Practice guidelines for ordering stool cultures in a pediatric population. Alberta Children's Hospital, Calgary, Alberta, Canada.

A comprehensive utilization review was done of all stool culture tests performed at a pediatric hospital during a 3-year period from June 1, 1989 through to July 1, 1992. A total of 4,460 stool culture specimens were surveyed from 3,420 children. Sixty percent (2,692) of the workload was from inpatients, 22% (1,001) was from emergency room (ER) visits and 18% (767) was from outpatient clinic/office visits. A total of 9% (294 of 3,420) of the children were confirmed to have enteric bacterial infection. Enteric infections in Southern Alberta follow a typical pattern, where most cases of Escherichia coli O157:H7, Salmonella sp and Campylobacter jejuni are diagnosed in the summer months between June and September. Marked differences were found in the overall stool culture positivity rate between hospitalized children (94 of 2,141, 4%), and ambulatory children attending the ER (176 of 892, 19%). Although a significant number of children had more than one stool culture done, most cases of enteric bacterial infections were diagnosed on the first stool sample (290 of 294, 98%). Most children (89 of 94, 95%) who were hospitalized because of an acute diarrheal illness had a bacteriological diagnosis confirmed within the first 4 days after admission (90 of 94, 96%), including immunocompromised children in the Oncology ward. Over the past year, significant sustainable cost savings have resulted from the implementation of practice guidelines for ordering pediatric stool cultures.