ABSTRACT
This study assessed the validity of dietary balance scores (DBSs) by investigating the association between DBSs and nutrient adequacy (NA) in two Japanese populations. The participants were 65 community-dwelling Japanese from Tokushima Prefecture and 2,330 community-dwelling Japanese from Aichi Prefecture. Based on food frequency questionnaires or 3-day dietary records, we obtained 18 food groups. The NA score integrates nine beneficial nutrients and two nutrients that should be limited. We calculated four different DBSs: DBS1 consisted of five food groups (score range:0?20), DBS2 consisted of nine food groups (score range:0?36), DBS3 consisted of eight food groups (score range:0?32), and DBS4 consisted of 10 food groups (score range:0?40). Both the Spearman rank correlation coefficient with NA and the area under the receiver operating characteristic curve (AUC-ROC) for the nine beneficial nutrients were then estimated to test the performance of each DBS in predicting nutrient intake. The results showed that DBS1 and DBS4 were positively correlated with NA, while the AUC-ROC showed that DBS4 could moderately discriminate individuals with adequate intake levels of all nine nutrients. These findings suggest DBSs (especially DBS4) are useful in assessing dietary balance in middle-aged and older community-dwelling Japanese. J. Med. Invest. 70 : 377-387, August, 2023.
