FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS	ESN:FAO/WHO/UNU/ EPR/81/18 August 1981
	WORLD HEALTH ORGANIZATION
	THE UNITED NATIONS UNIVERSITY

Item 2.4.2 of the Provisional Agenda

Joint FAO/WHO/UNU Expert Consultation on
Energy and Protein Requirements

Rome, 5 to 17 October 1981

FOOD ENERGY REQUIREMENTS OF INFANTS AND
PRE-SCHOOL CHILDREN

by

R.G. Whitehead, A.A. Paul and T.J. Cole

MRC Dunn Nutrition Unit, Cambridge, UK and Keneba, The Gambia, West Africa

1. Introduction and summary of findings

1.1. Carefully conducted surveys of food intake have been indicating that even in the affluent countries of the world, children and adults frequently do not consume amounts of food sufficient to meet WHO/FAO estimates of dietary energy requirements, yet average weights are well up to or even in excess of those considered desirable. This is usually assumed to be due to the low level of activity in such communities, but this conclusion is almost certainly not the whole truth and there is reason to believe that the previous recommendations overestimated the energy needs of many groups of people.

1.2. This paper concentrates on an analysis of the measured food energy intakes of healthy infants and preschool children living in Europe and North America in relation to their growth. It demonstrates that energy intakes in such children differ fundamentally from current recommendations. The conclusion is that during infancy recommendations are mostly in excess of needs for healthy growth along the NCHS 50th centile, particularly around 3–9 months. Between 1–5 years also there are differences between customary dietary energy intake and the previous recommendation but these are less marked. From 1–3 years measured values tend to slightly exceed the recommendation, while from then onwards the WHO/FAO 1973 value could well be an overestimate of needs.

2. Subjects

2.1. The subjects selected for this analysis were all from investigations carried out in either Canada, Sweden, the UK or USA. Details of the selected children and the food they ate are summarised in Table 1. No surveys from the developing countries have been used, to avoid the complications of a poor socioeconomic status affecting the availability of food, and the influence of repeated infections inhibiting food intake and growth.

2.2. The studies were all carried out on ‘free living’ individuals studied in their mothers' homes. In one or two cases the information was obtained from data which had been recorded graphically and thus minor errors may have arisen, but it is highly unlikely that these are of sufficient magnitude to affect the basic conclusions that have been reached. The total number of child data points in the literature used for assessing the infancy period was 9046 and for 1–5 years, 5327. All statistical analyses have been ‘weighted’ to take account of the different numbers in each study. The sexes have been analyzed together but where there proved to be a significant sex effect, separate terms have been fitted in the regression equations for boys and girls. In a few studies the results were originally published with the sexes combined. Where it was considered appropriate to use such data in the analysis it has been assumed there were equal numbers of boys and girls.

3. Results

Infancy

3.1. The data points used in the analyses of dietary energy intake for male and female infants are given in Tables 2 and 3: where no separation of the sexes was made in the original paper the data have been included in Table 4. The mean total energy intakes of the different groups of boys and girls are also shown in Fig. 1 and 2 in relation to the current recommendation according to age.

3.2. WHO/FAO (1973) and all preceding and subsequent expert committees have assumed that the energy intake of the average infant should increase with age in a curvelinear manner, essentially reflecting the pattern of growth in weight (see Fig. 1 and 2). In contrast, and of major importance, the data for actual measured intakes exhibited, when subjected to both linear and quadratic regression analysis, no significant curvature (t = 0.76).

The best fit linear relationship obtained by multiple regression analysis, allowing for sex as well as age, was Energy = 404 + 49 months for girls, and for boys: Energy = 456 + 49 months (R = 0.93, p<<0.001). In general, measured mean intakes were substantially below the previously recommended level except in very early and late infancy (Fig. 1 and 2).

