group that demonstrated little evidence of dehydration. However, the AIs for sodium and potassium were not based on the median intakes of apparently healthy groups as there is a significant proportion of the population in the United States and Canada with hypertension and cardiovascular disease. Their AIs were based on experimental observations of intakes in small groups of people (frequently subjects in experimental trials) that appeared to meet the needs for a defined indicator of adequacy or functional outcome, such as lowered blood pressure. Thus, confidence in the assessment that the prevalence of inadequacy is low when median intake exceeds the AI will be less than it would be if the AIs represented the median intake of a healthy group (IOM, 2000).
Group median intakes below the AI, however, cannot be assumed to be inadequate. The reason for this is the same as discussed earlier for individuals, namely that when the distribution of requirements for the chosen criterion is not known, the adequacy of an intake below a “recommended level” cannot be determined. For example, although median intakes of potassium are considerably below the AI, no statements regarding the adequacy of potassium intakes of individuals in Canada and the United States can be made.