Caballero B. (ed.) Encyclopaedia of Food Science, Food Technology and Nutrition. Ten-Volume Set

Подождите немного. Документ загружается.

(RQ). The RQ is the ratio of CO

expired to O

consumed during a defined period. The theoretical

RQ values are known for fat, carbohydrate, and pro-

tein oxidation, making it possible to determine total

fat, carbohydrate, and protein oxidation. When pro-

tein oxidation is estimated from nitrogen excretion,

fat and carbohydrate oxidation can be calculated as

follows:

NonproteinRQ ¼ 1:00  carbohydrate oxidation

þ 0:707  fat oxidation :ð14Þ

However, caution is needed when estimating fuel oxi-

dation from nonprotein RQ values, since the RQ

itself can vary considerably depending on the fat

and protein mixture being oxidized. In addition to

the problem of obtaining an accurate picture of fat

and carbohydrate oxidation from nonprotein RQ

values, it is also not possible to distinguish between

the oxidation of carbohydrate þ fat and the oxidation

of short-chain fatty acids produced by carbohydrate

fermentation. When measuring the apparent metab-

olizable energy value of a macronutrient that repre-

sents a small proportion of the total diet, there is a

high potential for error when macronutrient oxida-

tion profiles are measured. However, theoretical cal-

culations show that there is much less error associated

with measuring energy loss.

0025 An example of the use of indirect calorimetry for

measuring the metabolic energy of completely fer-

mentable substances is the determination of the net

energy value for lactitol, a disaccharide sugar alcohol

(4-O-(b-d-galactopyranosyl)-d-sorbitol). The total

energy balance was measured during the experimen-

tal period in the calorimeter. The retained energy of

each subject was then calculated as:

RE ðretained energy Þ¼IE

ðfecal E þ urine E

þ gaseous E þ expended EÞ:ð15Þ

The energy needed to maintain energy balance

(metabolizable energy for maintenance; ME

) could

then be determined as:

¼ metabolizable energy intake þ negative RE

 1 :11  positive RE :ð16Þ

The factor 1.11 represents the cost of depositing body

energy and was measured in an earlier experiment.

The difference in ME

on a diet supplemented with

sucrose versus one supplemented with lactitol is a

measure of the energy lost from lactitol, and can be

expressed as the change in ME

adjusted for the

metabolic weight of the individuals (weight

0.75

ME

¼ðME

=weight

0:75

lactitol

ðME

=weight

0:75

sucrose

, ð17Þ

and the NME can then be calculated as:

NME ¼ ME

part

ME

=ðIS =weight

0:75

lactitol

ð18Þ

where ME

part

is obtained from eqn (10), and IS is the

ingested mass of the supplement (lactitol). The ME

values are divided by the metabolic weight to account

for differences in metabolic mass. Using this pro-

cedure, a value of 2.4kcal g

1

(10 kJg

1

) for lactitol

was obtained. Note that some measure of the energy

expenditure is required in this experiment (eqn (15)).

The authors relied on diaries and ankle actometers

and corrected to equal actometer counts. They ac-

knowledged that this was not very satisfactory;

there is a potential 15% error in the final net energy

value for lactitol.

Metabolizable Energy Value of Dietary

Fiber in Mixed Diets

0026Up to this point, this article has focused on undigested

and fermented food components as a general group of

substances. This section will deal specifically with

dietary fiber, a partly fermented food component

that passes through the small intestine unacted upon

by human digestive enzymes. Dietary fiber is a mix-

ture of nonstarch polysaccharides and lignin that are

structural components of plant cell walls.

0027As indicated above, it is not appropriate to use

metabolizable energy values for dietary fiber since

many more energy losses are associated with fiber

digestion as compared with the digestion of fat,

carbohydrate, and protein. Equations have been de-

veloped to take account of these extra losses and

provide a method to calculate the NME value of a

fermentable dietary component when the fecal energy

losses are known. These equations show that ap-

proximately one half of the potential energy is avail-

able to the host when undigested carbohydrates are

completely fermented. This value is widely accepted.

0028Knowledge of the energy yield from the fermented

portion of dietary fiber is not sufficient in and of itself

to allow the calculation of dietary fiber NME since a

significant proportion of the ‘fermentable’ dietary

components are not fermented but appear in the

feces. In addition, some ‘extra’ energy (factor b, eqn

(13)) may be lost during digestion because of the

intimate physical interaction between dietary fiber

and the food matrix (See Dietary Fiber: Properties

and Sources). The aim of these studies is to determine

a general NME value for dietary fiber that can be

used to calculate the metabolizable energy of mixed

human diets. To overcome these problems, two

solutions have been employed. The first solution in-

volved determining individual digestibility factors for

1856 DIETARY FIBER/Energy Value

specific foods and using these to calculate metaboliz-

able energy intake. This method has the advantage

that macronutrient digestibility is measured directly

for specific foods and multiplied by the specific heats

of combustion when available. In this method, the

calculated digestible energies of the diets are close to

the measured values for mixed diets since the final

value reflects the sum of the individual components.

This method has some problems. In the system of

Merrill and Watt, the digestibility of carbohydrates

was determined by difference, so accurate values for

carbohydrate digestibility are lacking. This also

means that a digestibility value for the fermentable

food component portion of the carbohydrate fraction

(as determined by difference) is lacking. It is, there-

fore, not possible to calculate the metabolic energy of

a food of known composition since values for un-

digested and fermented components are not available.

