Context
Longitudinal data regarding vitamin D status in adolescence is scarce. This study presents population-based data from an Arctic adolescent population (n = 589) at 16 and 18 years.

Objective
The aims of this study were to investigate changes in vitamin D status during 2 years in adolescence, and whether lifestyle changes were associated with serum 25-hydroxyvitamin D (s-25(OH)D) at follow-up.

Methods
Fit Futures is a longitudinal study at 69°N in Norway. Participants had their s-25(OH)D levels analyzed in their first and third year of upper secondary school (median age 16 and 18 years), in Fit Futures 1 (FF1) and Fit Futures 2 (FF2), respectively. Self-reported lifestyle habits were registered through questionnaires. The association between lifestyle changes and s-25(OH)D levels at follow-up were calculated by regression analyses, controlling for baseline s-25(OH)D levels.

Results
Longitudinal data were available for 309 girls and 280 boys. The proportion of adolescents with s-25(OH)D <50 nmol/L were 73.7% in FF1 and 77.1% in FF2, while the proportion <30 nmol/L constituted 35.7% in FF1 and 40.9% in FF2. Of those with s-25(OH)D <30 nmol/L (severe vitamin D deficiency) in FF1, 73.3% remained severely deficient in FF2. Among boys, an increase in UV exposure was significantly associated with higher s-25(OH)D levels in FF2 (beta; CI [nmol/L] 12.9; 9.1, 16.7). In girls, decreased vitamin/mineral supplement intake was significantly associated with lower s-25(OH)D at FF2 (−6.7; −10.2, −3.1), while increased UV (10.8; 7.0, 14.7) and combined hormonal contraceptive exposure (12.1; 6.0, 18.1) in FF2 was significantly associated with higher s-25(OH)D levels in FF2.

Conclusion
Severe vitamin D deficiency was prevalent throughout adolescence. Lifestyle changes may alter s-25(OH)D levels in this age group.