The utility of serving size in the measurement of soft drink consumption

BACKGROUND: Many studies examining population differences in soft drink consumption or the association it has with fatness have not included serving size in its assessment. It is not clear what effect this has on their findings and our study aimed to investigate this by comparing the relationships that days (serving size unaccounted for) and cans/day (serving size accounted for) of consumption have with ethnicity/country and fatness. METHODS: Daily nutrient intakes were calculated from a self-administered food frequency questionnaire from a cross-sectional health screening study. Participants were Pacific (n=954) and New Zealand European (n=1745) people aged 35 to 74 years. RESULTS: Compared to Australian youth, NZ Pacific youth consumed soft drinks more frequently but a larger difference was observed for cans/day. In a dose-dependent manner, FMI was positively associated with days (P=0.015) and cans/day (P=0.024) of consumption. However, cans/day showed a stronger relationship, with a standardised regression coefficient of 0.066, compared to 0.033 for days of consumption. CONCLUSIONS: It is useful to include serving size in the assessment of soft drink consumption. Excluding it leads to underestimation of both ethnic/country differences in daily volume of intake and associations with fatness.