High beverage sugar as well as high animal protein intake at infancy may increase overweight risk at 8 years: a prospective longitudinal pilot study

Background: Combined effects of early exposure to beverage sugar and animal protein and later life overweight risk have not been studied. Methods: A prospective longitudinal study was initiated in 2001 with 226 infants between 4 and 13 months of age. Dietary intake was assessed with a 2 day food record. Also information on infant body weight and socioeconomic status was obtained at baseline. At 8 year follow-up in 2009, children were surveyed again. Main outcome measure was overweight at 8 years as defined by BMIsds > = +1.0. Also maternal BMI, present dietary intake and physical activity, were obtained by questionnaire and 2-day food record. Results: At the 8 year follow up, 120 children (53%) were surveyed again. Of those, questionnaires and food records were completed for 63 children, for the other 57 children only weight and height at 8 years was available; 20 out of 120 children (17%) were self-reported overweight at 8 years of age. Unadjusted odds ratios (ORs; 95% CI) for overweight at 8 years were 1.10 (1.02, 1.18) for beverage sugar intake per one percent of energy intake and 4.06 (1.50, 11.00) for the highest tertile of animal protein intake at infancy compared to the lowest two tertiles. After adjustment for sex, age, infant weight, breastfed at intake assessment, and socio-economic status, odds ratios were 1.13 (1.03, 1.24) for beverage sugar, and 9.67 (2.56, 36.53) for highest tertile of animal protein intake. In the subgroup with completed questionnaire (n = 63) ORs were also adjusted for current maternal overweight, more than 2 months full breastfeeding, physical activity, and energy intake, but ORs remained significantly associated with overweight at 8 years. Conclusions: A high intake of sugar containing beverages as well as animal protein in the first year of life may increase the risk of overweight at 8 years. The results of this pilot investigation should be confirmed in a larger cohort.