NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Research Council (US) Subcommittee on Criteria for Dietary Evaluation. Nutrient Adequacy: Assessment Using Food Consumption Surveys. Washington (DC): National Academies Press (US); 1986.

Cover of Nutrient Adequacy

Nutrient Adequacy: Assessment Using Food Consumption Surveys.

Show details

6Assessing Excessive Intake and Nutrient Energy Ratios

Excessive intake can present serious health problems. The subcommittee therefore discussed factors of major concern with regard to overconsumption and explored the use of the probability approach to assess excessive intakes.

Fat Intake

Approaches to adjusting the distribution of observed intakes to estimate the distribution of usual intakes can be applied to fats, which are of concern because of a possible effect of high intakes on serum lipids and coronary heart disease. Through the use of such approaches, subgroups can be compared or trends examined over time. Moreover, the distribution of intakes can be compared with recommended intakes promulgated by various groups. This technique may also be adequate for establishing whether overall population intakes should be increased or decreased. It would be desirable to assess prevalence of excess fat intake in such a manner in light of current concerns.

Present knowledge is inadequate for the subcommittee to offer specific guidance on assessing the intake of fat. However, it has considered a construct for assessing excessive intake of fat, carbohydrate, or indeed any nutrient. This construct, which is presented later in this chapter, can be used when data portraying the probability of excessive intake have been assembled in a manner similar to that used to assess probability of inadequate intake.


The U.S. public's interest in nutrition appears to have shifted in recent years from concern only about deficiency diseases to concern about both inadequate consumption and overconsumption. The number of dietary components of interest has also grown from only those whose intake is required for good health to those whose intake is somehow related to maintenance of optimal health. Included in the latter are nonnutritive compounds that are related to the onset and development of such diseases as cancer and cardiovascular disease. Many of these compounds have not traditionally been listed in food composition tables. The subcommittee suggests that future tables include listings of the concentrations of those compounds known to enhance or retard the development of chronic diseases. It believes that such additions to the food composition tables will significantly enhance the ability of the USDA to respond to anticipated questions from the U.S. public about the intake of food components and will provide the U.S. population with much better information on the adequacy and safety of its diet.

Assessment of the Prevalence of Excessive Intake

In Chapter 5, the probability approach was discussed in relation to estimating the prevalence of inadequate nutrient intake. The main feature of this approach is its recognition of requirement variations among similar individuals that are taken into account in estimating the probability of adequacy of a particular intake level and generalizing to the population or subpopulation by summing the estimates of prevalence for each intake level.

The same principles and approach can be used to analyze detrimentally high usual intakes of nutrients or food components. The risk of acute toxicity cannot be estimated in this fashion from survey dietary intake data. Individual variation in response to a detrimental factor is analogous to the variability of nutrient requirement, which has already been discussed. If a mean level of excess intake can be derived, it can be analyzed with a probability approach similar to the analysis of mean requirement. Thus, a distribution of intakes that would be deemed excessive can be conceptualized. If this distribution can be described or estimated, the probability approach can be applied in the same manner as described for inadequate intake, except that attention will be focused on the upper end of the adjusted intake distribution and the lower end of the requirement distributions.

The concepts embodied in this approach have been presented by two recent committees dealing with nutrient requirements (FAO/WHO/UNU, in press; Health and Welfare, Canada, 1983) and are illustrated in Figure 6-1. The left-hand curve of this figure is identical in concept and derivation to the curve presented in Figure 5-1 to portray the probability of inadequacy of protein intake. In Figure 6-1, the recommended intake is marked on the curve indicating the level of intake associated with a very low probability of inadequacy for a randomly selected member of the population. The curve on the right is meant to portray the analogous probability that a particular level of intake will be detrimental to the randomly selected individual. A point marking low risk is shown. Both this point and the recommended intake might be considered safe, i.e., intakes representing an acceptably low risk or probability of either inadequacy or detrimental excess. The range of intakes falling between these two points can be regarded as a safe range of intakes for individuals. By altering the definition of acceptable risk or probability, the range will be extended or contracted.

Figure 6-1.. The concept of a safe intake range.

Figure 6-1.

The concept of a safe intake range. Since there is individual variability in both requirement for a nutrient and tolerance for high usual intake, the risk or probability curves for inadequacy and for excess may be described as in Figure 5-1. The safe (more...)

Two implications and applications of these concepts are relevant to the present report. Consider first the presence of detrimental factors in foods. By using the right-hand curve, it is theoretically possible to estimate the prevalence of excess intakes in a manner directly parallel to that described for estimating the prevalence of inadequate intakes. Adjustment of intake distributions to eliminate the effects of day-to-day variation would be carried out in the same manner as described above. The prerequisite for applying the probability approach would be a description of, or judgment about, the frequency distribution of intakes that are detrimental for individuals in the population. The application of the probability approach in this area of research is at present constrained by the lack of attention to the examination of this distribution. Bearing in mind this constraint, the subcommittee recognizes that the probability approach can be used for excessive intakes as well as inadequate intakes whenever there is sufficient information on the distribution of excess intakes.

