Hacker News Reader

Comment by theferalrobot | Hacker News Reader

Like all things with the oxford vaccine - that is only if you look at one slice of the data (in which there were no elderly)... why they didn’t include elderly is beyond me but of course there aren’t going to be severe cases in vaccinated patients 18-55

https://blogs.sciencemag.org/pipeline/archives/2020/11/23/ox...

View on HN

I think the no serious cases includes the over 55s. The BMJ are clearer in their wording

>The analysis included 131 covid-19 cases and showed the vaccine was 62% effective when given as two standard doses (8895 participants) one month apart, but 90% effective when a half dose followed by a standard dose regime (2741) was used. No hospitalised or severe cases were found in anyone who received the vaccine. https://www.bmj.com/content/371/bmj.m4564

I think the 62% effectiveness which on the surface doesn't look that good, included basically asymptomatic people who PCR tested positive.

I'm not sure the other testing included asymptomatics so the difference between the vaccines may not be as large as it looks.

Were there no severe cases in the control group either? From my limited understanding of how these things are done, I assumed there must have been severe cases in the control group (which would have the same exclusions), and enough of them for some statistical test to show meaningful difference between the groups, for them to be trumpeting no severe cases in experimental group. No? If not, that would indeed to my non-statistician eye seem kind of sketchy. But if yes, thats how it works right?

Why is it so obvious that there wouldn't be any severe cases in vaccinated patients 18-55, when there have been severe cases in infected people 18-55 (outside this trial, I don't know enough about this trial to make a comment about that)?

Reply on news.ycombinator.com