OBJECTIVE: The present study analyzes the Lactose Tolerance Test in order to propose a reduction in the length of it. DESIGN: Cross-sectional, descriptive study. METHODS: Glycemia curves of patients considered tolerant and intolerant were compared. Results were analyzed for 56 patients. A cost minimization analysis was performed. RESULTS: Six patients had a normal curve of tolerance and the other 50 had flat curves. Both groups had similar basal levels of glycemia 93.3 and 90.3 (p=0.1). Glycemia levels measured at 30 minutes were 122 and 86.0 (p= 0.003) and at 60 minutes 108.3 and 87.5 (p=0.02), both groups being statistically different. Values at 120 minutes were not statically different between both groups (p=0.4) and not different from basal levels. CONCLUSIONS: Diagnosis of lactose intolerance could be made using a shorter test, without considering the 120-minute figure. A 25% reduction in user s cost could be attained using the modified method as compared with the standard test.