This limits the usefulness of this system to foods

that have been measured. In addition to this problem,

the NME value is appropriate for undigested and

fermented food components, whereas the digestible

energy value is not. Thus, this system does not

allow any correction for energy lost when food

components are resistant to digestion but fermented

in the lower gut.

002 9 The second solution is to use an average digestibil-

ity value determined by examining typical mixed

diets. This can be calculated by two separate

methods: using a factorial approach or using an em-

pirical model. Factorial models relate total protein,

fat, carbohydrate, and dietary fiber intake to digest-

ible energy through an equation of the type:

ME ðkca Þ¼4:0  protein ðgÞþ9:0  fat ðgÞþ3:75

 carbohydrate ðgÞþNME

 dietary fiber ðgÞ:ð19 Þ

In the development of eqn (19), carbohydrate was

determined directly (not determined by difference)

and is expressed as monomeric units. Dietary fiber

must also be measured directly through an approved,

reproducible method that is not affected by food

processing (See Dietary Fiber: Determination). The

factors 4 (protein) and 9 (fat) are the general Atwater

factors derived from mixed diets. The value of 3.75

for carbohydrate is slightly lower than the general

Atwater factor of 4. This reflects the fact that the

carbohydrate is expressed as the total monomeric

weight.

003 0 The value of NME

(the NME for dietary fiber)

comes from an analysis of the dietary fiber digestibil-

ity of mixed diets and the loss of energy to feces in

mixed Western diets. This value is determined as

follows. Firstly, the DE

part

value for dietary fiber is

obtained by comparing the digestible energies of

mixed diets to the metabolizable energy calculated

by eqn (19). The apparent digestible and metaboliz-

able energies for carbohydrate and fat are identical

since the macronutrient digestibilities were accounted

for, and urine and gaseous losses of these components

were negligible. Protein losses are corrected for urine

nitrogen and protein digestibilities. The digestible

energy for dietary fibre (DE dietary fiber) was esti-

mated by:

DE dietary fiber ¼H

 D

app

ð1  aÞ

¼ 4:1 kcal g

1

 0 :7  0 :7

¼ 2:0 kcal g

1

ð8:4kJg

1

Þ:ð20Þ

The apparent digestibility of dietary fiber is around

0.7. The value of factor a in eqn (20) needs to be

defined for dietary fiber. For sugar alcohols, the

value is approximately 0.2 (see above). A value of

0.3 has been adopted for dietary fibers in humans.

This value reflects the fact that the majority of energy

loss estimates lie between 0.2 and 0.4kcal per kcal of

fermented energy. As already mentioned, a value of

0.2 was obtained when measuring lactitol energy util-

ization in human subjects. It has also been suggested

that between 20 and 30% of the energy in undigested

and fermented carbohydrates is utilized by bacteria in

humans. A value of 0.2 kcal per kcal of fermented

material has been measured in rats fed an a-amyl-

ase-resistant pea starch and a value of 0.4kcal

per kcal of fermented material has been measured

in rats fed guar gum. In human diets with high intakes

of dietary fiber, an increased fecal energy output

of 0.3 and 0.39 kcal per kcal of energy fermented,

respectively, has been measured. Overall, these

data suggest that about 0.3kcal of energy is lost

to the feces per kcal of energy fermented. Thus,

1.0  0.3 ¼ 0.7kcal kcal

1

is retained.

0031As indicated above, the digestible energy value

does not give an accurate picture of the energy losses

that occur during fermentation of undigested food

components. Correction factors must be included

for energy losses to the heat of fermentation (factor

b), losses to fermentation gases (factor c), and the

relative efficiency of utilization of short-chain fatty

acids (factor d). Correcting the digestible energy for

these values gives:

NME

¼ H

 D

app

 0:54 ¼ 4:1kcalg

1

 0:7  0:54 ¼ 1:5kcalg

1

: ð21Þ

Eqn (21) shows that the NME value for dietary fiber

is approximately 1.5kcal g

1

. Eqn (21) applies strictly

to dietary fiber because the DH

value and digestibil-

ities were determined for dietary fiber. It is useful for

DIETARY FIBER/Energy Value 1857

calculating the metabolic energy values of mixed diets

because the digestibilities were determined for mixed

diets. It is comparable, therefore, with the general

Atwater factors of 4, 9, and 4 for carbohydrate, fat,

and protein.

0032 Empirical models represent another method for

determining total energy intake. Empirical models

are, in general, superior to the factorial models for a

number of reasons, the most important being that the

empirical models are much more robust to errors in

carbohydrate, fat, and protein determinations as well

as to the use of general Atwater factors. An example

of an empirical equation is as follows:

ME ¼ 0:96  total energy ðkcal Þ2:5kcal g

1

 dietary fiber ðgÞ12kcalg

1

 ingested nitrogen ðgÞ:ð22Þ

Eqn (22) relates the metabolizable energy to total

energy, protein, and dietary fiber only. The factor

0.96 represents the metabolizability of mixed diets

in the absence of dietary fiber. The value of 2.5kcal

1

represents the NME energy loss associated with

ingestion of dietary fiber. This value is, therefore the

difference between the heat of combustion of dietary

fiber and its NME

value (determined above):

