Secular trends in fruit intake among Danish schoolchildren, 1988 to 2006: Changing habits or methodological artefacts?

BACKGROUND: Intermittent monitoring of fruit and vegetable intake at the population level is essential for the evaluation and planning of national dietary interventions. Yet, only a limited number of studies on time trends in fruit and vegetable intake among children and adolescents have been published internationally. In Denmark, national comprehensive campaigns to enhance fruit and vegetable consumption were initiated in 2001. This paper describes secular trends in fruit intake among Danish adolescents by six comparable school surveys from 1988 to 2006. The paper demonstrates and discusses the consequences of measurement changes introduced in long-term trend analyses. METHODS: We used Danish data from the international Health Behaviour in School-aged Children (HBSC) study collected in 1988, 1991, 1994, 1998, 2002 and 2006. Analyses were conducted on comparable questionnaire-based data from students aged 11, 13 and 15 total (n = 23,871) from a random sample of schools. Data on fruit intake were measured by a food frequency questionnaire. Due to changes in number of response categories beween surveys, different cut-points were analysed. RESULTS: The prevalence of students eating fruit at least once daily ranged from 78.3% among 13-year-old girls in 1988 to 17.3% among 15-year-old boys in 2002. Based on the six data collections, analyses of trends showed a significant decrease in prevalence of students eating fruit at least once daily from 1988 to 2002 (all p-values < 0.0001). In all age and gender groups, a significant increase in intake occurred between 2002 and 2006 (all p-values < 0.0065). Analyses of alternative cut-points revealed similar results. CONCLUSION: Fruit consumption among Danish schoolchildren decreased from 1988 to 2002 with an increase since 2002. We suggest that the increase may be attributable to a nation-wide initiative conducted in Denmark since 2001 to increase the intake of fruit and vegetables in the population. Still, the results imply that a substantial proportion of Danish schoolchildren do not meet the nationally recommended daily intake of fruit. Our analyses indicate that the observed trends are not solely caused by methodological biases related to changes in measurements.