The impact of repeated chemotherapy on morbidity due to schistosomiasis mansoni was evaluated in Gihungwe (initial prevalence 58%) and Buhandagaza/Kizina (33%), two village clusters in Burundi. Surveys were carried out with reference to the first treatment (month 0) at months -6, -3, 0, 3, 6, 9, 12, 24, and 36. Praziquantel (40 mg/kg) was given at months 0, 12, 24, and 36 to those showing eggs in the feces with a single 28-mg Kato slide. At each survey, duplicate Kato smears were examined, and all participants responded to a standardized medical history interview and underwent a clinical examination. In the three preintervention surveys, spleen and liver rates remained stable at the community and the individual level. The frequencies of diarrhea and abdominal pain varied to some extent, but they were consistently higher in the most heavily infected villages and age groups and remained relatively stable at the individual level. At the final survey, the prevalence of infection had decreased to 25%, and the frequency of diarrhea from 19-26% to 10% in both village clusters. This impact was strongest in the younger age groups. The frequency of abdominal pain was reduced only at the short term and in selected age groups. Organomegaly decreased only to a limited extent in those treated, and increased in those not treated, possibly due to the impact of malaria. The net result was that no measurable impact of the treatments on organomegaly at the community level could be demonstrated. In the light of these results, the relevance of community-based chemotherapy in moderate foci is questioned.