Carbohydrate intake and key food sources of carbohydrates in early childhood are poorly understood. The present study described total carbohydrate intake and subtypes (i.e. starch, sugar), their primary food sources and their tracking among young Australian children. Data from children at ages 9 months (n 393), 18 months (n 284), 3·5 years (n 244) and 5 years (n 240) from the Melbourne InFANT Program were used. Three 24-hour recalls assessed dietary intakes. The 2007 AUSNUT Food Composition Database was used to calculate carbohydrates intake and food groups. Descriptive statistics summarised total carbohydrate and subtype intake and their main food sources. Tracking was examined using Pearson correlations of residualised scores between time points. Total carbohydrate, starch and sugar intakes (g/d) increased across early childhood. The percentage of energy from total carbohydrates (% E) remained stable overtime (48·4–50·5 %). From ages 9 months to 5 years, the %E from total sugar decreased from 29·4 % to 22·6 %, while the %E from starch increased from 16·7 % to 26·0 %. Sources of total carbohydrate intake changed from infant formula at 9 months to bread/cereals, fruits and milk/milk products at 18 months, 3·5 and 5 years. Across all time points, the primary sources of total sugar intake were fruit, milk/milk products and cakes/cookies, whereas main food groups for starch intake included bread/cereals, cakes/cookies and pasta. Weak to moderate tracking of total carbohydrates, total sugar and starch (g/d) was observed. These findings may have the potential to inform the refinement of carbohydrate intake recommendations and design of interventions to improve children’s carbohydrate intake.

The study aimed to describe trajectories of free sugar (FS) intake, its main sources and the associations with socio-economic status (SES) in Portuguese children/adolescents evaluated at 4, 7, 10 and 13 years of age from Generation XXI birth-cohort. Dietary intake was assessed through 3-day food diaries (n 5268). Added sugar intake was estimated following a systematic methodology, and FS was based on the WHO definition. A mixed-effects model with linear and quadratic terms for time was used to estimate FS trajectories and its association with SES, adjusting for children’s sex, age, BMI and SES. The FS mean intake (g/day) was 37, 47, 51 and 48 at 4, 7, 10 and 13 years, respectively. FS intake increased 4·6 g/year (CI 95 %: 4·20, 5·04), but velocity decreased by 0·3 g/year2 until 13 years. At all ages, the main food source was sweets. Some item’s consumption declined (sweets 25–21 % and yoghurts 22–7 %) as children grew older. Inversely, soft drink intake increased (9–18 %). Boys and children from younger mothers had higher FS intake, whereas higher maternal–SES was associated with lower children’s FS intake: occupational status (β = –3·5; 95 % CI: –4·97, –1·94), years of education (β = –3·7; 95 % CI: –4·93, –2·40) and household income (β = –4·9; 95 % CI: –6·50, –3·27). The FS trajectories were similar by SES categories but different by obesity status (interaction term β = –1·04; CI 95 % (–1·50, –0·59)). An increasing FS trajectory during childhood is mainly driven by an increasing intake of soft drinks and nectars. The FS trajectory pattern did not change according to SES categories, but children’s FS intake was always higher when their mothers had a lower SES.

Trend analyses based on dietary records suggest decreases in the intakes of total sugar (TS), added and free sugar since 2005 among children and adolescents in Germany. In terms of age trends, TS intake decreased with increasing age. However, self-reported sugar intake in epidemiological studies is criticised, as it may be prone to bias due to selective underreporting. Furthermore, adolescents are more susceptible to underreporting than children. We thus analysed time and age trends in urinary fructose excretion (FE), sucrose excretion (SE) and the sum of both (FE + SE) as biomarkers for sugar intake among 8·5–16·5-year-old adolescents. Urinary sugar excretion was measured by UPLC-MS/MS in 997 24-h urine samples collected from 239 boys and 253 girls participating in the Dortmund Nutritional and Anthropometric Longitudinally Designed (DONALD) study cohort between 1990 and 2016. Time and age trends of log-transformed FE, SE and FE + SE were analysed using polynomial mixed-effects regression models. Between 1990 and 2016, FE as well as FE + SE decreased (linear time trend: P = 0·0272 and P < 0·0001, respectively). A minor increase in excretion during adolescence was confined to FE (linear age trend: P = 0·0017). The present 24-h excretion measurements support a previously reported dietary record-based decline in sugar intake since 2005. However, the previously seen dietary record-based decrease in TS from childhood to late adolescence was not confirmed by our biomarker analysis, suggesting a constant sugar intake for the period of adolescence.