Fecal contamination of shanty town toddlers in households with non-corralled poultry, Lima, Peru.

We used direct observer techniques to measure the frequency with which toddler-aged children were contaminated by poultry feces in homes in a peri-urban shanty town in Lima, Peru. The mean number of fowl was 5.4 (SD 3.1), with 10.0 (SD 10.7) poultry defecations per 12 hours. Toddlers' hand contact with poultry feces occurred a mean of 2.9 (SD 3.0) times/12 hours. A mean of 3.9 (SD 4.6) feces-to-mouth episodes per household/12 hours occurred both by direct hand-to-mouth contamination and indirectly by handling soiled objects which were then placed in the mouth. There was a strong correlation between feces-to-hand contamination and feces-to-mouth contamination (r = 0.94). There was also an association between feces-to-mouth contamination and the number of stools deposited in the house (r = 0.66). For each additional chicken stool deposited during the day, there was an average increase of 0.27 in feces-to-mouth episodes. We collected feces from 68 infected chickens and found viable Campylobacter jejuni for up to 48 hours after deposition. Yet, a survey of 108 families demonstrated that free-roaming poultry were often not thought of as a health risk for children. An intervention program to reduce oral-fecal contamination should emphasize that all poultry be corralled and not allowed access into the house.