3.3. This analysis of the actual pattern of food energy intake of young babies clearly shows it to be quite different from that conventionally believed to be the case. This is of fundamental importance to the definition of food needs for young babies and infants. Table 5 gives the mean energy intakes predicted by the regression equations for each month during the first year of life, together with the WHO/FAO 1973 values. At 6 months, for example, the mean measured intake of boys was 750 kcal/d, while the recommendation is 864 kcal/d.

3.4. A critical question is how satisfactorily were the children growing? Fig. 3 and 4 show the corresponding mean body weights of the selected groups of girls and boys in comparison with the NCHS 25th, 50th and 75th centiles. In spite of theoretically low intakes, growth of both girls and boys corresponded favourably with this currently accepted international standard.

3.5. The fact that the pattern of change in food intake by young infants does not concur with current theory is further emphasized by Fig. 5, which shows the present recommendation plus the measured energy intakes re-analyzed on a per kg basis. Virtually all the measured data lie well below the current recommendation. In marked contrast to the essentially linear recommendation, the analysis of the data in Fig. 5 revealed a highly significant curvelinear relation between energy consumption on a body weight basis and age, the significance of the quadratic term being t = 8.5 (p<<0.001). The solid line in Fig. 6 is the quadratic best fit, Energy = 120 - 10.4 (months) + 0.76 (months)² R = 0.64 p<<0.001. There was no significant difference between the sexes and the results have therefore been combined. The energy intakes per kg at each month predicted by this regression equation are given in Table 6. Between 0 and 6 months measured intakes fell more rapidly than has previously been assumed to be the case. In the second half of infancy, however, this effect was reversed and intakes rose again, so that by one year actual intakes coincided again with the 1973 recommendation. For most of infancy, however, measured intakes were substantially lower than the theoretically required amount.

Pre-school children

3.6. The dietary background for the pre-school children is also given in Table 1, and details of the energy intakes in Tables 7, 8 and 9. The corresponding growth curves of the children are given in Fig. 7 and 8; once again the progress of both the boys and the girls followed closely the NCHS 50th centile.

3.7. Total energy intake data for the boys and girls are analyzed in Fig. 9 and 10; the WHO/FAO 1973 recommended values are also provided. The change in food energy intake with age was significantly curvelinear (t = 2.01 p = 0.05). The quadratic best fit for boys was Energy = 899 + 286 years - 25.2 years², and for girls 793 + 286 years - 25.2 years² (R = 0.90, p<<0.001). Mean monthly intakes derived from these regression equations are given in Table 10.

3.8. Corresponding data for energy intake analyzed on a kg body-weight basis are given in Fig. 11 and Table 11. Statistical analysis of the data revealed no significant sex effect (t = 0.8). The fitted quadratic regression equation for boys and girls combined was Energy = 99.8 + 7.95 yr - 2.15 yr² (R = 0.61, p < 0.001), the significance of the quadratic term being t = 2.01 (P = 0.05). These data demonstrate that the higher energy intake/kg body-wt observed at the end of the second half of infancy, compared with that at 6 months, was maintained, the actual maximum intake being at around 2 years. After 2 years, however, the intake/kg fell quickly. Up to 3 years mean intake was slightly in excess of the WHO/FAO 1973 recommendation but afterwards it was below. An analysis of data beyond this time has indicated that average intakes in childhood remain below the recommendation right up to 16 years, especially in girls. This age period, however, lies outside the scope of this position paper. (See Appendix).

4. Discussion

Infancy

4.1. WHO/FAO 1973 based their estimates for energy requirements during infancy largely on data published by Fomon in 1967. Since that time, particularly in the past 3 years, there have been a number of studies published on the food consumption of young infants, including breastfed ones. These new data show that current recommendations differ fundamentally from actually measured intakes.

4.2. The fact that the present analysis does not represent a statistical artifact is demonstrated by Fig. 12 where the data for infancy are viewed together with those for pre-school children. This shows not only the two fitted regression curves for energy intake/kg body weight but also a histogram derived from the simple averages of each data point. Both approaches clearly demonstrate essentially the same pattern.

