Objectives: To determine the age-specific hospitalization rate for rotavirus-associated diarrhea in Canadian children during the seasonal outbreak, and to characterize children and their households, for assessment of the need for a rotavirus vaccine. Design: Prospective multisite cohort study. Settings and Participants: Children with an admission diagnosis of diarrhea admitted to 18 hospitals serving 132 study census tracts of a major urban region, from November 1, 1997, through June 30, 1998. Prospective centralized testing of stools was performed, research nurses administered a follow-up questionnaire to parents. Main Outcome Measure: Age-specific diarrhea and rortavirus-associated hospitalization rates. Results: Of 224160 children younger than 5 years, the diarrhea hospitalization rate was 4.8 in 1000 (n=1086) during the seasonal epidemic. Based on testing of 65% of the hospitalized children, the rota virus-associated diarrhea hospitalization rate was 1.3 in 1000; the cumulative incidence to 5 years of age was 1 in 160. Rotavirus-associated diarrhea was reported in 37% of the 1001 hospitalized children undergoing testing inside and outside of the census tracts; in children aged 6 to 35 months, this hose to more than 70% during April and May. Ages of children with rotavirus-associated diarrhea were 0 to 2 months (2%), 3 to 5 months (5%), 6 to 23 months (60%), 24 to 35 months (15%), and 36 months or older (19%). Of children aged 0 to 5 and 6 to 11 months, 4 (19%) of 21 and 6 (10%) of 59, respectively, had been born prematurely; 20 (24%) of 83 younger than 1 year were breastfed at the time of illness. Of children younger than 36 months, 77% were cared for in their homes; 13%, in family day care homes; and 8%, in child care centers. The mean (+/- SD) duration of rotavirus hospitalization based on hospital records and parental questioning was 2.4+/-1.7 and 3.1+/-1.6 days, respectively; it was significantly longer (P less than or equal to.001) in children with an underlying medical condition. One child required intensive care unit hospitalization. Diarrhea occurred concurrently in 74% of household contacts younger than 3 years; 38%, aged 3 to 18 years; and 29%, older than 18 years. Seventy-six percent of parents were married. Household incomes in Canadian dollars in the 81% reporting were less than $20 000 in 20%, $20 000 to $60 000 in 44%, and greater than $60 000 in 36%. Ethnicity was reported as 53% white, 15% black, 10% Asian, 12% East Indian, and 11% other. Conclusions: Based on testing of 65% of children with diarrhea, rotavirus resulted in hospitalization in a minimum of 1 in 160 children by 5 years of age during the seasonal outbreak. Had 100% of young children with diarrhea undergone testing, the extrapolated cumulative incidence of rotavirus-associated diarrhea by 5 years of age may have been 1 in 106.
