Venom immunotherapy (VIT) is able to protect insect venom-allergic patients against life-threatening sting reactions. Standardized sting challenges can be used as a diagnostic tool to check whether VIT is required. No data are available on the long-term predictive value of sting challenge tests. The purpose of this study was to investigate the long-term predictive value of sequential bee-sting challenges with respect to the ability to predict future sting reactions in bee-venom (BV) allergic children. Between 1988 and 1992, 92 BV-allergic children had been challenged with sequential bee stings at intervals of 2–6 weeks to determine the necessity of VIT. In 1996, all 92 families were followed-up using standardized telephone interviews. Until the follow-up, 61 children (66.3%) had experienced at least one natural bee sting. Based on the results of the initial challenge tests, 13 of the 61 patients had been started on VIT. Two of these 13 (15.4%) developed systemic reactions 1 year after VIT of 5 years, of which one was mild and one was severe. Among the 48 re-stung patients who were not treated with VIT, three children (6.3%) experienced mild systemic reactions, whereas 45 children reported no more than a local reaction. The long-term predictive value of sequential bee-sting challenge tests for systemic reactions in children not treated with VIT remained at a level of 93.8% (95% confidence interval: 82.8–98.7%) even over a period of more than 6 years. Based on this data, we conclude that sequential bee-sting challenges are a powerful tool to determine the necessity for VIT in BV-allergic children.