COVID-19 statistics

[Zander]

The statistics for Covid 19 are skewed. The reported "death rate" for Covid 19 is based upon the number of deaths divided by the number of "Confirmed" cases. By contrast, the reported "death rate" for the flu is calculated like this:

How CDC Estimates the Burden of Seasonal Influenza in the U.S. | CDC

Links to key resources on the burden of influenza - CDC

www.cdc.gov

View attachment 316405

Why isn't this method being used for Covid 19?

because they don't have the data of years of flu seasons to provide the factors needed to extrapolate the COVID data to the entire US population.

Then they should stop comparing the "death rates" since they aren't apples to apples.

 

[TroglocratsRdumb]

The statistics for Covid 19 are skewed. The reported "death rate" for Covid 19 is based upon the number of deaths divided by the number of "Confirmed" cases. By contrast, the reported "death rate" for the flu is calculated like this:

How CDC Estimates the Burden of Seasonal Influenza in the U.S. | CDC

Links to key resources on the burden of influenza - CDC

www.cdc.gov

View attachment 316405

Why isn't this method being used for Covid 19?

because they don't have the data of years of flu seasons to provide the factors needed to extrapolate the COVID data to the entire US population.

Sure they do. It’s just not as comprehensive — yet. But it’s getting there. And the more it does the better the outlook.

Not the level of data they have for influenza, which even as it changes year to year is still basically the same virus.

For COVID-19 all you have is the current outbreak, and no one has done any sort of blind sampling for antibodies yet to just determine the viral penetration in the general population.

I don’t think it’s that far behind.

Data manipulation like extrapolation needs large sets from various cycles to be usable. We've had a century of flu seasons in modern times that can be used to clarify the "fudge" factors any model would use to determine nationwide infection/serious case/fatality data using just hospitalization/fatality data.

Sure you can make a model and run in on the limited data we have for COVID-19, but the results would be crap. GIGO applies to attempts at analysis like this.

I do modelling for wastewater treatment, which is more complex than epidemic modelling, but less complex than climate modelling, and GIGO applies no matter what.

The projected models that I have seen say that the virus will peak in early May and be mostly gone by August. But, sometimes there are second and third waves.

 

[martybegan]

The statistics for Covid 19 are skewed. The reported "death rate" for Covid 19 is based upon the number of deaths divided by the number of "Confirmed" cases. By contrast, the reported "death rate" for the flu is calculated like this:

How CDC Estimates the Burden of Seasonal Influenza in the U.S. | CDC

Links to key resources on the burden of influenza - CDC

www.cdc.gov

View attachment 316405

Why isn't this method being used for Covid 19?

because they don't have the data of years of flu seasons to provide the factors needed to extrapolate the COVID data to the entire US population.

Then they should stop comparing the "death rates" since they aren't apples to apples.

agreed. but since it's all they got, that's what they run with.

 

[martybegan]

The statistics for Covid 19 are skewed. The reported "death rate" for Covid 19 is based upon the number of deaths divided by the number of "Confirmed" cases. By contrast, the reported "death rate" for the flu is calculated like this:

How CDC Estimates the Burden of Seasonal Influenza in the U.S. | CDC

Links to key resources on the burden of influenza - CDC

www.cdc.gov

View attachment 316405

Why isn't this method being used for Covid 19?

because they don't have the data of years of flu seasons to provide the factors needed to extrapolate the COVID data to the entire US population.

Sure they do. It’s just not as comprehensive — yet. But it’s getting there. And the more it does the better the outlook.

Not the level of data they have for influenza, which even as it changes year to year is still basically the same virus.

For COVID-19 all you have is the current outbreak, and no one has done any sort of blind sampling for antibodies yet to just determine the viral penetration in the general population.

I don’t think it’s that far behind.

Data manipulation like extrapolation needs large sets from various cycles to be usable. We've had a century of flu seasons in modern times that can be used to clarify the "fudge" factors any model would use to determine nationwide infection/serious case/fatality data using just hospitalization/fatality data.

Sure you can make a model and run in on the limited data we have for COVID-19, but the results would be crap. GIGO applies to attempts at analysis like this.

I do modelling for wastewater treatment, which is more complex than epidemic modelling, but less complex than climate modelling, and GIGO applies no matter what.

The projected models that I have seen say that the virus will peak in early May and be mostly gone by August. But, sometimes there are second and third waves.

It depends on if it mutates to become a seasonal issue, or if it's a one shot bug that everyone will eventually have immunity to, and thus it won't have a reservoir to infect people in the future.