2:5 kcalg

1

¼H

 NME

¼ 4:0kcalg

1

 1:5kcalg

1

: ð23Þ

0033 A version of Eqn (22) was tested for bias and ac-

curacy of prediction by comparison with 43 human

diets containing varied intakes and sources of dietary

fiber. The equation showed no bias over a wide range

of dietary fibre intake (2–93g daily). By necessity, this

test was performed by comparing DE

part

values be-

cause ME

part

values cannot be measured directly for

fermentable food components. This analysis provides

further evidence for a general NME for dietary fiber

of 1.5kcalg

1

Overall Conclusion

0034 Eating food components that are not digestible but

that are fermentable lowers the amount of metaboliz-

able and digestible energy available from food. This

occurs because of two different processes. Ferment-

able food components, like dietary fiber, are not

100% fermentable, and so the NME for these com-

ponents is lower than that of protein, carbohydrate,

and fat. In addition, fermentation itself is associated

with many different energy losses, including losses to

increased bacterial mass, the heat of fermentation,

fermentation gasses, as well as a lower efficiency of

host utilization of short-chain fatty acids.Because the

digestion of fermentable food components is so

different from that of macronutrients, special equa-

tions are needed to determine the net metabolizable

energy yield.

0036It is possible to utilize data from several sources to

obtain an estimate of the relative losses that occur

during fermentative digestion and back-correct to

obtain an unbiased estimate of the energy value of

dietary fiber. Taking into account all these energy

losses, a NME value for dietary fiber of 1.5kcalg

1

(6.3kJg

1

) has been calculated. This appears to ac-

curately predict energy losses associated with mixed

Western diets when analyzed using either factorial or

empirical models. Thus, the NME of 1.5kcalg

1

(6.3kJg

1

) can be used in a fashion similar to the

general Atwater factors of 4, 9, and 4 for protein, fat,

and carbohydrate to predict the metabolizable energy

of mixed Western diets. Individually measured NME

values for fiber in whole foods may differ from this

value since dietary fiber is fermented to different

extents and is intimately associated with the food

matrix, which may limit macronutrient digestibility.

Thus, individual measurements are needed for direct

comparison and may be required for specific types of

foods.

See also: Dietary Fiber: Properties and Sources;

Determination

Further Reading

r A (1990)Factorial calculation model for the estimation

of the physiological caloric value of polyols. In: Hosoya

N (ed.) Proceedings of the International Symposium on

Caloric Evaluation of Carbohydrates, pp. 209–257.

Tokyo: Research Foundation for Sugar Metabolism.

Bernier JJ and Pascal G (1990) Valeur e

nerge

tique des poly-

ols (sucres-alcools). Me

dicine et Nutrition XXVI: 221–

238.

British Nutrition Foundation (1990) Complex Carbo-

hydrates in Foods. The Report of the British Nutrition

Foundation’s Task Force, pp. 56–66. London: Chapman

& Hall.

Calloway DH and Kretsch MJ (1978) Protein and energy

utilization in men givena rural Guatemalan diet andegg

formulas with and without added oat bran. American

Journal of Clinical Nutrition 31: 1118–1126.

Cummings JH (1983) Fermentation in the human large

intestine: evidence and implications for health. Lancet

1: 1206–1208.

FASEB (1994) The Evaluation of the Energy of Certain

Sugar Alcohols used as Food Ingredients. Bethesda,

MD:LifeSciences Research Office, Federation ofAmeri-

can Societies for Experimental Biology (FASEB).

Goranzon H, Forsum E and Thilen M (1983) Calculation

and determination of metabolizable energy in mixed

diets to humans. American Journal of Clinical Nutrition

38: 954–963.

1858 DIETARY FIBER/Energy Value

Livesey G (1989) Procedure for calculating the digestible

and metabolizable energy values of food components

making a small contribution to dietary intake. Journal

of the Science of Food and Agriculture 48: 475–481.

Livesey G (1990) Energy values of unavailable carbo-

hydrate and diets: an inquiry and analysis. American

Journal of Clinical Nutrition 51: 617–637.

Livesey G (1991) Calculating the energy values of foods:

Towards new empirical formulae based on diets with

varied intakes of unavailable complex carbohydrates.

European Journal of Clinical Nutrition 45: 1–12.

Livesey G (1992) The energy values of dietary fibre and

sugar alcohols for man. Nutrition Research Reviews 5:

61–84.

Livesey G (1993) Comments on the methods used to deter-

mine the energy values of carbohydrates: dietary fibre,

sugar alcohols and other bulking agents. International

Journal of Food Sciences and Nutrition 44: 221–241.

Livesey G (1995) Metabolizable energy of macronutrients.

American Journal of Clinical Nutrition 62(supplement):

1135S–1142S.

Livesey G and Elia M (1988) Estimation of energy expend-

iture, net carbohydrate utilization, and net oxidation

and synthesis by indirect calorimetry: evaluation of

errors with special reference to the detailed composition

of fuels. American Journal of Clinical Nutrition 47:

608–628.

Merrill AL and Watt BK (1973) Energy Value of Foods.

Basis and Derivation. Agriculture Handbook #74,

Washington, DC: US Department of Agriculture and

US Department of Health and Human Services.

Mongeau R and Brassard R (1993) Enzymatic–gravimetric

determination in foods of dietary fibre as sum of insol-

uble and soluble fibre fractions: summary of collabora-

tive study. Journal of the AOAC International 76:

923–925.

Mongeau R, Brooks SPJ, Lampi BJ and Brassard R (1997)

Hard wheat bran and hard wheat bran fiber energy

values measured in rats after 6 and 16 weeks. In: Kritch-

evsky D and Bonfield C (eds) Dietary Fiber in Health

and Disease, pp. 267–289. New York: Plenum Press.

