic. Population has increased so the US did lose a smaller percentage of population but on the other hand Spanish Flu didn't have a vaccine to stop it one year in. Hopefully enough of the population takes the vaccine now to stop this. Note this…

I had no idea about HIV deaths, I would have guessed it would be 2x-3x COVID deaths. I don't think that means "HIV had a smaller impact than I thought", I think that means COVID is bigger than I thought. We get desensitized to the numbers. Thanks for the perspective.

Re: Estimation of total mortality due to COVID-19

#30

It looks like Florida is two shades "better" than California in the diagram, which would put the total Florida death rate as better than California. That would be interesting because Florida did almost no lockdown. How could that be?

There are two images you could be referring to.

Figure 5: The (estimated) ratio of reported versus total deaths. It appears from it that FL was more accurate in their reported numbers than CA. That has absolutely nothing to do with lockdowns.

Figure 7: The cumulative deaths, where FL and CA are the same color (that is, in the same range of cumulative deaths).

Either way, neither indicates that FL's death rate was better than CA's.

https://www.statista.com/statistics/1109011/coronavirus-covi...

Shows that FL has a higher death rate than CA. Though if you take the adjustment from this paper into consideration, then CA's would be worse.

Post reply on HN