Objective: To determine whether supplemental zinc, with or without additional micronutrients, affects the severity and duration of persistent childhood diarrhea and the rate of nutritional recovery. Design: The study was a community-based, double-blind, randomized trial implemented in a shanty town in Lima, Peru. Children aged 6 to 36 months with persistent (≥ 14 days) diarrhea received daily, for 2 weeks, a placebo (group P, n = 136) or a supplement of 20 mg of zinc, either with (group Z+VM, n = 157) or without (group Z, n = 139) additional vitamins and minerals. Symptoms of illness were recorded daily; and biochemical and anthropometric assessments were completed at baseline and on day 15. Results: The treatment groups were similar at baseline with regard to the characteristics of the presenting episode, anthropometric data, and plasma zinc concentration. The children consumed, on average, 95% (group P), 94% (group Z); or 88% (group Z+VM) of the supplement (P < .001). The plasma zinc concentration did not change significantly from baseline to clay 15 in group P (4 μg/dL) but increased by 38 μg/dL in group Z and 14 μg/dL in group Z+VM. The median duration of diarrhea after starting treatment was 1 day; among children who continued to have diarrhea, there was a significant effect of treatment on diarrheal duration (P = .04, analysis of covariance). Specifically, the duration of illness was significantly reduced by 28% in children in group Z (P = .01) and by 33% in girls in group Z+VM (P = .04). There were no differences in the severity of the episode by treatment group. Conclusion: There was a significant reduction in the duration of persistent diarrhea in selected subgroups of zinc-supplemented ambulatory patients in this population.