A multivariate non-parametric test and a semi-parametric regression model via counting process are proposed for detecting the heterogeneity of a disease spread through a community. The infection rates are allowed to depend on time in an arbitrary manner. Infectious data usually are not completely observed, nevertheless only partial information of the epidemic is needed for the suggested methods. The testing procedures and the associated methods of analysis are illustrated with reference to epidemics of respiratory disease on the Island of Tristan da Cunha in the South Atlantic