An assessment of the test–retest reliability of the New Nordic Diet score

BACKGROUND There is a growing interest in the New Nordic Diet (NND) as a potentially health promoting, environmentally friendly, and palatable regional diet. Also, dietary scores are gaining ground as a complementary approach for examining relations between dietary patterns and various health outcomes. A score assessing adherence to the NND has earlier been published, yet not tested for reliability. OBJECTIVE To assess the test-retest reliability of the NND score in a sample of parents of toddlers, residing in Southern Norway. DESIGN A questionnaire survey was completed on two occasions, approximately 14 days apart, by 67 parents of toddlers [85% females, mean age 34 years (SD=5.3 years)]. The NND score was constructed from 24 items and comprised 10 subscales that summarize meal pattern and intake of typical Nordic foods. Each subscale was dichotomized by the median and assigned values of '0' or '1'. Adding the subscales yielded a score ranging from 0 to 10, which was further trichotomized. Test-retest reliability of the final NND score and individual subscales was assessed by Pearson's correlation coefficient and Spearman's rank correlation coefficient, respectively. Additionally, cross tabulation and kappa measure of agreement (k) were used to assess the test-retest agreement of classification into the NND score, and the subscales. RESULTS Test-retest correlations of the NND score and subscales were r=0.80 (Pearson) and r=0.54-0.84 (Spearman), respectively, all p<0.001. There were 69% (k=0.52) and 67-88% (k=0.32-0.76) test-retest correct classification of the trichotomized score and the dichotomized subscales, respectively. CONCLUSION The NND score and the 10 subscales appear to have acceptable test-retest reliability when tested in a sample of parents of toddlers.