Nutrition Council of Holland (1987) The Energy Value of

Sugar Alcohols. Recommendations of the Committee on

Polyalcohols, The Hague, Netherlands.

van Es AJH, de Groot L and Vogt JE (1986) Energy bal-

ances of eight volunteers fed on diets supplemented with

either lactitol or saccharose. British Journal of Nutrition

56: 545–554.

Dietary Goals See Community Nutrition; Nutrition Policies in WHO European Member States

DIETARY REFERENCE VALUES

J Pollard, University of Leeds, Leeds, UK

Background to the Dietary

Recommendations in the UK

0001 The need to quantify aspects of the nations’ diet has

occupied scientists for many years. When one con-

siders that an adult eats about onetonne of food each

year, it is easy to see why nutrition has become such a

public health issue. The food we eat should provide

all the nutrients we need, yet in some countries, we

have an excess of food and in others insufficient. In

fact, it isonly relativelyrecently in evolutionary terms

that we have the variety of food in this country avail-

able today. For example, as recently as the 1930s, it

was estimated that up to one-third of households in

Britain could not afford an adequate diet. It is in this

context that dietary recommendations were first de-

veloped, to ensure that people were able to consume

an adequate diet. During the war years, recommen-

dations were developed with the objective of provid-

ingenoughfood to enablea work forcetoachieve and

maintain its productivity. However, as the science of

nutrition has developed, and the publics’ awareness

and understanding has increased, the objectives have

changed again towards one of maintaining health.

The recommended dietary intakes (RDIs) were pub-

lished in 1969 and were defined as ‘the amount suffi-

cient, or more than sufficient, for the nutritional

needs of practically all healthy persons in a popula-

tion.’ These recommendations were subsequently re-

placed by the recommended daily amounts (RDAs).

These were published in 1979, and were defined as

‘the average amount of the nutrient, which should be

provided per head in a group of people if the needs of

practically all members of the group are to be met.’

Both of these definitions attempt to make it clear that

the standardsare foruse in populationestimates only.

DIETARY REFERENCE VALUES 1859

In other words, consumption of nutrients at the level

set would be more than adequate for most people.

The problem with both these sets of recommenda-

tions was that they were misused. People using the

standards did not understand the proposed use, how

they had been derived, or how accurate they were

intended to be. People widely interpreted these stand-

ards as minimum amounts of nutrients that everyone

had to consume, rather than amounts that would

prove adequate for the majority. With this, then,

comes the interpretation that more is better, leading

to potential overdosing of some vitamins and min-

erals. Most of the confusion in using the standards lies

in the difference between the terms ‘recommenda-

tions’ and ‘requirements.’ People who did not under-

stand the previous standards used them to assess

requirements. A requirement is the amount of a par-

ticular nutrient that an individual must consume in

order to avoid deficiency. Deficiency may be defined

by clinical, physiological, biochemical, or socio-

logical criteria, and each of these may give different

values at an individual level for nutrient intake needs.

Requirements will differ from individual to individ-

ual. What the standards actually give us are recom-

mendations, that is the amount of a nutrient judged

to be the appropriate amount to maintain health in

virtually all individuals in that group.

0002 In 1987, the Committee on Medical Aspects of

Food Policy (COMA) convened a Panel to review

the 1979 RDAs. The result of this was the new rec-

ommendations, published in 1991, and entitled Diet-

ary Reference Values for Food Energy and Nutrients

for the United Kingdom. These new recommenda-

tions attempted to clarify the confusion in the usage

of the previous recommendations and covered more

nutrients, particularly the macronutrients, than the

previous reports. The Report also took the objective

of maintaining health one step further to try and

provide an ‘optimal diet’. An optimal diet can be

defined as one which ‘maximises health and longevity

and therefore, prevents nutrient deficiencies, reduces

risks of diet-related chronic diseases, and is composed

of foods that are safe and palatable’.

Dietary Reference Values – Definition and

Derivation

0003 The term ‘dietary reference values’ (DRVs) refers to a

number of different parameters that describe the dis-

tribution of the requirements for a particular nutrient

found within a population. When one considers that

people differ from one another in the amounts of

nutrients they need, it would be impossible to repre-

sent these differences with just one value, as the pre-

vious standards attempted to do. Therefore, when

COMA designed the DRVs, they came up with a

range of values for each nutrient to represent the

differing needs of the population.

0004The Panel used different criteria for the assessment

of nutritional adequacy for individual nutrients,

depending on what information was available and

on the physiological effects of different nutrients.

Evidence came from observational studies of healthy

populations, clinical studies of deficiency symptoms,

balance studies, or functional tests of a nutrient. The

Panel identified a key functional parameter for each

nutrient that was considered to be essential for healthy

life. That functional parameter was then used to meas-

ure the range of needs for that nutrient within a com-

munity. Functional parameters used for each nutrient

are shown in Table 1. The DRVs are derived from a

statistical association describing the variation around

the average seen when making measurements of a

large number of people. As there is little evidence on

the distribution of requirements in a population, the

Panel made the assumption that they are normally

distributed in almost all cases. Hence, the average

requirement for each nutrient was calculated, and

the upper and lower limits of its distribution could

be estimated. The DRVs are based on this distribu-

tion, as shown in Figure 1.

