OBJECTIVE: To determine changes in the last 5 years of the fraction of acute and chronic symptoms attributable to schistosomiasis japonica in fishing communities with moderate endemicity in the Dongting Lake region of China. METHOD: Complete medical histories, physical examinations, and stool samples were obtained from 1909 individuals (53% male) ranging in age from 4 to 81 years. Age, sex, occupation, frequency of water contact, number of times treated for schistosomiasis, and last year of treatment were tested as potential effect modifiers and confounders. RESULTS: Overall, there were very few infected cases and very little variation in any prevalence ratio with infection intensity, sex, age, year of last treatment, number of times treated or frequency of water contact. The attributable fraction of liver enlargement in one village was estimated at 0.114 (11.4%), although this was not significantly > 0(95% CI; -0.008-0.237). CONCLUSION: The chemotherapy-based schistosomiasis control programme in this locality has succeeded in reducing attributable morbidity to undetectable levels, but many uncertainties remain about sustaining control efforts in the future. Both old and new control strategies will have to be examined if the health and well-being of these people are to be maintained into the next millennium, given that periodic chemotherapy will not be accepted indefinitely.