Non-reproducibility of positive patch test reactions is common. In this retrospective study, we investigated the reproducibility of previously positive patch test reactions and the appearance of new positive reactions in a consecutive series of 264 eczema patients tested at least twice over a 12-year period using the TRUE Test standardized patch test system. Out of 297 positive reactions, 66% remained positive in a 2nd test, 10% were scored as doubtful and 24% as negative. Among the previously recorded doubtful reactions, 13% were positive in the 2nd test, 13% were again doubtful and 74% negative. In relation to the degree of patch test reactivity, 48% of the + reactions were reproducible, compared to 87% of ++ and +++ reactions. No difference in reproducibility was observed between patients tested with a short or a long-time interval between tests. With the use of the TRUE Test panels 1 and 2, methodological factors are minimized. The variation in patch test positivity may then be attributed to biological variation, which depends on several individual factors such as patch test reading, patient-related factors such as changed immunological sensitivity, regional variation of the skin on the back and the presence of more or less active dermatitis.