The medical team performed daily rounds to identify events of interest. All physicians involved were blinded to the study. The project was registered at the National

Commission on Research Ethics (CONEP) on 10/28/2008, and approved by the Research Ethics Committee of the Hospital Universitário Professor Edgard Santos, under protocol No. 042/2008, on November 6, 2008. The analyses were performed considering only the data collected at the daycare. To assess the main variables (presence of DD and ARI), measures of occurrence (incidence) and of association (relative risk – RR) were calculated, as well as the relative risk reduction (RRR) Veliparib mw and the respective confidence intervals (CI). Fisher’s exact test was used to compare characteristics between groups at baseline. The parametric Student’s t-test was used to analyze the continuous variables Everolimus concentration (anthropometric

indicators) with normal distribution. A Poisson regression model was used to identify interactions and confounding factors for DD and ARI. The Kaplan-Meier test was used to assess duration of DD episodes. The overall significance level was set at 5%. The database was built using the EpiData software, release 3.3.1, and the analyses were performed using R. Of the 150 children enrolled in the study, seven were excluded: three for age less than six months and four for no longer attending the daycare. Thus, 143 were randomly assigned to the two groups. Of these, 52.45% (n = 75) were assigned to group A (test) and 47.55% (n = 68) to group B (control). All children completed the study. The groups were similar regarding admission characteristics – gender, age, selleck chemical and nutritional status (Table 2). Severe anemia was not identified in any of the participants. The anthropometric indicators W/H, W/A and H/A were also similar between groups at baseline (Table 3). However, after the intervention (p-values not shown in Table), there was an increase in

mean z-scores of W/H and W/A in both groups, whereas in controls this improvement was statistically significant for both W/H (p = 0.033) and W/A (p = 0.005). The same result was not observed for the H/A ratio: in the test group, it maintained the same mean (p = 0.9634) and in controls, it showed a slight decrease (p = 0.007). The calculation of ingestion, including the addition of sprinkles, showed no difference in the mean consumption of energy (A: 72.09 kcal/kg vs. B: 72.96 kcal/kg; p = 0,728), fibers (A: 6.7 g vs. B: 6.83 g; p = 0.564), proteins (A: 2.08 g/kg vs. B: 2.11 g/kg; p = 0.654), carbohydrates (A: 9.31 g/kg vs. B: 9.48 g/kg; p = 0.602) and lipids (A: 2.95 g/kg vs. B: 2.95 g/kg; p = 0.949) between the groups. Only zinc levels were different between the groups, due to the supplementation in the test group (A: 7.16 mg vs. B: 2.3 mg; p < 0.001).