Standard methods for analysing survival data or case-control data normally concern factors affecting a subject as a whole. However, in a study of a condition that might develop in one or both of a pair of bodily organs information on response and on covariates may be available for each separately. This information can be used to investigate whether the presence of a factor on one side is associated with the condition on that same side, and thus leads to more direct evidence concerning aetiology. In this paper methods of analysing data on paired organs are described. The techniques are illustrated using data from a case-control study of breast cancer, in which the role of breast calcifications was of interest.