Figure 2.

Graph represents adjusted percentages from individual participant data of 7 growth studies where infants were exclusively formula-fed PHF-W or CMF (Nan) from 2 weeks of age to 4 months. For 2 days before study visits at 1, 2, 3, and 4 months of age, caregivers kept a daily diary and recorded the consistency of each stool passed. Statistical significance was determined using marginal probabilities for stool consistency based on a longitudinal multinomial regression model with adjustment for treatment, age and treatment-by-age interaction, and accounting for clustering within study and infant. Statistical differences were observed for soft, formed, and hard stools at all time points.