@DroidEngineer @grammargirl @clubantietam

Tufekci's one of the biggest COVID minimizers on planet earth. I wouldn't listen to her on anything. Period.

All you have to do is open the actual paper that's been linked here and look at figure 5, as has been pointed out to you, and you'll see quite clearly that all of the data points match perfectly. Just in the paper it's a bar graph and here it's been put into a different form for easier viewing for a non-scientific audience. It's not a "made up graph" in any way other than it's not the graph that was in the paper.

It's not something that's up for debate and too much time's been wasted on this discussion already, frankly.

Now, if you want to discuss what the data itself was discussing and what the limitations were, I think that's a valid discussion, but you'd have to read the paper before we could have that discussion.

EDITED TO ADD - I'm going to pull this back, because I can(yay Mastodon) and say I'm not trying to be aggressive about it, as that did sound when I re-read it, but Tufekci's not a reliable source and I do actually know that not everyone keeps up with the latest COVID news, but this articles been discussed ad nauseam for 4 months now. I generally believe anyone still casting doubt on the overall info to be a bad faith actor, but I'm going to pull back and assume you're not.

What it *does* discuss is exactly what most people would think it discusses, and rather than explain it myself, here's what the paper states quite clearly:

"In this study of 5,819,264 people, including 443,588 people with a first infection, 40,947 people who had reinfection and 5,334,729 noninfected controls, we showed that compared to people with no reinfection, people who had reinfection exhibited increased risks of all-cause mortality, hospitalization and several prespecified outcomes.

...

Compared to noninfected controls, assessment of the cumulative risks of repeat infection showed that the risk and burden of all-cause mortality and the prespecified health outcomes increased in a graded fashion according to the number of infections (that is, risks were lowest in people with one infection, increased in people with two infections and were highest in people with three or more infections). Altogether, the findings show that reinfection further increases risks of all-cause mortality and adverse health outcomes in both the acute and postacute phases of reinfection. The findings highlight the clinical consequences of reinfection and emphasize the importance of preventing reinfection by SARS-CoV-2."

IMO Tufekci simply creates a straw man argument that sicker people will get COVID more often because they have more issues, rather than looking at the fact that millions upon millions of people are on their 3rd infection(at least) already and no matter how you slice it those people have worse health outcomes.

The limitation here is the same as all of the studies done by this group, in that they pull most of their data from the VA so it skews heavily male(although females tend to have worse COVID outcomes) and older. Take it for what you will.