Our primary outcome was estimated dietary sodium intake, which can be calculated based on 24-hour urinary sodium excretion. We compared three previously published methodologies for estimating 24-hour sodium excretion by using spot urine samples (formulas shown in Supplementary Table 1) [35 (link)38 (link)39 (link)40 (link)].
The first two formulas are based on the principle that the ratio of sodium to creatinine in the spot urine sample is proportionate to the ratio excreted in 24 hours, so an estimated 24-hour creatinine excretion (based on various demographic factors) can be used to ‘scale up’ the results from the spot urine sample, as follows [35 (link)]:
The Kawasaki method was developed in 1993 based on a sample of 159 healthy Japanese adults (age 20–79 years) [38 (link)]. It uses sodium and creatinine concentrations from a second morning urine sample and estimates 24-hour creatinine excretion based on age, sex, height, and weight. The Tanaka method was developed in 2002 based on a sample of 591 Japanese adults (age 20–59 years) from the INTERSALT study (data collected in 1987–1988) [39 (link)]. It uses sodium and creatinine concentrations from a casual urine sample (i.e., time of day not specified) and estimates 24-hour creatinine excretion based on age, height, and weight.
The third formula uses a different methodological approach. The INTERSALT method was developed in 2013 based on a sample of 2,948 participants (age 20–59 years) from a variety of North American and European sites in the INTERSALT study (data collected in 1984–1987) [40 (link)]. Unlike the other formulas, INTERSALT is based on a regression analysis that uses urine sodium and creatinine concentrations from a casual urine sample, age, sex, and BMI as independent variables to predict 24-hour urine sodium excretion. This eliminates the step of estimating 24-hour urine creatinine excretion.
All three methods provide an estimate of 24-hour urine sodium excretion. Dietary sodium intake can be calculated by dividing the urine sodium excretion by 0.9, based on the assumption that 10% of sodium intake is lost through sweat and feces, and thus urinary excretion accounts for 90% of intake [41 (link)42 (link)]. All results are reported here in terms of dietary sodium intake; if desired, dietary salt intake can be calculated by multiplying the dietary sodium intake by 2.5 [42 (link)].
After comparing the estimated population mean across all three estimation methods, we used the Kawasaki formula for subsequent analyses in this paper. This decision was based on a recent large-scale study suggesting that the Kawasaki formula may be most accurate in non-European populations [33 (link)].
The first two formulas are based on the principle that the ratio of sodium to creatinine in the spot urine sample is proportionate to the ratio excreted in 24 hours, so an estimated 24-hour creatinine excretion (based on various demographic factors) can be used to ‘scale up’ the results from the spot urine sample, as follows [35 (link)]:
The Kawasaki method was developed in 1993 based on a sample of 159 healthy Japanese adults (age 20–79 years) [38 (link)]. It uses sodium and creatinine concentrations from a second morning urine sample and estimates 24-hour creatinine excretion based on age, sex, height, and weight. The Tanaka method was developed in 2002 based on a sample of 591 Japanese adults (age 20–59 years) from the INTERSALT study (data collected in 1987–1988) [39 (link)]. It uses sodium and creatinine concentrations from a casual urine sample (i.e., time of day not specified) and estimates 24-hour creatinine excretion based on age, height, and weight.
The third formula uses a different methodological approach. The INTERSALT method was developed in 2013 based on a sample of 2,948 participants (age 20–59 years) from a variety of North American and European sites in the INTERSALT study (data collected in 1984–1987) [40 (link)]. Unlike the other formulas, INTERSALT is based on a regression analysis that uses urine sodium and creatinine concentrations from a casual urine sample, age, sex, and BMI as independent variables to predict 24-hour urine sodium excretion. This eliminates the step of estimating 24-hour urine creatinine excretion.
All three methods provide an estimate of 24-hour urine sodium excretion. Dietary sodium intake can be calculated by dividing the urine sodium excretion by 0.9, based on the assumption that 10% of sodium intake is lost through sweat and feces, and thus urinary excretion accounts for 90% of intake [41 (link)42 (link)]. All results are reported here in terms of dietary sodium intake; if desired, dietary salt intake can be calculated by multiplying the dietary sodium intake by 2.5 [42 (link)].
After comparing the estimated population mean across all three estimation methods, we used the Kawasaki formula for subsequent analyses in this paper. This decision was based on a recent large-scale study suggesting that the Kawasaki formula may be most accurate in non-European populations [33 (link)].
Free full text: Click here
