Thread by @Peoples_Pundit

2. You can't compare mortality rates for the Chinese Coronavirus in South Korea to U.S. seasonal flu in the current season.

1) Healthcare systems are not comparable.

2) Burden estimates are adjusted significantly post-season.

Here are burden estimates by age for flu 2018-2019.

3. Why are burden estimates — mortality rates, hospitalizations, etc. — adjusted significantly post-season?

Few reasons depending on the season we're talking about.

Re: mortality rates, CDC relied on death certificate data during the 2018–2019. But...

4. But until more complete death certificate data were available, the CDC used average frequencies of each cause from previous seasons.

Also, testing data from participating FluSurv-NET sites are often not available for up to two years after the end of a season, thus revised.

5. Further, CDC burden estimates for 18-19 season were derived from rates of influenza-associated hospitalization, a newer method than statistical models used in older published reports.

It also makes certain years difficult or flatly unable to compare to one another.

6. For outside hospitals, CDC model "uses the frequency of influenza-related deaths that have cause of death related to pneumonia or influenza (P&I), other respiratory or cardiovascular (other R&C), or other non-respiratory, non-cardiovascular (non-R&C) to account for deaths."

7. Point being, this is all very convoluted and fluid over a season, and even beyond.

That's kind of my point and why I'd caution newfound Twitter experts on mortality rates — in and out of media — to talk to a real experts before making definitive claims that mislead people.