Sixty-six patients with human immunodeficiency virus infection and diarrhoea caused by Cryptosporidium parvum were enrolled in a double-'blind' placebo-controlled study to evaluate the safety and efficacy of nitazoxanide in the treatment of cryptosporidiosis related to the acquired immune deficiency syndrome. Patients were randomly assigned to one of 3 treatment groups and received either 500 mg twice daily of nitazoxanide, 1000 mg twice daily of nitazoxanide, or placebo orally for 14 d; the patients on nitazoxanide then crossed over to placebo while the placebo patients crossed over to nitazoxanide therapy at either the high or low dose depending on their randomization. Three post-treatment faecal examinations were conducted on days 15, 22 and 29 following initiation of treatment: patients were considered 'cured' if none revealed any C. parvum oocysts. Both doses of nitazoxanide produced parasitological cure rates superior to the placebo responses (12/19 [63%, P = 0.016] for patients receiving 1 g/d and 10/15 [67%, P = 0.013] for those receiving 2 g/d). Parasitological cure was correlated with the complete resolution of the diarrhoeal syndrome in 19 of the 22 treated patients who were considered parasitologically cured (86%). Both doses of nitazoxanide were well tolerated by the patients.