Scientific Opinion on Dietary Reference Values for energy


EFSA Journal 2013;11(1):3005 [112 pp.].
EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA)
Panel Members
Carlo Agostoni, Roberto Berni Canani, Susan Fairweather-Tait, Marina Heinonen, Hannu Korhonen, Sébastien La Vieille, Rosangela Marchelli, Ambroise Martin, Androniki Naska, Monika Neuhäuser-Berthold, Grażyna Nowicka, Yolanda Sanz, Alfonso Siani, Anders Sjödin, Martin Stern, Sean (J.J.) Strain, Inge Tetens, Daniel Tomé, Dominique Turck and Hans Verhagen

The Panel wishes to thank the members of the Working Group on Population Reference Intakes: Carlo Agostoni, Jean-Louis Bresson, Patrick Even, Susan Fairweather-Tait, Albert Flynn, Ambroise Martin, Monika Neuhäuser-Berthold, Hildegard Przyrembel, Sean (J.J.) Strain, Inge Tetens and Daniel Tomé for the preparatory work on this scientific opinion and EFSA staff: Anja Brönstrup for the support provided to this scientific opinion.

Opinion of the Scientific Committee/Scientific Panel
On Request From
European Commission
Question Number
29 November 2012
10 Juli 2013
European Food Safety Authority (EFSA), Parma, Italy
Article (3.42 MB)3.42 MB

Following a request from the European Commission, the Panel on Dietetic Products, Nutrition and Allergies (NDA) derived dietary reference values for energy, which are provided as average requirements (ARs) of specified age and sex groups. For children and adults, total energy expenditure (TEE) was determined factorially from estimates of resting energy expenditure (REE) plus the energy needed for various levels of physical activity (PAL) associated with sustainable lifestyles in healthy individuals. To account for uncertainties inherent in the prediction of energy expenditure, ranges of the AR for energy were calculated with several equations for predicting REE in children (1-17 years) and adults. For practical reasons, only the REE estimated by the equations of Henry (2005) was used in the setting of the AR and multiplied with PAL values of 1.4, 1.6, 1.8 and 2.0, which approximately reflect low active (sedentary), moderately active, active and very active lifestyles. For estimating REE in adults, body heights measured in representative national surveys in 13 EU Member States and body masses calculated from heights assuming a body mass index of 22 kg/m2 were used. For children, median body masses and heights from the WHO Growth Standards or from harmonised growth curves of children in the EU were used. Energy expenditure for growth was accounted for by a 1 % increase of PAL values for each age group. For infants (7-11 months), the AR was derived from TEE estimated by regression equation based on doubly labelled water (DLW) data, plus the energy needs for growth. For pregnant and lactating women, the additional energy for the deposition of newly formed tissue, and for milk output, was derived from data obtained by the DLW method and from factorial estimates, respectively. The proposed ARs for energy may need to be adapted depending on specific objectives and target populations.


Following a request from the European Commission, the Panel on Dietetic Products, Nutrition and Allergies (NDA) was asked to deliver a scientific opinion on dietary reference values (DRVs) for the European population, including energy.

The DRVs for food energy provide a best estimate of the food energy needs of population groups within Europe. They are given as average requirements (ARs) of specified age and sex groups and are of limited use for individuals. The reference values need to be adapted to specific objectives, such as dietary assessment, dietary planning, labelling dietary reference values, or development of food-based dietary guidelines. In addition, there is a need to define and characterise the target population.

In this Opinion, total energy expenditure (TEE) in the steady state of a healthy body mass was chosen as the criterion on which to base the AR for energy. In practice, the adequacy of usual energy intakes is best monitored by measuring body mass. In terms of regulation of body mass, the overall energy balance over a prolonged period of time needs to be considered. TEE expended over 24 hours is the sum of basal energy expenditure, the energy expenditure of physical activity and the thermic effect of food. In this Opinion, resting energy expenditure (REE) was used as a proxy for the slightly lower basal energy expenditure, as most studies measured REE. TEE is best measured with the doubly labelled water (DLW) method, which provides energy expenditure data over biologically meaningful periods of time and under normal living conditions.

One approach to determine the AR for energy is to use regression equations which describe how TEE measured with the DLW method varies as a function of anthropometric variables (such as body mass and height) for defined population groups and for an activity constant that accounts for the level of physical activity. However, this approach has been criticised because of the inability of such TEE prediction models to account for the variations in energy expenditure of physical activity in a transparent way. In addition, limited TEE data generated with the DLW method are available, and they may not be representative for the European population; moreover, some age groups are under-represented. Another approach for estimating TEE is by the factorial method in which the energy spent in various activities is added to measured or predicted REE. This is achieved by using the physical activity level (PAL), which is defined as the ratio of TEE to REE per 24 hours and reflects the part of TEE that is due to physical activity. Accordingly, TEE is predicted as PAL x REE. During growth, pregnancy and lactation, additional energy is needed for the synthesis and deposition of new tissues, and for milk production.

In this Opinion, TEE of children and adults was estimated factorially to account for the diversity in body size, body composition and habitual physical activity among children and adult populations with different geographic, cultural and economic backgrounds.

