Back

Validity and reproducibility of a culturally tailored dietary screening tool for hypertension risk in Nigerian healthcare

Batubo, N. P.; Auma, C. I.; Moore, J. B.; Zulyniak, M. A.

2024-01-24 cardiovascular medicine
10.1101/2024.01.24.24301732 medRxiv
Show abstract

Food frequency questionnaires (FFQ) are commonly used tools in dietary assessment but require validation. This study aimed to assess the relative validity and reproducibility of a culturally tailored FFQ for estimating food intake among Nigerian adults in clinical settings. The FFQ was administered to 58 patients at the Rivers State University Teaching Hospital, Nigeria, on two occasions, two weeks apart. Three repeat non-consecutive 24-hour dietary recalls (24DR) were also conducted as a reference method to evaluate the validity of the FFQ. Spearmans rank correlations, Wilcoxon signed-rank tests, cross-classification agreement, intraclass correlation coefficients (ICC), and Bland-Altman analysis were performed in R to evaluate the relative validity and reproducibility. The trial was registered with ClinicalTrials.gov: NCT05973760. The correlation coefficient (rs) between the FFQ and 24DR ranged from 0.20 for fats and oils to 0.78 for vegetables, with an average rs of 0.60 (p<0.05). The Wilcoxon signed-rank tests indicated no significant differences in the 19 food groups queried (p>0.05), except for fats and oils (p<0.05). The exact agreement between FFQ and 24DR for classifying individuals into quartiles ranged from 17% for salt to 88% for processed meats and alcoholic drinks, with 90% of individuals classified into the same or neighbouring quartile. Additionally, the Bland-Altman analysis demonstrated acceptable agreement, with > 96% of observations within the acceptable limits of agreement for all food groups. For reproducibility, the ICC ranged from 0.31 for stew to 0.98 for fruit, with an average ICC of 0.77 between the FFQs delivered two weeks apart. These data demonstrate good agreement between our culturally tailored FFQ and 24DR, and moreover, robust reproducibility for quantifying intakes of food groups associated with hypertension in Nigeria. This confirms that this novel FFQ is a valid and reliable tool for assessing the intake of key food groups among Nigerian adults.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.