In 1991, a major epidemic of diarrhoea broke out in Bangladesh. To estimate the extent of cholera during diarrhoea epidemics and to focus on the public health issues related to cholera in Bangladesh, we have used the government figures of the 1991 epidemic and data from our own experience of epidemic interventions in nearly 400 rural upazilas (sub-district) between 1985 and 1989 and in 1991. Our data showed that V. cholerae 01 was the most frequently (40%) isolated enteropathogen during the epidemics. The disease is widely distributed in the country. Only 24% of the total 1,648 laboratory confirmed cholera patients were below 5 years of age, and children below 2 years of age accounted for only 10% of the total. Access difficulty to medical care and absence of a reliable surveillance were thought to be the constraints to early detection and appropriate intervention, thus, there were more deaths during the epidemics. We have shown that a high proportion (59%) of cholera patients during their illness in the rural areas were not visited by the government surveillance staff and that most (80%) were treated at home. Access to treatment by qualified physicians was limited to 23% of the patients, whereas a large proportion of the patients were treated by the unqualified rural practitioners (68%), and the others (9%) had no access to any health care providers. Our experience also indicated a higher case fatality ratio (14%) prior to intervention by qualified physicians during epidemics and an overall fatality ratio of 4%, despite the significant reduction (less than 1%) achieved by the intervention. Cholera is highly epidemic in Bangladesh.(ABSTRACT TRUNCATED AT 250 WORDS)