0005The average amount of a nutrient required is called

the estimated average requirement (EAR). The range

of normal consumption is taken as the EAR plus or

minus two notional standard deviations. The EAR

plus two standard deviations is called the reference

nutrient intake (RNI) and is defined as ‘‘an amount

of a nutrient that is enough for almost every individ-

ual, even someone who has high needs.’’ This is the

equivalent of the old RDAs. The EAR minus two

standard deviations is the lower reference nutrient

intake (LRNI) and is defined as ‘‘the amount of a

nutrient that is enough for only the small number of

people with low needs.’’ Statistically, we know that the

RNI indicates the amount of a nutrient sufficient to

meet the needs of up to 97.5% of the population.

Similarly, we known that only about 2.5% of the

population will have needs lower than the LRNI. So,

the requirements of about 95% of the population will

lie between the LRNI and the RNI. It should be

pointed out that the EAR is not a recommended intake

for an individual. It is the estimated average require-

ment of a nutrient throughout a large group of people:

some will need more than this, and some will need less.

0006For some nutrients, a reference value called the ‘safe

intake’ was assigned. This term was used if there was

not enough information to ascertain requirements.

This intake level was one that was judged by the

COMA Panel to be adequate for almost everyone’s

needs but not so large as to cause undesirable effects.

1860 DIETARY REFERENCE VALUES

Energy, Protein, Vitamins, and Minerals

0007 The RNI is designed to be adequate to meet the needs

of most individuals in a population, even those with

high needs. If a person with low needs consumed the

RNI for most nutrients, this excess would not be

harmful, but this is not the case with energy. If some-

one with low energy needs consumed excess, this

would lead to an undesirable weight gain. Therefore,

the Panel set energy standards at the average need,

knowing that some people will need more and some

will need less.

0008Safe intakes were set for pantothenic acid, biotin,

vitamin E, vitamin K, molybdenum, manganese,

chromium, and fluoride, as insufficient evidence was

available for DRVs.

Fats, Carbohydrate and Nonstarch Polysaccharides

(NSP)

0009As there are no specific deficiency signs or symptoms

for fats and carbohydrates, and no absolute require-

ment for any of these individually, population averages

have been derived from the likely effects that changes

in their average consumption could be expected to

have on certain health outcomes. These are given in

the Report as percentages of the EAR for energy that

they should, on average, provide. No ranges around

the EAR are given.

0010The DRVs for NSP are based on an estimated de-

sirable average intake, and the expected range of

individual intakes around that figure.

Lower

reference

nutrient intake

Reference

nutrient intake

Frequency distribution of

individual requirements

2 s.d. 2 s.d.

Estimated average requirement

Nutrient requirements

fig0001 Figure 1 Dietary reference values – definition. From Depart-

ment of Health (1991) Report on Health and Social Subjects 41.

Dietary Reference Values for Food Energy and Nutrients for the

United Kingdom. London: HMSO, with permission.

tbl0001 Table 1 Functional parameters used by the COMA panel to assess DRVs

Nutrient Functional parameter

Energy Energy expenditure

Essential fatty acids Prevention of essential fatty acid deficiency

Fat Changes from current intakes that would be expected to result in specific changes in physiological

and/or health outcomes, particularly in relation to coronary heart disease

Nonstarch Polysaccharides Healthy bowel function and stool weight

Nonmilk extrinsic sugars Effect on dental caries

Protein Nitrogen balance

Vitamin A Adequate body pool size, based on the amount of vitamin A in the liver

Vitamin B

(thiamin) Urinary thiamin output, thiamin loading, glucose loading, and transketolase activation

Vitamin B

(riboflavin) Measurement of urinary and red cell riboflavin; the determination of erythrocyte glutathione reductase

activation coefficient – a measure of riboflavin tissue saturation and long-term riboflavin status

Niacin Observational requirements to prevent or cure pellagra

Vitamin B

Estimates of body pool size and blood vitamin B

levels in depletion and repletion of adults on

controlled diets

Vitamin B

Serum vitamin B

levels

Folate Serum levels, red cell folate, and liver reserves

Vitamin C Plasma ascorbate levels and prevention of scurvy

Calcium Growth, calcium retention, urinary excretion, and absorption

Magnesium Magnesium balance

Phosphorus 1 mmol P: 1 mmol Ca

Sodium Sodium balance and balance of risks and benefits that might practically be expected to occur, given the

prevailing sociocultural environment

Potassium Amount needed for growth plus losses (fecal, urinary, via skin and hair)

Iron Estimated gut absorption and losses

Zinc Systemic needs plus losses (fecal, urinary, skin, hair, semen, and menstruation) and absorption

Copper Balance studies

Selenium Glutathione peroxidase activity

Iodine Prevention of deficiency symptoms of goiter

DIETARY REFERENCE VALUES 1861

Dietary Reference Values – Uses,

Limitations, and Assumptions

0011 The DRVs set out to reduce confusion over use, and in

order to do this, they deliberately used the term ‘ref-

erence’ in the title to try and convey the message that

these were not individual recommendations but a

point of reference only. The main proposed uses of

the DRVs were for food planners, health policy per-

sonnel, and the food industry, although many other

people may wish to refer to the reference values.

