Objectives: The main aim of this longitudinal study was to describe dietary behaviors and Body Mass Index (BMI) in an adolescent cohort from age 15 to age 18 in Oslo, Norway. There were three sub-questions: ‡@ To what extend did dietary behaviors change in the period, and were there any differences in dietary behaviors at age 18 by gender, sociodemographic background or ethnicity when controlling for dietary behaviors at age 15 ‡A To what extend did BMI change in the period from age 15 to age 18 and were there any differences in BMI at age 18 by gender, sociodemographicbackground or ethnicity when controlling for BMI at age 15. ‡B To what extent were dietary behaviors at age 15 able to predict BMI at age 18, and were there any differences by gender, sociodemographic background or ethnicity.
Materials and Methods: The youth part of the Oslo Health Study (UNGHUBRO) constitutes the baseline of this longitudinal study. It was a questionnaire based study conducted in schools in 2000-2001 (N= 3811). The follow-up study, Youth 2004, was conducted partly in school and partly by mail in 2004. A total of 2489 (1112 boys and1377 girls) or 65 % of the participants in the baseline study also participated in the follow-up study and constitutes the study population. Almost 20 % of the participants were of Non-Western origin. Mean frequencies of intake, correlation coefficients andcrosstab analysis were used to describe changes and stability in dietary behaviors. Changes in BMI were described by the use of BMI-percentiles based on the WHO reference and the mean BMI z-score at age 15 and age 18. Multiple linear regression analysis was preformed to study if dietary behaviors and BMI at age 18 differedbetween adolescents of different sociodemographic background and ethnicity when controlling for dietary behaviors or BMI and physical activity at age 15. Indicators on sociodemographic background were parents f educational level and marital status.
Results: There were reduced mean frequency of intake of most of the registered food and beverage items among both genders in the period from age 15 to age 18. Among the boys, the reduction was significant for intake of juice, fruit/berries, raw vegetables,chocolate/sweets and chips. Among the girls, there were significant reductions in intake of soda with added sugar, juice, fruit/berries, chocolate/sweets and chips, and a significant increase in the consumption of boiled vegetables. Girls of Norwegian/Western origin with married parents/common law partners, showed a significant higher intake of fruit/berries, boiled vegetables and chocolate/sweets compared to the Norwegian/Western girls with parents of unmarried/other marital status. Girls of Non-Western ethnicity with parents of unmarried/other marital statusshowed significant higher intake of soda with added sugar compared to the Non-Western girls with married parents/common law partners. At age fifteen, 12.6 % of the boys were at or above the 85th percentile and could be classified as overweight or obese. At age 18, the proportion of overweight boys was 13.2 %. Among the girls,6.7 % were at or above the 85th percentile at age 15, while 7.5 % of them were at or above the 85th.
Conclusion: The mean decrease in frequency of intake of fruit and vegetables in the period from age 15 to age 18, support the need for identifying factors that could enhance the adolescents consumption of these healthy food items also in late adolescence. Overweight adolescents showed more variation in weight compared to the adolescents with a BMI < 85 percentile at age 18. The slightly increase in the proportion of overweight adolescents was non-significant in both genders. Mean BMI z-score values were 0.34 for boys 15 years of age and 0.03 for boys 18 years old. The mean BMI zscore values for girls were -0.01 at age 15 and -0.09 at age 18. Results therefore showed a mean decrease in age- and gender adjusted BMI both among girls and among boys. Girls of Non-Western ethnicity with parents of unmarried/other marital status were found to have significant higher BMI z-score at age 18 compared to theNon-Western girls with married partners/common law partners. In addition, there was a significant inverse association between intake of boiled vegetables and intake of chocolate at age 15 and BMI z-score at age 18 among girls of Norwegian/Western origin.th percentile in the period from age 15 to age 18.Further researchers should therefore seek to identify factors that contribute to weight variations and weight stability in late adolescence; especially among overweight youth. In addition, the relationship between girls f ethnicity, parents' marital status anddietary behaviors and BMI are interesting and should be further researched.