I read this recently: I've noticed a fair number of people assuming the herd immunity threshold, 1 - 1/R0, is the total proportion of the population infected.

I read this recently: I've noticed a fair number of people assuming the herd immunity threshold, 1 - 1/R0, is the total proportion of the population infected. It isn't; it's the proportion of the population immune (i.e. has had the disease; in this model that's assumed to confer immunity) at which the epidemic stops growing. The total proportion of the population infected is called the epidemic final size (I'll call it 'F'), and is given by F = 1 - e-R0*F This is higher. Lots higher. For R0 = 2, it's 80%. For R0 = 3.4 (the WHO estimate), it's 96%.