In other words, each additional copy of the risk allele increases the odds of disease by the same multiplicative factor. This is variously referred to as either the additive model or the multiplicative model. We use the latter term throughout but will refer to β as the additive parameter or effect since it naturally operates additively on the log scale. The widely used Cochran-Armitage trend test [Armitage, 1955] is the score test of the null hypothesis (β = 0) under this model [Sasieni, 1997].
