Water Cost and Availability: Key Determinants of Family Hygiene in a Peruvian Shantytown Robert H. Gilman, MD, Grace S. Marquis, MS, Gladys Ventura, BS, Miguel Campos, MD, William Spira, PhD, and Fernando Diaz, MD Introduction Hand washing is an effective means of interrupting fecal-oral transmission and decreasing the intrafamiliar spread of diar- rheal disease pathogens such as shigella.1-4 Poor hygiene practices may be due to ig- norance of sanitary principles, high cost, scarcity of clean water, or distance from it.5-8 When water is plentiful, programs that emphasize hygiene education in- crease personal sanitation practices and result in decreased rates of diarrhea.9'10 However, educational messages may not succeed if adequate and convenient water supplies are not available.'1 Water is a scarce resource in Lima shantytowns called pueblo jovenes. 12'13 To describe the effect of water scarcity on sanitary behavior, we used direct obser- vation to determine hygienic practices in such a town. Methods Study Sites The study was carried out in 1987, prior to the cholera epidemic, in two well- described periurban pueblo jovenes of Lima, Peru: Huascar, Canto Grande, and Las Pampas de San Juan de Miraflo- res.13,14 Differences in income (based on fam- ily expenditure data) in apueblojoven are small, varying by less than $1500 per year between families. Seventeen families were selected with equal numbers in each wealth category to determine if that small difference in income affected water and soap consumption. No difference was found, and the remaining 36 families were selected from the randomly generated list without further stratification for wealth. There was no difference between the two groups in their mean socioeconomic sta- tus, family size, or water and soap use; therefore, both data sets were pooled and analyzed together. Detailed In-House Methodology Hygiene activities. Three consecu- tive days of 12-hour (7 AM to 7 PM) in- house observation were scheduled with mothers. During the observation period, the field-worker noted the activities of each family member, including prepara- tion and consumption of food and water, defecation, handling of children's feces, and bathing and washing of hands. The field-workers engaged in small talk and avoided all comments concerning soap and water use. Children were asked Family Selection Our study population consisted of families living in Canto Grande who were previously selected at random to partici- pate in a longitudinal diarrhea study.13"14 Fifty-three of these families, all with at least one child less than 3 years of age, were chosen for the current study. Al- though all families were impoverished, a skilled field-worker classified them into three ascending grades of wealth. Robert H. Gilman and William Spira are with The Johns Hopkins School of Hygiene in Bal- timore, Md. Drs Gilman and Spira and Miguel Campos and Fernando Diaz are with the Uni- versidad Peruana de Cayetano Heredia, Lima, Peru. Grace S. Marquis and Gladys Ventura are with the Instituto de Investigacion Nutri- cional in Lima, Peru. Dr Diaz is also with A. B. PRISMA, Lima. Requests for reprints should be sent to Robert H. Gilman, MD, Department of Inter- national Health, The Johns Hopkins School of Hygiene and Public Health, 615 N Wolfe St, Baltimore, MD 21205. This paper was accepted April 29, 1993. November 1993, Vol. 83, No. 11 ............................................................................ ._7- -.1 M., .4-sm. ..