What does 95% efficacy for the vaccines mean?

in #covidlast month


The efficacy number is calculated as a ratio between vaccinated and unvaccinated (placebo) for some outcome. For the RNA vaccines, the outcome they were looking for in the big trial was symptomatic covid. For J&J the outcome was moderate-to-severe covid. We also have an experiment in Israel showing that the Pfizer vaccine reduces hospitalizations by about the same amount as it reduces symptomatic covid, if you want to to a closer comparison.

The ratio for a certain outcome between the vaccinated and unvaccinated people gives us a relative risk. If you are fully vaccinated with one of the RNA vaccines, your relative risk of symptomatic covid is about 0.05, or to put it another way, the reduction of your risk is 95%.

If you want to know absolute rather than relative risk, you need to know what your background risk is. We know that at least with the older variants, your risk of catching covid if you have a close contact with an infectious person (15 min face-to-face conversation indoors with no masks) is about 20%. That's an absolute risk. If you reduce that absolute risk by 95% because you are vaccinated, your risk becomes 5% of 20%, or 1%. So your absolute risk of catching covid after a close contact with an infectious person is 1%. It's probably lower if the other person is also vaccinated, but at this point we don't know by how much.

If you wanted to know your absolute risk for a certain time period, you'd have to know how much exposure you have in that time period.

The vaccines also reduce the symptoms and the amount of virus shed if you do get covid, so there still is a benefit, even if you are unlucky enough to get covid in spite of being vaccinated.