Assessing the Diet of Individuals

0012 DRVs, as with previous standards, are not designed for

assessment of individual diets. It is not possible to know

an individuals’ requirement for a particular nutrient

without measuring the functional parameter for that

nutrient. However, the DRVs, unlike the previous

standards, can tell us something about the likely ad-

equacy of an individuals nutrient intake. If a person’s

consumption of a particular nutrient is somewhere

between the LRNI and the RNI, it would not be pos-

sible for us to say whether or not that persons diet is

adequate, as we do not know whether that person has a

low, medium or high requirement. However, what we

do know is that if an individual’s consumption is above

the RNI, they are likely to be meeting their needs, but

the closer consumption gets to the LRNI, the more

likely it is that that person is not meeting their needs.

If a person is routinely consuming amounts of a nutri-

ent less than the LRNI, they are likely to be deficient,

but functional tests would be required to confirm this.

DRVs can only give us an indication of nutritional

adequacy in an individual, and great care needs to be

taken when using them in this manner.

Assessing the Diet of Groups of People

0013 When assessing the adequacy of diet in groups of people,

the RNI should be used where possible to ensure that the

risk of any deficiency in the group is small. When assess-

ing the adequacy of the diet of a group of people, some

assumptions have to be made. These are that the distri-

bution of intakes in the group is identical to the distribu-

tion of requirements andthat the people inthe groupwith

the highest requirements are generally those with the

highest intakes, also that the measured diet is typical of

habitual intake. Because of these assumptions, we are

unable to say with certainty that intakes below the

LRNI are inadequate, and, as with assessing the diets

of individuals, measurement of functional parameters is

required to identify deficiencies.

Planning Food Supplies

0014 When planning food supplies, it is necessary to ensure

that everyone (even those people with high require-

ments) obtains enough of every nutrient. Therefore,

the RNI should be used as a reference point. Even

though some people will be receiving in excess of their

requirements, the risk of deficiency will be minimal.

Nutrition Labeling

0015The nutrition information appearing on a food label

is used by individuals, and it would be inappropriate

therefore to use the RNI as a reference level. If the

RNI appeared on food packaging, many people might

try and consume that level of the nutrient, even

though they might have much lower requirements.

Although this may not be harmful, it may be expen-

sive and perhaps wasteful. Therefore, the EAR is used

as a reference in nutrition labeling, and the amount of

a nutrient in a product is expressed as a percentage of

the EAR. However, this still needs some interpret-

ation and is potentially misleading, as some people

will still need more than the EAR and some less.

Identifying ‘At Risk’ Subgroups

0016By using the LRNI as a reference value, ‘at risk’

groups in the population may be identified. If people

are routinely consuming less than the LRNI, they are

likely to be ‘at risk’ of deficiency. These people then

can be examined and their nutritional status assessed

using the appropriate functional parameter.

0017There are numerous assumptions that have to be

made, and therefore limitations to using DRVs.

When deriving the DRVs, the COMA Panel empha-

sized that there was insufficient information to be

able to set the RNI, EAR, and LRNI for most nutri-

ents with any great confidence. The Panel stressed

that some of the functional parameters used were

far from ideal and emphasized the need for further

research.

0018The DRVs only apply to healthy people and

may not be appropriate for use with some clinical

conditions or those with special requirements.

DRVs are also not appropriate for use in planning

nutritional requirements for intravenous therapy. It

also needs to be taken into consideration that the

requirement for a particular nutrient is dependent

on the composition of the diet as a whole, as the

presence or absence of other dietary constituents

can affect the absorption and utilization of some

nutrients. The DRVs for one particular nutrient there-

fore assume that DRVs are also being met for all other

nutrients and energy.

0019Ultimately, the value of the DRVs when used as a

frame of reference will also be affected by the quality

of the intake data against which they are to be

compared. The DRVs indicate how the amounts of

1862 DIETARY REFERENCE VALUES

different nutrients should average out over time. It is

not necessary for an individual to be consuming these

amounts of every nutrient at every meal, or even on a

daily basis. Therefore, when collecting dietary data

for comparison with DRVs, the data should reflect

long-term habitual intakes, whether in the individual

or at the community level.

0020 In conclusion, the DRVs are a major advance in the

use of reference nutrient levels in the science of

human nutrition. They build on the old methods

whilst expanding the uses and relevance of recom-

mendations. It is still necessary for people to be

aware of the derivation and limitations of the DRVs,

but they do go some way towards reducing the poten-

tial misuse and abuse of these values.

See also: Carbohydrates: Requirements and Dietary

Importance; Dietary Requirements of Adults; Energy:

Intake and Energy Requirements; Fats: Requirements;

Food Labeling (Labelling): Applications; Millets;

Protein: Requirements; Vitamins: Overview

Further Reading

British Nutrition Foundation Briefing Paper 19 (1992)

Dietary Reference Values – What are They and How

Should They be Used? BNF: London

Buttriss J (2000) Nutrient requirements and optimisation of

intakes. British Medical Journal 56(1): 18–33.

Department of Health. Report on Health and Social Sub-

jects 41 (1991) Dietary Reference Values for Food

Energy and Nutrients for the United Kingdom. London:

HMSO.

Jackson AA (2001) Human protein requirement: policy

issues. Proceedings of the Nutrition Society 60: 7–11.

Waterlow JC (1979) Uses of recommended intakes. The

purpose of dietary recommendations. Food Policy 4:

107–114.

DIETARY REQUIREMENTS OF ADULTS

J Dwyer, Jean Mayer US Department of Agriculture,

Human Nutrition Research Center on Aging at Tufts

University, Boston, MA, USA

History and Current Status

0001 Nutrients are substances that are not synthesized in

the body in sufficient amounts to perform critical

functions such as growth, reproduction, or health

maintenance and therefore they are required from

the diet.

