Five years after a single dose treatment, prevalence, intensity and morbidity of schistosomiasis mansoni were evaluated in Agua Branca, a low endemic community in the South East Brazil (cure rate 94%). At community level, prevalence showed a decrease from 38.9 to 24.5% and the intensity of infection dropped from 119.5 to 38.9 eggs per g of faeces (epg). However, after the exclusion of immigrants, newborn children and individuals that had left the area after the first evaluation, the prevalence among the treated and followed population was not significantly affected. Multivariate analysis showed that the 10-29 age group and water contact for agricultural purposes were independently associated with the presence of infection on post treatment evaluation [OR 3.9 and 5.09, respectively]. A previous treatment among subjects older than 15 years was inversely associated [OR 0.58]. The authors wish to draw attention to the fact that mobility may lead to a serious bias in evaluating the impact of the control programme.