4.3. Food energy intakes per kg fall off much more rapidly during the first six months than has previously been believed to be the case. The pattern of change during the second half of infancy differs even more from current recommendations. The data indicate that instead of continuing to fall, intakes per kg rise again between 7 and 12 months, presumably reflecting the marked increase in physical activity of the child which occurs at this time. The initial fall, during early infancy, may be assumed to be partly the result of a rapidly decelerating velocity of growth, a reduction in the rate of fat storage in the adipose tissues plus the general physiological fact that the larger the baby becomes relative to his surface area, the less energy per unit mass is required for homoeostasis.

4.4. It is recommended that the committee seriously considers the revision of current estimated recommendations for energy intake during infancy in the light of the new data presented and analyzed here.

Pre-school children

4.5. From 1–5 years the 1973 report used intakes collected by Widdowson in the UK before the Second World War, plus American data made available to the 1956 FAO committee on energy requirements. Since that time one or two further major studies have been carried out in the UK and USA. The indication is that children with a perfectly acceptable rate of growth as judged by the NCHS standards consume on average a little more dietary energy than the WHO/FAO 1973 recommendations between 1 and 3 years. This situation is only a temporary one, however, and after 3 years it would appear that the previous recommendation may overestimate needs, especially in girls. The committee may wish to consider some modification to their recommendations for pre-school children but more data are required before an equivocal conclusion can be reached. It is our opinion that data for children of school age also need to be re-examined in a critical manner.

References

Beal, V.A. (1970): Nutritional intake. In Human Growth and Development, ed. R.W. McCammon. Springfield: C.C. Thomas. 63–100

Black, A.E., Billewicz, W.Z. & Thomson, A.M. (1976): The diets of preschool children in Newcastle-upon-Tyne 1968-71. Br. J. Nutr. 35, 105–113.

Chandra, R.K. (1981): Breast feeding, growth and morbidity. Nutrition Research 1, 25–31.

DHSS (1975): A nutrition survey of pre-school children 1967-8. Rep. Hlth. Soc. Subj. No. 10.

Ferris, A., Vilhjalmsdottir, L.B., Beal, V.A. & Pellet, P.L. (1978): Diets in the first six months of infants in western Massachusetts. J. Amer. diet. Ass. 72, 155–160.

Fomon, S.J. (1967): Infant Nutrition. Philadelphia: Saunders.

Fomon, S.J., Thomas, L.N., Filer, L.J., Ziegler, E.E. & Leonard, M. (1971): Food consumption and growth of normal infants fed milk based formulas. Acta Paed. Scand. Suppl. 223, 36 pp.

Fryer, B.A., Lamkin, G.H., Vivian, V.A., Eppright, E.S. & Fox, H.M. (1971): Diets of preschool children in the North Central Region. J. Amer. diet. Ass. 59, 228–232.

Hamill, P.V.V. (1977): NCHS Growth Curves for Children. U.S. Dept. Health, Education and Welfare, Pub. No. PHS. 78–1650. National Center for Health Statistics, Hyattsville, Maryland.

Hofvander, Y., Hagman, U. & Sjolin, S. (1979): Breast milk consumption in healthy Swedish infants. Vår Fda 31, Suppl. 3, 182.

McKillop, F.M. & Durnin, J.V.G.A. (1981): A comparison of the energy and nutrient intakes of 305 Glasgow infants with current recommendations for the UK. Proc. Nutr. Soc. (in press).

Morgan, J. & Mumford, P. (1977): Feeding practices and food intake of children under two years of age. Proc. Nutr. Soc. 36, 47A.

Morgan, J., Mumford, P., Evans, E. & Wells, J.K.C. (1976): Food intake and anthropometric data on children under 4 years old living in the south of England. Proc. Nutr. Soc. 35, 74A–75A.

Nelson, M. & Nettleton, P.A. (1980): Dietary survey methods. 1. A semi-weighed technique for measuring dietary intake within families. J. Hum. Nutr. 34, 325–348.

