The aim was to assess the reliability of bulky DNA adducts measurement by means of the 32P-post-labelling assay. The research design consisted of an intramethod reliability study. Buffy coats from 41 subjects were used to obtain two aliquots of 1-5 microg DNA for each subject; bulky DNA adducts were measured using the nuclease P1 32P-post-labelling technique. The reliability of the measurement was assessed by means of the intraclass correlation coefficient (ICC), the distribution of the differences between the two measurements and the limits of agreement. The estimated ICC was 0.977, with a 95% confidence interval between 0.921 and 0.977. The limits of agreement were +/- 0.44 (DNA adducts per 10(8) nucleotides). Only three subjects had differences lying out of such limits. Bulky DNA adduct levels measured by the 32P-post-labelling technique showed good reliability. Only one measurement is needed to use DNA adducts as a biomarker of exposure and, possibly, cancer risk. Besides, as a validation analysis, 32P-post-labelling measurements can be repeated in only 20-30% of samples.