Abstract
OBJECTIVE: We examined whether drinking water per se is associated with drinking less of other beverages and whether changes in BMI are associated with the intake of water and other beverages. DESIGN: Secondary analysis of a randomized trial of fourth graders followed over 1 year. SETTING: Public schools in the metropolitan area of Rio de Janeiro, Brazil. SUBJECTS: Participants were 1134 students aged 10-11 years. RESULTS: At baseline, a higher frequency of water consumption was associated with a greater daily intake of fruit juice (P = 0.02) and a higher daily frequency of milk (P = 0.005). In the intervention group, the baseline frequency of water consumption was negatively associated with weight change over 1 year but without statistical significance (coefficient = -0.08 kg/m2; 95 % CI -0.37, 0.24 kg/m2), whereas fruit juice intake frequency was positively associated with weight change: each increase in fruit juice intake of 1 glass/d was associated with a BMI increase of 0.16 (95 % CI 0.02, 0.30) kg/m2. CONCLUSIONS: Our findings do not support a protective effect of water consumption on BMI, but confirm consumption of juice drinks as a risk factor for BMI gain. Students who reported high water consumption also reported high intake of other beverages; therefore, the promotion of water consumption per se would not prevent excessive weight gain.