To estimate REE, predictive equations were used, derived from regression analysis of measured REE, body masses and heights from groups of subjects. Body mass is the most important determinant of REE and all predictive equations use this parameter. In addition to body mass, height, sex, age and ethnicity can affect REE significantly and numerous equations have been developed to take into account one or several of these parameters. Based on the accuracy of various equations in specified population groups, five widely used equations (Harris and Benedict, 1919; Henry, 2005; Mifflin et al., 1990; Müller et al., 2004; Schofield et al., 1985) were considered as equally valid for estimating the REE of healthy adults in Europe. For healthy children and adolescents, the equations of Schofield et al. (1985) and Henry (2005) derived from large datasets and covering wide age groups were considered to be the most suitable.

PAL can be estimated either from time-allocated lists of daily activities expressed as physical activity ratio values or from the ratio of TEE (measured by the DLW method) to REE (either measured or estimated). However, the same limitations apply to the derivation of PAL values from DLW data as to the estimates of TEE with this method. Within the general population, PALs associated with sustainable lifestyles have been observed to range between 1.35 and 2.5, and to decrease only marginally with age. When assigning PAL values to descriptions of activities/lifestyles (such as light, moderate or heavy activity), the range of PAL values in each lifestyle category is large. Thus, the allocation of lifestyles to defined PAL values can only be considered a rough indication of PAL, but may be useful for decisions about which PAL values to apply in various circumstances and applications.

In the absence of arguments for the selection of one predictive equation best fitted to adults in the European Union (EU), REE was calculated with five widely applied predictive equations using individual data of measured body heights of adults obtained in 13 representative national surveys in EU Member States, with corresponding body masses calculated for a body mass index (BMI) of 22 kg/m2, i.e. the midpoint of the range of healthy BMI of adults as defined by the WHO. This yielded a range of ARs calculated for PAL values from 1.4 to 2.4 in steps of 0.2, and demonstrated the magnitude of uncertainty inherent in these values. However, for practical reasons, only one AR is proposed for a defined age and sex group with a healthy BMI of 22, and for PAL values selected to approximate corresponding lifestyles. The predictive equations of Henry (2005) were used to estimate REE because, at present, the underlying database is the most comprehensive as regards number of subjects, their nationalities and age groups. To derive TEE as REE x PAL, PAL values of 1.4, 1.6, 1.8 and 2.0 were chosen to approximately reflect low active (sedentary), moderately active, active and very active lifestyles. Because of a lack of anthropometric data from EU countries for age groups from 80 years onwards, average requirements were not calculated for adults ≥80 years.

For infants from birth to six months of age, energy requirements were considered to be equal to the energy supply from human milk, and no DRV is proposed. For infants aged 7-11 months, the ARs were estimated from equations for TEE, adding the energy needs for growth. TEE was based on measurements using the DLW method in healthy, full-term infants, exclusively breast-fed for the first four months of life and with adequate body mass. Body masses from the WHO Growth Standards were used to derive ARs for infants growing along the trajectory of this standard. Estimates of the energy requirement for growth were based on protein and fat gains reported in the literature.

The ARs of children from one year upwards are based on predicted REE and adjusted PAL for growth. REE was calculated using the predictive equations of Henry (2005) and Schofield et al. (1985) and median body masses and heights taken from the WHO Growth Standards (for children up to two years) or from harmonised growth curves of EU children (for children from 3 to 17 years). For the same reasons as outlined for adults, and because the results obtained with these two equations were very similar, only the predictive equations of Henry (2005) were applied for the estimation of REE values. PAL values of 1.4, 1.6, 1.8 and 2.0 were used for three age groups (1-3 years, >3-<10 years, and 10-18 years). Energy expenditure for growth was accounted for by a 1 % increase in PAL values for each age group.

For pregnant women, a mean gestational increase in body mass of 12 kg was considered to be associated with optimal maternal and fetal health outcomes. The additional amount of energy required during pregnancy to support this increase in body mass was estimated using the cumulative increment in TEE estimated with the DLW technique plus the energy deposited as protein and fat. Based on these data, the average additional energy requirement for pregnancy is 320 MJ (76,530 kcal) which equates to approximately 0.29 MJ/day (70 kcal/day), 1.1 MJ/day (260 kcal/day) and 2.1 MJ/day (500 kcal/day) during the first, second and third trimesters, respectively.

For women exclusively breastfeeding during the first six months after birth, the additional energy requirement during lactation was estimated factorially as 2.1 MJ/day (500 kcal/day) over pre-pregnancy requirements, taking into account a requirement of 2.8 MJ/day (670 kcal/day) for milk production and an energy mobilisation from maternal tissues of 0.72 MJ/day (170 kcal/day). No additional energy requirement is proposed for women lactating beyond the sixth month because volumes of milk produced during this period are highly variable and depend on the infant’s energy intake from complementary foods.

Energy, resting energy expenditure, prediction equation, physical activity level, total energy expenditure, factorial method, average requirement
Print on demand
Number of Pages