Vobecky, J.S., Vobecky, J., Demers, P-P., Shapcott, D., Blanchard, R. & Black, R. (1980): Food and nutrient intake of infants in the first fifteen months. Nutr. Rep. Internat. 22, 571–580.

Wallgren, A. (1944-5): Breast-milk consumption of healthy full-term infants. Acta Paed. Scand. 32, 778–790.

Whitehead, R.G. & Paul, A.A. (1981): Infant growth and human milk requirements: a fresh approach. Lancet (in press).

WHO/FAO (1973): Energy and protein requirements. Wld. Hlth. Org. Tech. Rep. Ser. No. 522. WHO: Geneva.

Yeung, D.L., Hall, J. & Leung, M. (1980): Adequacy of energy intake of infants. J. Canad. Diet. Ass. 41, 48–52.

^* SW = SwedenCAN = Canada

^** F = formula fedFS = formula + solid foodB = breastfedBC = breastfed + complementary foodS = solid food only

† Assumed that the 556 children studied were evenly distributed between boys and girls.

See Table 2 for footnotes

^* Boys: Energy = 456 + 49 monthsGirls: Energy = 404 + 49 months(R = 0.93, p<0.001)

† Calculated using NCHS 50th centile body weights

* Energy = 120 - 10.4 months + 0.76 months²

(R = 0.64, p < 0.001)

The number of child data points in this analysis are more than for that in Table 2 because of the extra information available from study 15 in which energy/kg but not total energy nor body weights were provided

* Boys = 899 + 286 (years) - 25.2 (years)²Girls = 793 + 286 (years) - 25.2 (years)²(R = 0.90, p < 0.001)

* Energy = 99.8 + 7.95 (years) - 2.15 (years)²

(R = 0.61, p < 0.001)

Fig. 1. Energy intake (kcal/d) of male babies in the first year of life compared to WHO/FAO (1973) recommendations. Numbers and letters indicate the reference numbers given in Table 1. Where sexes were not separated in refs. 2, 3 and 10 (A), the data are included in both Fig. 1 and 2. The solid line is the fitted linear regression line.

Energy = 456 + 49 (months).

Fig. 2. Energy intake (kcal/d) of female babies in the first year of life compared to WHO/FAO (1973) recommendations. Numbers and letters indicate the reference numbers given in Table 1. Where sexes were not separated in refs. 2, 3 and 10(A), the data are included in both Fig. 1 and 2. The solid line is the fitted linear regression line.

Energy = 404 + 49 (months).

Fig. 3. Weights of male babies in the first year of life compared to the NCHS (Hamill, 1977) 25th, 50th and 75th centiles. Numbers and letters indicate the reference numbers given in Table 1.

Fig. 4. Weights of female babies in the first year of life compared to the NCHS (Hamill, 1977) 25th, 50th and 75th centiles. Numbers and letters indicate the reference numbers given in Table 1.

Fig. 5. Energy intakes per kg body weight of boys and girls in the first year of life compared to WHO/FAO (1973) recommendations. Numbers and letters indicate the reference numbers given in Table 1.

Fig. 6. The quadratic regression line of energy intake per kg body weight of boys and girls in the first year of life calculated from the data in Fig. 5, compared to the WHO/FAO (1973) recommendation. Energy = 120 - 10.4 (months) + 0.76 (months)².

Fig. 7. Weights of boys aged 1 to 5 years compared to the NCHS (Hamill, 1977) 25th, 50th and 75th centiles. Numbers and letters indicate the reference numbers given in Table 1.

Fig. 8. Weights of girls aged 1 to 5 years compared to the NCHD (Hamill, 1977) 25th, 50th and 75th centiles. Numbers and letters indicate the reference numbers given in Table 1.

Fig. 9. Energy intakes (kcal/d) of boys aged 1 to 5 years compared to the WHO/FAO (1973) recommendations. Numbers and letters indicate the reference numbers given in Table 1. Where sexes were not separated in refs. 10(A) and 11 (B), the data are included in both Fig. 9 and 10. The solid line is the fitted quadratic regression line. Energy = 899 + 286 (years) - 25.2 (years)².

