In developing countries, poor families in urban slums often do not receive municipal services including water. The objectives of our study were to characterize families who purchased drinking water and to examine the relation between purchasing drinking water and child morbidity and mortality in urban slums of Indonesia, using data collected between 1999 and 2003. Of 143,126 families, 46.8% purchased inexpensive drinking water from street vendors, 47.4% did not purchase water, i.e., had running or spring/well water within household, and 5.8% purchased more expensive water in the previous 7 days. Families that purchased inexpensive drinking water had less educated parents, a more crowded household, a father who smoked, and lower socioeconomic level compared with the other families. Among children of families that purchased inexpensive drinking water, did not purchase drinking water, or purchased more expensive water, the prevalence was, respectively, for diarrhea in last 7 days (11.2%, 8.1%, 7.7%), underweight (28.9%, 24.1%, 24.1%), stunting (35.6%, 30.5%, 30.5%), wasting (12.0%, 10.5%, 10.9%), family history of infant mortality (8.0%, 5.6%, 5.1%), and of under-five child mortality (10.4%, 7.1%, 6.4%) (all P<0.0001). Use of inexpensive drinking water was associated with under-five child mortality (Odds Ratio [O.R.] 1.32, 95% Confidence Interval [C.I.] 1.20-1.45, P<0.0001) and diarrhea (O.R. 1.43, 95% C.I. 1.29-1.60, P<0.0001) in multivariate logistic regression models, adjusting for potential confounders. Purchase of inexpensive drinking water was common and associated with greater child malnutrition, diarrhea, and infant and under-five child mortality in the family. Greater efforts must be made to ensure access to safe drinking water, a basic human right and target of the Millennium Development Goals, in urban slums.