*Geek Box: Measuring Sodium in Nutrition Research
It is critical to understand sodium measurements in nutrition research in order to contextualise some of the findings in different studies. Sodium is a nutrient that tends to correlate quite poorly from dietary assessment methods such as food frequency questionnaires and other recall methods.
However, over up to 95% of ingested sodium is excreted in urine, which means that urinary sodium excretion has the potential to be an excellent biomarker for dietary intake of sodium. But this potential comes with a catch; to estimate actual daily 24 h intake of sodium would require collecting all urine excreted by an individual on that day. This is clearly burdensome on both investigators and participants, as well as costly to research.
Not only is this burdensome, however, there is also another catch, which is that sodium excretion is also highly variable from day-to-day in the same individual, i.e., within-person variation. This means that if only one single 24 h urine collection is taken, it will be unrepresentative of average [mean] intake in that individual. With one single 24 h urine collection estimates of sodium intake can vary by between 30–50% between individuals in a study.
As a result, researchers have employed other methods. For example, a “spot sample” is where the investigators take a single urine sample at a random point during the day – this will also vary widely from the true intake for that day, and therefore introduce a significant measurement error.
From such single “spot samples”, researchers have attempted to construct mathematical formulas for estimating 24 h intake. Two widely used formulas are the Tanaka and Kawasaki formulas, respectively. However, these formulas are generally inaccurate and produce overestimations of risk at lower levels of estimated sodium and underestimations of risk at higher levels of estimated sodium.
The issue of sodium measurement is at the heart of the apparent controversy in the literature on sodium and CVD. To increase the reliability of estimates of sodium intake, multiple 24 h collections are required, with three in the same person being desirable as a minimum. To visualise and hear further discussion of the importance of the method of measuring sodium, I recommend you watch this Research Lecture.