Summary The aim of this study was to employ the TRUE test® assay to confirm the presence or absence of the ‘angry back’ phenomenon, i.e. that a strong positive patch-test reaction heightens adjustment patcht-est response. In addition, we wished to establish the dose-response relationship for nickel sulphate patch tests among nickel-sensitive patients. Seventy-two nickel-sensitive patients. 36 in Odense and 56 in Stockholm, were tested with a 10-step nickel sulphate dilution series (TRUE test® and two placebo patches. The position of the patches was rotated, to provide a balanced spatial distribution of the different concentrations. Readings were performed blind. The results were analysed by means of polynomial multiple-regression methods and a logistic dose–response model. Half the patients (38/72) had a threshold patch-test concentration for nickel sulphate in the range of 3.0—3 μg/cm2. The ‘angry back’ phenomenon was not apparent in this study, as the spill-over effect was not statistically significant. Strong reactions to high concentrations of nickel sulphate did not enhance the response to adjacent lower concentrations of nickel sulphate.