Abstract
Background: Cardiovascular diseases are the leading cause of death globally, with hypertension and high sodium intake being major contributors. Accurate estimation of sodium intake is essential, but 24 h urine collection, the gold standard, is cumbersome and impractical for routine clinical use. Existing spot urine-based prediction formulae lack accuracy at the individual and population level. Objective: To develop and validate population-specific formulas for estimating 24 h urinary sodium excretion from spot urine samples using data from a representative Swiss adult population. Methods: Models with and without urea and potassium were developed incorporating age, sex, and anthropometry-based, population-specific, estimated urinary creatinine excretion values. Data quality was rigorously controlled, and model performance was compared to the INTERSALT, Kawasaki, and Tanaka formulae and to a nocturnal timed urine collection used to calculate hourly creatinine excretion. Results: Models based on first morning urine demonstrated improved accuracy (AUCs: Swiss anthropometric model 0.85 (95% CI: 0.80-0.90), Swiss anthropometric model with urea 0.86 (95% CI: 0.81-0.91)) and lower bias (-5.5 mmol/24 h for the Swiss anthropometric model and -2.86 mmol/24 h for the Swiss anthropometric model with urea) compared to existing equations. Performance was consistent across clinically relevant sodium intake thresholds and the models were therefore suitable for use in clinical settings. A timed nocturnal urine collection further improves accuracy. Conclusions: These new simple and reliable formulae provide a promising and practical tool for estimating sodium intake from first morning urine spot in adult European populations, and are potentially applicable in clinical settings.