Abstract
This cross-sectional study evaluated the validity of three alternative methods compared to the gold standard 24-h urine collection for estimating dietary sodium intake, a modifiable risk factor for hypertension, among middle-aged and older adults with elevated blood pressure. These included spot urine collection (using Kawasaki, Tanaka, and INTERSALT equations), 24-h dietary recall, and food frequency questionnaire responses, compared to 24-h urine collection in a subset of 65 participants (aged 50-75 years, 58.5% women, 61.6% hypertensive) from the DePEC-Nutrition trial. The validity of the methods was assessed using bias, the Spearman correlation coefficient (SCC), the intraclass correlation coefficient (ICC), and Bland-Altman analysis. Among the alternative methods, spot urine collection using the Kawasaki equation showed the strongest correlation (SCC 0.238; ICC 0.119, 95% CI -0.079 to 0.323), but it exhibited a significant bias (1414 mg/day, p-value < 0.001) relative to 24-h urine collection. Conversely, dietary surveys had a smaller bias but wider limits of agreement. These findings underscore the complexities of accurately estimating dietary sodium intake using spot urine collection or dietary surveys in this specific population, suggesting that a combination or the refinement of existing methodologies might improve accuracy. Further research with larger samples is necessary to develop more reliable methods for assessing sodium intake in this high-risk group.