Construction and reproducibility of a questionnaire aimed for evaluation of dietary habits in physically active individuals.

The aim of this study was to develop a comprehensive food frequency questionnaire (FFQ), and a complementary general questionnaire associated to it, specifically tailored for the assessment of dietary habits among athletes and other physically active individuals in Croatia, and to determine the reproducibility of the questions and nutrient intake estimations. The study included 83 students (42 males and 41 females) from the Faculty of Kinesiology, University of Zagreb. Construction of the initial version of the questionnaire was done according to the literature and the results of a pre-testing interview conducted on 15 subjects. The final version of FFQ consisted of 74 questions on usual consumption frequency of particular food and beverage items and usual food preparation methods. The final version of the general questionnaire consisted of 30 questions on age, gender, body height, body weight, energy expenditure, specific dietary regimen, general dietary habits, diet characteristics before, during and after training, and cigarette smoking habits. Reproducibility of the questionnaires was assessed by a repeated administration of the questionnaire after one month. Pearson correlation coefficients and proportions of agreement were used to present reproducibility of quantitative and qualitative data, respectively. In the sample of male students, Pearson correlations ranged from 0.51 for copper to 0.95 for alcohol, with an average correlation of 0.68. In the sample of female students, correlations ranged from 0.30 for alcohol to 0.78 for calcium, with an average correlation of 0.60. Average reproducibility of items in the general questionnaire was 0.83 and 0.86 for male and female students, respectively. In conclusion, the newly developed FFQ enables cost-effective assessment of dietary habits in Croatian athletes and other physically active individuals, as well as demonstrates acceptable reproducibility.