95% efficacy doesn’t mean what you think it means. From
A -short- look at how vaccine makers like Pfizer and Moderna get to claim a 90% or even 95% efficacy for their products, with the help of regular Automatic Earth commenter Doc Robinson (not a medical doctor) and his quotes from the British Medical Journal (BMJ). The way the companies report their efficacy may be normal in their circles, but will, in the “normal” world, be experienced as confusing if not outright misleading.
What they do -simplified-, let’s take the Pfizer report, is they have 20,000 volunteers who get a vaccine, in this case 8 are infected, and that gives a result of -well- over 90% efficacy. But that is largely meaningless, because it appears to assume that all remaining 19,992 volunteers would have gotten infected if not for the vaccine.
To give this meaning, the world of science has long insisted on control groups (placebo groups), in this case also 20,000 strong, who don’t get a vaccine. If you know how many in that group are infected, you know -much better- hoe effective the vaccine is. Turns out, in the control group 86 out of 20,000 were infected. More than 8, but much less than 20,000. 19,914 unvaccinated people never got infected.
The 90%-95% numbers “measure” relative risk reduction. The absolute risk reduction is completely different. In the Pfizer case, 99.57% of the unvaccinated people did not become infected, while 99.96% of the vaccinated people did not become infected. Therefore, the absolute risk reduction is 99.96% – 99.57% = 0.39%. While there remain many questions swirling around the mid- to long term effects of taking the vaccine.
You would think this is the most relevant information out there for those thinking about being vaccinated or not, and not the 95% relative risk reduction. But the latter info is what is reported. And sure, it sounds much better.