OBJECTIVE: The EORTC QLQ-C30 is one of the most widely used QL measures in cancer clinical trials. This study aimed to look at the psychometric properties of the QLQ-C30 in 32 countries with a database of 9000 patients. METHOD: All EORTC studies incorporating the EORTC QLQ-30 were systematically selected for this study. Inclusion criteria for trials were if the trial contained the QLQ-C30 responses had been coded into the EORTC database. 114 EORTC studies were reviewed of which 52 met the criteria for being included in the final analysis. RESULTS: The majority of cancer patients were receiving palliative for primary cancers including melanoma, prostate, head and neck, breast and lung and 90% of the patients were distributed over 10 out of the 32 countries. At least one item of data was missing in 14 to 17% of patient questionnaires completed though the average percentage of missing items per patient ranged from 1.1% to 1.5% Particular items relating to Role Functioning and Financial Difficulties were the most common items missing (3%). Factor analyses for all 3 versions of the questionnaire are similar, though some difference has been found when examining individual cancers. Interdomain correlations, in each of the 3 versions of the QLQ-C30 were strongest in Role Functioning, Pain and Global Health Status. Cronbach's reliability measure shows improved reliability in newer versions of the questionnaire, for example, the pain scale increases from 0.82 to 0.86 from version 1 to version 3. CONCLUSION: We believe this is one of the first studies to examine the scales of all three versions of the QLQ-C30 with a large sample across a large number of countries and cancer sites. We found that in general all 3 versions of the QLQ-C30 have similar psychometric structures and that the measure proves to a be a useful tool to use within a clinical trials setting and that missing data is continually reducing over time, indicating increasing compliance among staff/patients.