Assessment of sodium intake in epidemiological studies on blood pressure.

Previous studies suggested that seven to fourteen 24-hour urine collections are necessary to minimize the attenuation due to the large day-to-day variation in an individual's Na intake and due to inadequate numbers of 24-hour urinary collections. Difficulties and inconveniences in collecting such a large number of 24-hour urine specimens are likely to result in a relatively large proportion of people who will either refuse to participate in the study or will collect incomplete specimens. The latter problem is especially serious since no satisfactory criteria are currently available to determine the completeness of urine specimen. There is no perfect solution to the problem of feasibility of multiple 24-hour specimens. Suggestions given in this paper can only help to partially solve the problem. The average Na/K and Na/creatinine ratios calculated from multiple, timed overnight urine specimens deserve some attention in future research.