Inherently Low Precision of Infectivity Titrations Using a Quantal Response

Abstract

Experimental evidence is reviewed which suggests that infective (i.e. self-reproducing) particles act independently after inoculation with the result that (1) the dose-response curve for either completely susceptible hosts or partially susceptible hosts of identical resistance will be derived from the first term of the Poisson series when a quantal response is observed; (2) considerable variability in response will always be present in an infectivity titration using a quantal response, even if the hosts are uniform in resistance, and this will be considerably greater than that observed in many pharmacological titrations; (3) variability in response probably will not be greatly reduced by using hosts bred for uniformity as the Poissonian curve is fairly insensitive to heterogeneity in host resistances. A review of published infectivity titrations has shown that nearly all are compatible with the above predictions.