Objective: To explore determinants of serum 25-hydroxyvitamin D (s-25(OH)D) during autumn in young, Caucasian children not consuming vitamin D-fortified foods or supplements, and explore differences in sun behaviours between pre-school and school children.

Design: In September-October, s-25(OH)D was measured by LC-MS/MS; physical activity, sun behaviours and vitamin D intake were assessed with questionnaires.

Setting: Baseline data from the ODIN Junior trial at 55°N.

Subjects: Children aged 4-8 years (n 130), of whom 96% gave blood samples.

Results: Mean s-25(OH)D was 56·8 (_SD 12·5) nmol/l and positively associated with fat-free mass index (P=0·014). Children being active 6-7 h/week had 5·6 (95% CI 1·1, 10·0) nmol/l higher s-25(OH)D than less active children (P=0·014). Children seeking shade sometimes or rarely/never had 7·0 (95% CI 1·2, 12·9; P=0·018) and 7·2 (95% CI 0·8, 13·6; P=0·028) nmol/l higher s-25(OH)D, respectively, than children always/often seeking shade. Pre-school children had more sun-safe behaviour than school children in terms of use of a hat, sunscreen and sunscreen sun protection factor (P<0·05). In school but not pre-school children, using a hat rarely/never was associated with 12·1 (95% CI 2·5, 21·7; P=0·014) nmol/l higher s-25(OH)D v. always/often (P_interaction=0·019). Vitamin D intake was not associated with s-25(OH)D (P=0·241).

Conclusions: Physical activity and sun behaviours are associated with s-25(OH)D in young children. Identifying factors influencing autumn s-25(OH)D is relevant to optimize levels before sun exposure diminishes. Strategies to reduce risk of inadequacy should consider risk of skin cancer and sunburn, and could include fortification and/or vitamin D supplementation.