The aim of this study was to investigate how the Scientific Committee on Occupational Exposure Limits (SCOEL) of the European Commission uses uncertainty factors when proposing health-based indicative occupational exposure limit values (IOELVs). In total, 75 IOELVs in 62 summary documents published from 1991 to 2003 were analyzed. For 31 of the IOELVs, no explicit uncertainty factor (EUF) was stated. For these, we calculated an implicit safety margin (ISM) as the ratio between the point of departure (POD, derived from the NOAEL or LOAEL of the critical effect) and the proposed IOELV. We further analysed whether date of recommendation, type of critical effect, nature of POD or amount of available data influenced the magnitude of the EUFs and ISMs. The ISMs varied little (range 1-5), while the EUFs showed more variability (range 1-50). The EUFs remained unaffected over time and the ISMs decreased slightly. Significant differences in the magnitude of the EUFs, but not ISMs, were found between critical effects, however, contrary to expected the average EUFs and ISMs for irritation were similar to those for more severe systemic effects. The nature of the POD affected the ISMs and EUFs only slightly and less than expected. Both EUFs and ISMs showed a weak but significant negative correlation with the amount of available toxicological data, measured as the number of relevant publications in PubMed, whereas SCOEL statements on data sufficiency had no influence. Overall, the most striking difference was that between EUFs and ISMs, the former being on average 2.1 times higher.