Development and Reproducibility of a Food Frequency Questionnaire to Assess Diets of Older Children and Adolescents

https://doi.org/10.1016/S0002-8223(95)00086-0Get rights and content

Abstract

Objective To develop a self-administered food frequency questionnaire for older children and adolescents and to demonstrate reproducibility over a 1-year period.

Design The youth/adolescent questionnaire (YAQ) was based on the validated Nurses’ Health Study food frequency questionnaire and was developed to reflect the eating habits of this age group.

Subjects The multiethnic sample consisted of 179 youths (ages 9 to 18 years) who completed the questionnaire twice, 1 year apart.

Statistical analysis Pearson correlation coefficients were calculated on nutrient data adjusted in relation to energy intake and on unadjusted food data.

Results Reproducibility for nutrients ranged from .26 for protein and iron to .58 for calcium, and for foods it ranged from .39 for meats to .57 for soda. Mean reproducibility was higher among girls than boys for energy and nutrients and for foods; no consistent pattern was observed for age.

Conclusion A self-administered food frequency questionnaire has a reasonable ability to assess the eating habits of older children and adolescents over time. J Am Diet Assoc. 1995; 95:336-340.

Section snippets

Development of Questionnaire

The youth/adolescent questionnaire (YAQ) was developed from a preliminary semiquantitative food frequency questionnaire for children, which was designed in 1986 according to the same format as the validated, adult Nurses’ Health Study frequency food questionnaire (1), (10). The original children's questionnaire was self-administered, and had 116 food items. It was modified from the adult version to include ice cream sundaes, milk shakes, candy bars, corn chips, and onion rings.

We pilot-tested

Results

Our study sample consisted of 179 children and adolescents who completed the YAQ twice, 1 year apart. Ages ranged from 9 to 18 years (mean=14 years); 44% were boys and 56% were girls. The youths who completed both YAQs were slightly younger and slightly heavier at the time of the first completion than the 51 youths who completed only the first YAQ. (These differences were not significantly different; P >.05.) The responders to both YAQs had a higher percentage of boys, Asian-Americans, and

Discussion

Reproducibility of the YAQ for energy-adjusted nutrients ranged from .26 for protein and iron to. 58 for calcium. The reproducibility was higher for girls than for boys, particularly for total energy, total fat, fiber, and calcium. We observed no consistent pattern of correlation between age and nutrients. Across age and gender, protein and iron had the lowest reproducibility.

When serving frequency was assessed, the reproducibility of the YAQ ranged from .57 for soda to .39 for meats. The

Applications

The self-administered instrument described in this article shows reproducibility in a range observed for similar food frequency questionnaires used with the adult population and performs consistently down to age 10 years. Thus, the YAQ may be used in school-based settings or in community studies of diet, lifestyle, and health. Further studies that compare the YAQ with multiple 24-hour recalls would allow quantification of measurement error when this instrument is used with adolescents from a

References (24)

  • E.M Pao et al.

    Foods Commonly Eaten by IndividualsAmount per Day and per Eating Occasionit>

    (1982)
  • Food and Nutrient Intakes by Individuals in the United States, 1 Day, 1987-88

    (1993)
  • Cited by (427)

    • The Association of Residential Mobility With Weight-Related Health Behaviors

      2022, Journal of Nutrition Education and Behavior
      Citation Excerpt :

      Fruit and Vegetable Consumption was measured as daily servings of fruit and vegetables. The YAQ was used to measure servings of fruit and vegetables at wave 2.20,21 The FFQ was used to measure servings of fruit and vegetables at wave 3 and 4.22

    View all citing articles on Scopus
    View full text