Decay and root rot caused by Heterobasidion annosum (Fr.) Bref. s. lato is the most serious disease of Norway spruce [Picea abies (L.) Karst.]. Mathematical models of disease development have recently been developed for forestry planning purposes. Functions for predicting the probability of decay were developed from two data sets, one comprising trees and another comprising stumps. From the years 1983–2001, 45,587 Norway spruce trees from the Swedish National Forest Inventory (NFI) were analysed for decay incidence at 1.3 m height and correlation with environmental conditions. The decay frequency increased in all studied regions from the first to the second half of the period for trees with comparable tree and environmental attributes. In a stepwise logistic regression, sets of functions were developed showing significance regarding stand age, site index class, temperature sum, height above sea level, diameter at 1.3 m, soil moisture and texture, proportion of spruce and eastern coordinates. The functions were calibrated and validated with a data set from the Swedish NFI from the years 1993–2002 comprising 7,893 stumps. The calibration of decay incidence at breast height to stump height doubled the decay incidence (R 2=0.85). The developed functions could be used to establish initial conditions for dynamic modelling of disease and in strategic planning.