0002 Although humans have known since ancient times

that they must eat to live, only recently has it been

possible to specify the chemicals and amounts in-

volved with certainty. The rise of this knowledge has

led to the growth of nutrition as a science. In the

nineteenth and early twentieth centuries, the concepts

of energy metabolism and essential nutrients were

formulated, and techniques for quantifying nutrient

requirements were developed. Experimental and

observational work determined the amounts of

nutrients required to prevent frank dietary deficiency

disease. During the middle and later twentieth cen-

tury, attention turned to nutrient functions associated

with chronic degenerative diseases and additional

roles for nutrients. This chapter describes the recent

work on dietary requirements as they have been for-

mulated by the Health Canada/Food and Nutrition

Board, National Academy of Sciences in the USA in

the dietary reference intakes. These deliberations and

parallel efforts by expert groups of the World Health

Organization/Food and Agricultural Organization of

the United Nations and many national or regional

governments occupied most of the 1990s. They will

conclude early in the twenty-first century.

0003The landmark Dietary Reference Intake project

arose because much new work on nutrient needs,

the balance and interactions between nutrients, and

the development of chronic degenerative diseases had

become available by the early 1990s. New techniques

for evaluating the biological functions of nutrients

were present. In addition, many nonnutrients in food

with biological activities of health significance were

being described. The need for upgrading data on the

validity and reliability of existing dietary standards

was evident. Moreover, new statistical techniques for

adjusting nutrient intake data for assessment and

planning purposes had recently become available,

and these needed to be integrated into existing

concepts. Finally, it had become apparent that, in

the past, some uses of dietary requirements and nutri-

ent standards had been inappropriate, and new

guidance was needed.

0004Estimates of nutrient requirements are now well

delineated for healthy adults, but they are still not so

for special groups, such as very premature infants, the

very old, and the ill. This remains a challenge for the

future.

DIETARY REQUIREMENTS OF ADULTS 1863

What Nutrients are Required?

0005 All human beings require the energy-providing

nutrients (amino acids, fatty acids and fat, glucose,

and other carbohydrates), vitamins, minerals, and

water. The specific essential nutrients include organic

compounds such as nine essential amino acids

(leucine, isoleucine, histidine, lysine, methionine,

phenylalanine, threonine, tryptophan, and valine).

Cystine, which can replace some of the methionine

requirement, and tyrosine, which can replace some of

the phenylalanine requirement, are also sometimes

considered to be essential. Taurine is thought to be

essential in premature infants. There are two essential

fatty acids (linoleic and arachadonic), four fat-soluble

vitamins (A, D, E, and K), and 10 water-soluble vita-

mins (niacin, thiamin, riboflavin, folate, vitamin B

vitamin B

, biotin, pantothenic acid, ascorbic acid,

and choline). A source of glucose is also required, but

this can be synthesized from glycerol and glycogenic

of the amino acids if they are present in sufficient

amounts. Among the inorganic substances, three

minerals (calcium, phosphorus, and magnesium),

nine trace mineral elements (iron, iodine, selenium,

copper, zinc, manganese, fluoride, chromium, and

molybdenum), three electrolytes (sodium, chlorine,

and potassium), and water are required.

0006 There are many other substances that are naturally

present in foods and are known to be required in the

diets of other species. For example, several ultratrace

elements are essential in experimental animals and

may also be essential in humans (arsenic, nickel, sil-

icon, boron, cadmium, lead, lithium, tin, vanadium,

and cobalt), but at present, evidence is lacking that

they are. Dietary requirements, if any, for many other

compounds, the so-called candidate nutrients, are

also uncertain, and they are currently being explored.

They include carnitine, myoinositol, cholesterol, the

nonprovitamin A caratenoids, and dietary fibers. In

addition to these substances that appear to be essen-

tial for some higher animals but not for humans,

scores of other organic phytochemical and zoochem-

ical compounds are present in foods that have various

health effects, and some of them eventually may

prove to be essential. These include growth factors

and coenzymes such as dietary nucleotides, coenzyme

Q, lipoic acid, and many others.

How Much is Required?

0007 The amount of a nutrient required depends on the

health criterion or indicator of adequacy that is used.

For most nutrients, multiple possible criteria exist

that are based on various health parameters. Levels

of nutrients that are adequate to meet one of these

criteria, such as prevention of night blindness, may

not be so for reaching another, such as repletion of

liver stores of vitamin A. The types of evidence and

the criteria used to establish the quantitative require-

ments for a specific compound vary from nutrient to

nutrient. They also vary from one age or physio-

logical stage to another.

0008The amounts of nutrients that are needed, and

sometimes the types of the essential nutrients that

must be provided, differ depending on age and

physiological state. For example, taurine is an amino

acid that may be essential in premature infants but is

not so in older infants or adults. Also, the forms in

which nutrients occur in foods, fortified foods, and

supplements affect their bioavailabiliy and thus their

requirements.

How are Nutrient Requirements

Expressed?

0009The quantitative recommendations for nutrient

intakes used as reference values or standards for plan-

ning and evaluating the nutrient intakes of healthy

people are collectively referred to as the dietary refer-

ence intakes, or DRIs. The DRIs include the estimated

average requirement for nutrients (EAR) and three

other reference values: the recommended dietary

allowances, the adequate intake, and the tolerable

upper level.

0010The process for establishing estimated average

requirements and other dietary reference intakes

used in the USA and Canada is summarized below.

