Validering av et nettbasert frekvensspørreskjema : blant ungdom

Abstract

As a number of today's health problems are related to nutrition, expanded knowledge about nutrition is essential. The Norwegian Mother and Child Cohort Study aims at implementing a nutrition investigation among 13-14-year-old adolescents during the next 2–10 years. The purpose of this nutrition study is to investigate the validity of a food frequency questionnaire (FFQ) applied to 13-14-year old adolescents. After having been tested for its validity, the FFQ is to be used in the Norwegian Mother and Child Cohort Study. The validation study is carried out on 93 Norwegian 13-14-year-old pupils at four schools in Vest-Agder, who filled in a web-based FFQ about nutrition and subsequently accomplished 2x24-hour recalls at an interval of five weeks. A selection of 14 food intake variables was subjected to data processing. Correlation was measured by the Spearman correlation coefficient (r) between FFQ and the 2x24-hour recall. Further, the methods were tested statistically for the proportion classified in the same tertil by Cohen's kappa. Statistical testing by Spearman's r showed a relatively strong correlation between FFQ and the 2x24-hour recall for the foodstuffs cheese, sweet drink, water and fruit/berries (with r from 0.34-0.45) and a very strong correlation for juice (r 0.55). The proportion classified in the same tertil varied from 28 % to 54.8 %. Cohen's kappa values varied between – 0.016 and 0.325 depending on type of foodstuff. Comparison of the validity testing of the FFQ and the 24-hour recall showed statistically significant correlations with 8 among the 14 selected foodstuff variables. Cheese, juice and water had better correlations than the other types of foodstuff. The questionnaire is an appropriate method for surveying frequently consumed foodstuff. Keywords: food frequency questionnaire, 24-hour recall, adolescents, validation