“…We included a random interaction between farm and period in the first two models, to account for non-genetic effects of infectivity. This interaction serves to avoid overestimation of the genetic variance in infectivity (Anche, 2016), similar to the inclusion of a random group effect in the analysis of indirect genetic effects (Bijma et al, 2007;Bergsma et al, 2008). The genetic effect for infectivity and the interaction term were partly confounded, because both effects reflect the number of susceptibles that became a case within a certain period on a certain farm.…”