That oft-cited paper out of Israel isn't yet published, and seems to have some issues on (informal) review: https://twitter.com/DGBassani/status/143139018892476

That oft-cited paper out of Israel isn't yet published, and seems to have some issues on (informal) review: https://twitter.com/DGBassani/status/1431390188924768269 https://twitter.com/DGBassani/status/1431390188924768269 Or from the article you just linked: > As for the Israel medical records study, Topol and others point out several limitations, such as the inherent weakness of a retrospective analysis compared with a prospective study that regularly tests all participants as it tracks new infections, symptomatic infections, hospitalizations, and deaths going forward in time. “It will be important to see these findings replicated or refuted,” says Natalie Dean, a biostatistician at Emory University. She adds: “The biggest limitation in the study is that testing [for SARS-CoV-2 infection] is still a voluntary thing—it’s not part of the study design.” That means, she says, that comparisons could be confounded if, for example, previously infected people who developed mild symptoms were less likely to get tested than vaccinated people, perhaps because they think they are immune.