Predictors for misreporting sodium and potassium intakes by overweight and obese participants in a food-based clinical trial: implications for practice.

This study compared self-reported sodium and potassium intakes with urinary biomarkers and identified predictive factors. Secondary analysis of the 3-month intensive phase of the HealthTrack study with control (C), interdisciplinary intervention (I), intervention plus 30 g walnuts/day (IW) arms (n = 149). Dietary data was derived from diet history (DH) interviews and biomarker measures from urine. Urine-derived sodium (all, p = 0.000) and potassium (C: p = 0.011; I: p = 0.000; IW: p = 0.004) measures were significantly greater than self-reported intakes over the three months. Multiple linear regression showed body weight at baseline, body mass index (BMI) at baseline, and combined BMI at baseline and DH interviewer significantly negatively predicted the differences in sodium intake and excretion for C (β = -21.226, p = 0.016), I (β = -106.140, p = 0.002) and IW (F (9.530, 2df), p = 0.000), respectively. Where intakes of sodium and potassium are of interest in a trial, both reported intake and urinary biomarker measures are recommended.