Although Shigella dysenteriae type 1 has been responsible for large outbreaks of severe dysentery in many parts of Asia, relatively few cases of this disease have been reported from Thailand and have generally not involved nalidixic acid resistant strains. Beginning March 1991, all patients with diarrhoea seen at the hospital outpatient department (OPD) in Suan Phung, Thailand (a western district near the Burmese border) were cultured for enteric pathogens. Shigella dysenteriae 1 was first recognized in July 1992, and an OPD-based case-control study was conducted to pinpoint the source of the outbreak in the community. For each case of culture confirmed S. dysenteriae 1, one control person without diarrhoea, matched by age and date of visit, was randomly selected from the OPD registry. Of 197 patients treated for diarrhoea at the hospital OPD in July and August 1992, 79 (40%) had bloody diarrhoea, compared with 86/561 (15%) patients seen during 16 months of previous surveillance (PShigella dysenteriae 1 was isolated from 33/197 (17%) patients. Compared to matched controls, patients with S. dysenteriae 1 were more likely to attend one of the local elementary schools (odds ratio = 6.74, P= 0.025), or live in the community surrounding this school (odds ratio for non-school age people = 18.0, P = 0.008). A cross-sectional study conducted at the school indicated that 50 (10%) of 485 students had dysentery in July. A coconut milk dessert prepared at the school was identified as the vehicle of transmission (relative risk = 24.9, PS. dysenteriae 1 emerged in a community in Thailand, and was traced to a point source outbreak at a local school.