Thread by @whippletom

Demographics of the South African study into Oxford vaccine

Criteria for mild or moderate cases

Neutralization data - immune responses "similar if not better"

Efficacy, pre November.

South African scientists explicitly say that immune pressure, due to high levels of immunity, caused it to favour escape mutations.

42 cases in analysis

Significant drop in antibodies.

And here are the results at last: 19 cases in the vaccine group, 23 in placebo.

Raw data: 22 per cent efficacy, but not significant.

Note: this is with two doses.

Two thirds mild illness, one third moderate. None serious.

Past infection *does not protect* against variant.

"Data fairly compelling will not protect against mild covid. But will be of some use against severe covid"

"We believe that t cell responses will remain intact...."

"These results are a reality check. We've been somewhat euphoric about efficacy. It is time for us to recalibrate our expectations"