Fig. 10. Energy intakes (kcal/d) of girls aged 1 to 5 years compared to the WHO/FAO (1973) recommendations. Numbers and letters indicate the reference numbers given in Table 1. Where sexes were not separated in refs. 10 (A) and 11 (B), the data are included in both Fig. 9 and 10. The solid line is the fitted quadratic regression line. Energy = 793 + 286 (years) - 25.2 (years)².

Fig. 11. The fitted quadratic regression line of energy intake/kg body weight in boys and girls aged 1 to 5 years calculated for the studies shown in Fig. 7 to 10, compared to the WHO/FAO (1973) recommendations. Energy = 99.8 + 7.95 (years) - 2.15 (years)².

Fig. 12. Mean energy intake/kg body weight calculated from the data at each age point from 1 month to 5 years, compared to the fitted quadratic regression lines shown in Fig. 6 and 11. The number of data points at each age is shown.

Appendix

Recent studies of dietary energy intakes of school
children in the UK

Introduction

The analysis of dietary energy intakes of preschool children has indicated that after about 3 years of age, measured intakes fall below the WHO/FAO 1973 Recommendations, especially in girls. An extension of the examination with recent studies carried out on large numbers of school children in the UK shows that intakes remain below the recommendation particularly in girls continuously from 5 to 18 years, and also in 13 to 16 year old boys. The weights of the children are unfortunately presented in only one study; but no children were reported to be undernourished (Cook et al. 1970; Darke et al. 1980) and only a small proportion (4–7%) were classified as obese (Darke et al. 1980).

Results

Details of the children studied are given in Table 1A and their energy intakes in tables 2A and 3A. For comparison, data from the longitudinal study of Beal in the USA (Study 5) are also included.

The dietary energy intakes are shown in Fig. 1A. Data from 1 year of age are also included to provide a continuous spectrum throughout childhood. Details of the studies from 1 to 5 years are contained in the main body of this position paper.

The intakes of the UK boys are almost all below the WHO/FAO 1973 recommendation, and the American ones also deviate below the recommendation after 13 years. With the girls the change in energy intake with age follows the same pattern as the recommendation, but from the age of 5 all intakes fall on a curve consistently below the recommendation by some 250 kcal/d.

Discussion

The lower energy intakes of children reported in recent studies are presumably a reflection of decreased physical activity. It is anticipated that studies from other industrialised countries would reveal a similar trend, but a search for such studies has not been attempted for this position paper. The present recommendations appear to be a considerable overestimate of energy needs for school girls of all ages, and somewhat in excess of those for adolescent boys.

References

Beal, V.A. (1970). Nutritional intake. In Human Growth and Development, ed. R. W. McCammon. Springfield: C.C. Thomas. 63–100.

Cook, J., Altman, D.G., Moore, D.M.C., Topp, S.G., Holland, W.W. and Elliot, A. (1973). A survey of the nutritional status of schoolchildren: relation between nutrient intake and socioeconomic factors. Brit. J. Prev. Soc. Med. 27, 91–99.

Darke, S.J., Disselduff, M.M. and Try, G.P. (1980). Frequency distributions of mean daily intakes of food energy and selected nutrients obtained during nutrition surveys of different groups of people in Great Britain between 1968 and 1971. Brit. J. Nutr. 44, 243–252.

Durnin, J.V.G.A., Lonergan, M.E., Good, J. and Ewan, A. (1974). A cross-sectional nutritional and anthropometric study, with an interval of 7 years, on 611 young adolescent school children. Brit. J. Nutr. 32, 169–179.

Nelson, M. and Nettleton, P.A. (1980). Dietary survey methods. 1. A semi-weighed technique for measuring dietary intake within families. J. Hum. Nutr. 34, 325–348.

Fig. 1A. Energy intakes (kcal/d) of boys and girls from 1 to 18 years compared to WHO/FAO 1973 recommendations. Numbers and letters refer to reference numbers given in Tables 1 and 1A.