The approach can also be used to assess the appropriateness of nutrient intake when there is reason for concern about both inadequate and excessive intake. A specific and important example would be the assessment of fat intake in the NFCS survey. On one hand, inadequate intake of fat or fatty acids can result in specific fatty acid deficiencies and too low an energy density in the diet. On the other hand, excess fat intake can produce detrimental effects on serum lipids and has been implicated in the development of atherosclerosis. To determine the optimal level of fat intake for a population, one should take into account both risk curves portrayed in Figure 6-1. The application of this type of analysis in nutrition programming would logically be directed toward the encouragement of dietary intake within the safe intake range. This assessment is limited, however, by the absence of descriptions of the distributions of intake levels that would be inadequate or excessive—the two requirement distributions. Before there can be any scientific approach to the assessment of biologically appropriate or safe intake levels, these curves must be defined.

Energy from Protein, Fat, and Carbohydrate

A technical approach to defining criteria for protein-to-energy ratios has been presented by the Committee on Energy and Protein Requirements of FAO/WHO/UNU (in press). This approach requires knowledge of the distribution of protein requirements, of energy requirements, and of the correlation between the two. The same technical approach could be applied to fat and carbohydrate, if requirement distributions become known (PAO/WHO/UNU, in press).

The major problem in applying nutrient energy ratios is that the distribution of energy requirements changes with the distribution of physical activity characteristic of the population or population subgroup (FAO/WHO/UNU, in press). In theory, this might mean that a distribution of the ratio of nutrient to energy requirements (nutrient density) should be determined for each population subgroup based on a distribution of energy requirements for that subpopulation. For this reason, the subcommittee sees no advantage in assessing the biological suitability of observed nutrient density in population studies of dietary intake and offers no guidelines for this type of evaluation, because more information about requirements is required to apply this approach than for the probability approach.

The concept of nutrient density, which relates nutrients to energy or volume of food, may be useful for other purposes, such as in considering the nutritional quality of individual foods or in providing prescriptive recommendations for diet modifications. For community diagnoses or needs assessment, however, there is no apparent advantage to examining nutrient-to-energy ratios rather than nutrient intake, except when energy is a determinant of requirement for a nutrient as for thiamin. These other applications of nutrient density ratios do not fall within the scope of the mandate assigned to the subcommittee.

If it can be assumed that activity levels and anthropometric status are to remain constant, the distribution of usual energy intake is taken to describe approximately the distribution of energy requirements. If, however, observed energy intake describes the distribution of energy requirement, there will be no purpose to the use of this information for population assessment since this same distribution serves as the intake distribution. The final assessment would be identical with that obtained by assessing observed protein intake per kilogram of body weight. Although the same principles would hold for assessment of energy derived from fats or carbohydrates, the situation is more difficult because no descriptors of the requirement distribution are available.

In view of the other uses of nutrient density information and the increasing popularity of this method of describing the diet of a population, the subcommittee recognizes that the USDA may wish to publish descriptive information about the energy derived from fat, carbohydrate, and protein. For this purpose it suggests that centile distribution may be appropriate. Such centile distributions should only be generated after the intake distributions have been adjusted to remove the effects of day-to-day variations.

The interpretation of such information in relation to epidemiological studies or similar studies requires caution, because fats, protein and carbohydrates together with alcohol make up the total energy intake, and if the intake of one decreases there must be compensatory increases in the intake of one or more of the others. To interpret such information, one should take into account the covariances among these macronutrients, which will not be apparent from simple centile distributions of the individual macronutrient-to-energy ratios.

Use of the Probability Approach to Assess Energy Intake

As was discussed in Chapter 3, the probability approach to analyzing dietary adequacy requires either a knowledge of the joint distribution of dietary intake and requirement or an assumption of independence. For the nutrients, it is reasonable to assume that intake is determined to a very large extent by psychosocial factors affecting the selection of particular foods rather than by physiological factors. Studies in animal models indicate, however, that there are specific regulatory mechanisms for nutrient intake and provide limited evidence that these mechanisms continue to operate as an important factor in determining nutrient intake by free-living subjects. Therefore, it is reasonable to assume independence of intake and requirement for nutrients, provided the obvious factors affecting both intake and requirement (e.g., age, sex, and major differences in body size) have been taken into account.

Energy intake and expenditure are highly correlated among individuals, and regulatory mechanisms adjust intake to expenditure or adjust expenditure to intake (PAO/WHO/UNU, in press). The probability approach cannot be used to assess dietary energy intake and to interpret NFCS data on energy intake until specific knowledge has accrued on the magnitude of this correlation. In the assessment of dietary adequacy, therefore, energy intake must be interpreted in a manner completely different from that used for nutrient intake data.

For example, since activity is a major variable of energy expenditure, observed energy intake may be used as a descriptor of the usual expenditure levels and of status quo activity levels (Beaton, 1983). Examination of the energy status of a population requires measures of energy stores, such as anthropometric measurements, which are collected in the United States by the National Health and Nutrition Examination Survey.

Image img00059
Copyright © National Academy of Sciences.
Bookshelf ID: NBK217541


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (2.8M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...