It is described in detail in monographs on groups of

nutrients published by the Standing Committee on the

Scientific Evaluation of Dietary Reference Intakes of

the Food and Nutrition Board, Institute of Medicine,

National Academy of Sciences. They replace both the

1989 Recommended Dietary Allowances, the single

reference values used in the USA, and previous refer-

ence data used by Canada in the past.

0011The DRI paradigm has now been used to develop

current estimates of nutrient requirements for vita-

mins and minerals. Requirements for energy, macro-

nutrients, water, and other compounds of possible

health significance are now being considered, but as

of this writing, no paradigm has been published for

these dietary constituents.

Estimated Average Requirement

0012The estimated average requirement (EAR) is the

amount of a nutrient that is estimated to meet the

requirement for a specific criterion of adequacy of

half of the healthy individuals of a specific age, sex,

and life-stage. In setting the EAR, the evidence for

1864 DIETARY REQUIREMENTS OF ADULTS

each possible criterion is considered, and the reason

for selecting the criterion that is finally chosen is

justified. The amount of the nutrient necessary to

meet the appropriate criterion of adequacy varies

from one individual to the next, but the data are

usually distributed normally or can be transformed

to achieve a normal distribution. The EAR is not

useful as an estimate of nutrient adequacy in individ-

uals, because it is a mean requirement for a group,

and the variation around this number is considerable.

At the EAR, 50% of the individuals in a group are

below their requirement, and 50% are above it. Thus,

a person whose usual intake is at the EAR has a 50%

risk of an inadequate intake during the reporting

period. An individual with an intake between the

RDA and the EAR would have a risk of inadequacy

between 50 and 2–3%. An individual with a usual

intake below the EAR would have a risk of inad-

equacy between 50 and 100%. This is because the

EAR is derived from a group estimate. The precise

amount of a nutrient that will be adequate for any

given individual is therefore unknown. It can be

stated only in terms of probabilities, and thus it is

rarely used in clinical practice. For healthy individ-

uals whose usual nutrient intakes are accurately

described, the EAR can be used to assess the approxi-

mate probability of inadequacy, although the range of

error in the estimate is considerable.

Recommended Dietary Allowances

0013 The recommended dietary allowance (RDA) is the

average daily dietary intake level that suffices to

meet the nutrient requirements of nearly all (97–

98%) healthy persons of a specific sex, age, life

stage, or physiological condition (such as pregnancy

or lactation). The RDA is a nutrient intake goal for

planning the diets of individuals.

0014 To ensure that the needs of any given individual are

met, a measure of variability around the EAR, usually

the standard deviation or, if data are insufficient,

an estimated coefficient of variation (CV) of 10 or

15%, is assumed. Then, an amount that covers this

variation is calculated. The RDA is therefore defined

statistically as two standard deviations above the

EAR, assuming that the distribution of nutrient

requirements is normal, as it may be for most nutri-

ents that have been studied. That is, the RDA is the

EAR þ2 SD. If the SD is not available, using a CV

of 10%, the RDA is set at 1.2 times the EAR. If the

CV is assumed to be 15%, the RDA is set at 1.3 times

the EAR.

0015 The risk, but not the certainty, of inadequacy

increases as intakes fall further and further below

the RDA. However, the RDA is an overly generous

criterion for evaluating nutrient adequacy. By defin-

ition, the RDA exceeds the actual requirements of all

but about 2–3% of the population. Therefore, many

individuals who are below the RDA may still be

getting enough of the nutrient in question to be

above their requirement level.

0016The 1989 RDA consisted of a single reference

point, with a focus on achieving sufficiency only, for

individuals. However, the reference was often mis-

takenly applied as a standard for the evaluation of

population nutrient intakes. In fact, the RDA is not

appropriate for assessing the diets of individuals, or

for assessing or planning the diets of groups.

Adequate Intake

0017For some nutrients, it is not possible to set RDA

because data on the requirement (EAR) for the func-

tion of health significance that is judged to be the

most important are not available. However, enough

information is available to make some helpful state-

ments about healthful nutrient intake levels. An ad-

equate intake (AI) is set when experts believe that

sufficient data on the requirement using the criterion

that has been chosen to determine the requirement are

still preliminary. More research is needed before an

EAR and an RDA for that particular criterion or

function can be determined. The AI is also an appro-

priate goal for the nutrient intake of individuals.

0018The AI is a value, based on observed or experi-

mentally determined approximations of nutrient

intakes by groups of healthy people, that is used

when an EAR and an RDA cannot be determined.

For infants, the AI is always based on the mean

intakes of groups of healthy infants. For adults, in

some instances, the AI is set as the mean of diets of

healthy individuals in some reference groups. For

other nutrients, the criteria are less precisely deter-

mined but always chosen to be generous enough for

good health. In the DRIs that have been established

to date, AIs rather than RDAs are proposed for all

nutrients for infants up to age 1, and for calcium,

vitamin D, fluoride, pantothenic acid, biotin, choline,

vitamin K, chromium, and manganese) for persons of

all ages. The AI is not a requirement, and it should not

be interpreted as such. Rather, it is an average or

median intake in a group of healthy people, all of

whom are assumed to be meeting their nutrient

requirements.

Tolerable Upper Intake Levels of

Nutrients (UL)

0019Until recently, dietary requirements for nutrients were

thought to consist solely of assuring adequacy. Now,

DIETARY REQUIREMENTS OF ADULTS 1865