Abstract

OBJECTIVE--To investigate the association between where young children defecate, where stools are disposed of, and the presence of human stools on the ground in the compound and the rate of hospital admission with diarrhoea. DESIGN--This was a case-control study with two control groups. SETTING--The study took place in Bobo-Dioulasso, the second city of Burkina Faso in West Africa. PARTICIPANTS--Three groups of children aged 36 months and under, and living in Bobo-Dioulasso were studied. Cases were 757 children admitted to hospital with symptoms of diarrhoea or dysentery. The first control group comprised 757 neighbourhood control children approximately matched on age and date of recruitment, and the second, 631 children admitted to the same hospital without symptoms of diarrhoea or dysentery. MAIN RESULTS--There was no evidence of any association between where the child was reported to defecate and hospital admission with diarrhoea or dysentery (odds ratio = 1.10; 95% confidence interval (CI) 0.78, 1.57, cases v neighbourhood controls; odds ratio = 0.84; 95% CI 0.60, 1.18, cases v hospital controls). There was evidence of an association between where the mother reported disposing of the child's stools and hospital admission with diarrhoea or dysentery (odds ratio = 1.50; 95% CI 1.09, 2.06, cases v neighbourhood controls; odds ratio = 1.31; 95% CI 0.96, 1.79, cases v hospital controls). Human stools were more frequently observed in the yards of cases than controls (odds ratio = 1.38; 95% CI 0.98, 1.95, cases compared with neighbourhood controls; odds ratio = 1.33; 95% CI 0.96, 1.84, cases compared with hospital controls). CONCLUSIONS--The findings suggest that it is not where the child defecates that matters but how the mother then deals with the child's stools.