Agreement of Food Records and 24-Hour Urine Studies in Clinical Practice

The objective of this study was to determine the level of agreement between 3-day food records obtained as part of clinical care with 24-hour urine collections specifically assessing sodium, potassium, phosphorus, calcium, protein, and fluid intake.

Data were collected from patients at a nephrology clinic in a metropolitan, academic medical center. Patients who completed both a 3-day food record and a 24-hour urine collection were analyzed. Food record and urine collection measurements were compared using a simple ratio, Pearson's correlation, and general linear models.

Patients (n = 85) were 47.9 ± 15.2 years of age, 54% were female, with a mean serum creatinine of 1.3 ± 0.7 mg/dL and estimated glomerular filtration rate of 64.2 ± 25.6 mL/min. Patients had autosomal-dominant polycystic kidney disease (48.2%), nephrolithiasis (31.1%), chronic kidney disease (4.7%), or other genetic or cystic conditions impacting the kidney (12.9%). Nutrient intake was measured utilizing a 3-day food record. Food records and urine collections were compared using the values, correlations, and general linear models. Fluid intake demonstrated the highest agreement (ratio 1.01) and calcium demonstrated the least agreement (ratio 6.30). Significant correlations were demonstrated for phosphorus (r = 0.321, P = .003), magnesium (r = 0.256, P = .018), protein (r = 0.555, P < .000), and fluid (r = 0.277, P = .010) intake. Food record intake of potassium (P = .046), protein (P = .004), and fluid (P = .010) were significant predictors of 24-hour urine excretion.

3-day food records are useful tools to determine patient dietary patterns, but should be used with caution when assessing specific nutrient intake in clinical settings.