To the Editor.
—Dr Avorn and colleagues1 report that daily ingestion of cranberry juice may reduce the frequency of bacteriuria and pyuria in older women. We believe, however, that problems in experimental design and data analysis lessen the validity of their conclusions.First, the power calculations have been performed for an infection rate that is not reflected in the study population. The authors assume a 50% infection rate, which is apparently the proportion of subjects that are expected to have at least one infection during the 6-month study period. This rate seems high in light of their own baseline data indicating that no more than 25% of treatment or control subjects had a urinary tract infection (UTI) in the preceding 6 months. Furthermore, the authors never present the observed proportion of UTIs for either group, even though the study was apparently designed around this endpoint.Second, the authors note