Visits to household during a census in an impoverished area of north Jakarta were used for exploring the four-week prevalence of diarrhoea, factors associated with episodes of diarrhoea, and the patterns of healthcare use. For 160,261 urban slum-dwellers, information was collected on the socioeconomic status of the household and on diarrhoea episodes of individual household residents in the preceding four weeks. In households with a reported case of diarrhoea, the household head was asked which form of healthcare was used first. In total, 8,074 individuals (5%)--13% of children aged less than five years and 4% of adults--had a diarrhoea episode in the preceding four weeks. The two strongest factors associated with a history of diarrhoea were a diarrhoea episode in another household member in the four weeks preceding the interview (adjusted odds ratio [OR] 11.1; 95% confidence interval [CI] 10.4-11.8) and age less than five years (adjusted OR 3.4; 95% CI 3.2-3.5). Of the 8,074 diarrhoea cases, 1,969 (25%) treated themselves, 1,822 (23%) visited a public-health centre (PHC), 1,462 (18%) visited a private practitioner or a private clinic, 1,318 (16%) presented at a hospital, 753 (9%) bought drugs from a drug vendor, and 750 (9%) used other healthcare providers, such as belian (traditional healers). Children with diarrhoea were most often brought to a PHC, a private clinic, or a hospital for treatment. Compared to children, adults with diarrhoea were more likely to treat themselves. Individuals from households in the lowest-income group were significantly more likely to attend a PHC for treatment of diarrhoea compared to individuals from households in the middle- and higher-income groups.