OBJECTIVES: Epidemiological studies indicate that acute uncomplicated urinary tract infections (UTI) in women can be successfully treated with short treatment regimens. However, the findings from the literature do not match experiences in daily practice. METHODS: A randomised, controlled trial evaluating the microbiological and clinical (self-reported) cure rates of a three-day vs. five-day treatment regimen with trimethoprim for UTI in women. RESULTS: No statistically significant difference in bacteriological cure rate was found between the three-day and five-day regimen. One day after the shorter regimen 44% of women considered themselves as 'not-recovered' due to persistence of the symptoms compared with 35% after the five-day treatment (p > 0.05). Three days after therapy these percentages were 30 and 25% respectively. CONCLUSION: The relatively high percentage of persistent symptoms one day after the three-day regimen might be responsible for general practitioners believing that short regimens are not successful. It is therefore advisable that if urine samples are controlled to wait at least three days after finishing treatment.