Children will receive 10-40 billion CFUs/day of Bio-K+ (Lactobacillus acidophilus CL1285, Lactobacillus casei LBC80R and Lactobacillus rhamnosus CLR2), with dose based on their weight. The product will be administered as a strawberry flavored tub of milk.

Strawberry flavored tub of milk, identical (taste, color, odor) to the active Bio-K+ treatment.

Probiotic (BioK+) with 3 stains of Lactobacillus

The investigators will employ a questionnaire addressing Pediatric Acute Diarrhea (qPAD), a measure of diarrhea involving the assessment of stool frequency and consistency for each bowel movement. To measure consistency, the qPAD contains a stool consistency classification system. Our registered sample size is based on pilot data from The Hospital for Sick Children indicating that the incidence of AAD is 33% (95% CI 23% to 43%) according to the qPAD. Given an estimated baseline AAD risk of 32.9% and a 48% relative risk reduction in AAD (Johnston et al, Cochrane Library, 2011), a randomized trial with 80% power and a 2-sided alpha of 0.05 comparing Bio-K+ with placebo would require a total sample size of 118 patients per group (236 total).

Given that parents have knowledge of the child's typical bowel movements per day, the investigators will also employ a Global Rating Scale for diarrhea (GRSd), using a 0 to 4 scale (0 = no diarrhea, 1 = mild diarrhea, 2 = moderate diarrhea, 3 = severe diarrhea, 4 = worse imaginable diarrhea). The investigators will ask participants to select values based on diarrhea severity, which is a combination of stool frequency and consistency over 24 hours. Our registered sample size is based on the incidence of AAD (33%; 95% CI 23% to 43%) according to the qPAD. However, the incidence of AAD according to GRSd is 23%, which corresponds to the lower bound of the 95% CI for the qPAD. The investigators are currently applying for additional peer-reviewed funds. If these funds are obtained the investigators will power the trial based GRSd, a more conservative AAD incidence (23%). The investigators will also power the trial for a smaller treatment effect (39% relative risk reduction).

The investigators will assess severity using the qPAD and the investigators will employ definitions for diarrhea from the Canadian Nosocomial Infection Surveillance Program, modified for use in children, defined as: severe: ≥6 loose/unformed/liquid stools; moderate: ≥3 loose/unformed/liquid stools; mild: a change in stooling pattern to 1-2 loose/unformed/liquid stools daily (Gravel 2007). The investigators will classify severity according to the 24-hour period with the highest degree of severity.