BACKGROUND Despite its clinical relevance, sodium intake is seldom monitored by physicians, largely because of shortcomings of the 24-h urine collection for sodium excretion. In a prior study, sodium excretion was shown to be accurately estimated from a late afternoon/early evening spot urine sodium/creatinine ratio, adjusted for 24-h creatinine excretion. In this study, we assessed a more convenient and inexpensive method, using chloride and creatinine dipsticks. METHODS Subjects submitted 24-h urine collections along with an "AM sample," collected at the beginning, a "PM sample" collected in the late afternoon/early evening before dinner, at roughly the midpoint, and a "random sample," collected after completion, of the 24-h collection. Predicted 24-h sodium excretion was then determined from the spot urine dipstick chloride/creatinine ratio, measured by two independent observers, and from the spot urine laboratory sodium/creatinine ratio. Both ratios were adjusted for 24-h creatinine excretion. RESULTS For PM samples, predicted sodium excretion correlated strongly with actual 24-h sodium excretion, both for the dipstick method (r = 0.71; observer 1 and r = 0.65; observer 2; both P < 0.001), and the laboratory method (r = 0.86, P < 0.001). PM samples also differentiated subjects with sodium excretion <100 mEq/day vs. > or =100 mEq/day (sensitivity and specificity: dipstick method: 83 and 82%, respectively for observer 1, 89 and 77%, respectively, for observer 2; laboratory method: 100 and 82%, respectively). AM samples and random samples correlated less strongly. CONCLUSIONS The dipstick method appears promising as a convenient and inexpensive means to serially assess sodium excretion.