Inter-laboratory variability of anti-beta2-glycoprotein I antibody measurements (IgG and IgM) was investigated in the frame of the European Forum on Antiphospholipid Antibodies and its Standardization Group. Twenty-eight samples from patients with autoimmune diseases, two samples from blood donors and a set of six calibrators obtained by dilution with normal plasma of a pool of patient samples were sent to 21 European centers. Six of them used commercial kits and the others home-made assays. Marked differences in the steepness of the calibration curves obtained with the calibrator provided were observed. The standard deviations of sample measurement were high. Cut-off of positivity varied from 7 to 90 Forum Units (FU) for IgG and from 10 to 138 FU for IgM, whereas the rate of positivity varied from 50 to 93% for IgG and from 13 to 70% for IgM. No clear relationship between cut-off values and positivity rate could be established for either isotype. Adopting a common cut-off did not markedly improve the overall agreement between centers in positive/negative sample classification. Because of the majority of low positive samples, excellent concordance between centers (as defined by kappa values from 0.8 and 1) occurred only in 13% of cases for IgG and in 6% of cases for IgM, because many selected samples were low-positive. Despite the large variability of anti-beta2-glycoprotein I measurements between centers, the agreement on results with high- and medium-positive samples was good.