A review of the diarrhoeal disease literature reveals considerable variability in the definition of diarrhoeal episodes. The use of various definitions of diarrhoea and episodes leads to misclassification, affects the estimates of the disease burden in communities and reduces comparability of the findings from different studies. This study is an attempt to validate the definition of diarrhoeal episodes using prospectively collected community-based surveillance data. In comparative validation analyses, three or more loose stools or any number of loose stools containing blood in a 24-hour period seemed to be the best definition of diarrhoea. Three intervening diarrhoea-free days seemed to be the optimum to define a new episode. The implications of using differing definitions and the importance of using a validated definition are discussed.