Thread by @hildabast

J&J ENSEMBLE data is here - what. a. day.

It works with a single, simple to distribute, shot, which is huge. And it's complicated.

Largest set of events yet: 468 people with symptomatic Covid-19 in 43,783

...1/n jnj.com/johnson-johnso…

...34% were over 60 (14,672 people). Diverse, from South Africa & the Americas. 41% had co-morbidities putting them at risk of severe Covid-19. Protection was consistent across age, race, & part of the world - even though 95% in SA were infected with the new "SA" variant...2/n

...First trial powered to answer the severe Covid-19 question: efficacy was 85% at 28 days after the injection (no event breakdown or confidence intervals). For moderate to severe, efficacy was 66%, varying a bit in the 3 regions (from 57% in SA to 72% in the US...3/n

...Efficacy against severe disease increased over time - all the severely ill people were in the first 49 days. There were 0 people hospitalized, 0 deaths. But they don't report the mild/moderate/severe breakdown of the 468 illnesses...4/n

...This is clearly a "quick fix" for the worst that this disease does, to people & to healthcare systems & that's fabulous & so badly needed. But the absence of the data the others reported raises questions. Good they have 2-dose trial results coming soon...5/n

...On safety & adverse events, the only specific data is for fever: only 0.2% rate for grade 3 (moderate) (presumably 0 severe). And they report no anaphylaxis (allergic reaction)...6/n

...How does it compare to others? Tricky when they're such different trials & this one is particularly different: it's even harder without more detailed data. Single dose may cost some efficacy, but with such a huge advantage, too. Hard to tell. 7/n

...Another important piece of data on J&J vaccine in an NIH press release: 5 deaths from Covid-19 in this trial, all in the placebo group nih.gov/news-events/ne…

Between the other 2 big trials - Tozinameran (BNT/Pfizer) & Moderna - there was 1 Covid death (placebo gp).