September 1981


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

Rome, 5 to 17 October 1981



N.G. Norgan
University of Technology
Leics, U.K.

Paper presented at the Workshop on
‘Energy Expenditure under Field
conditions’, Charles University,
Prague, Czechoslovakia

The FAO/WHO 1973 report on Energy and Protein Requirements states that ‘the energy requirements of individuals depend on four variables interrelated in a complex way: (a) physical activity, (b) body size and composition, (c) age, and (d) climate and other ecological factors (1). It continues ‘Individuals of the same size living in the same environment and with the same mode of life have a similar energy requirement whatever their ethnic origin.’

Since this report was published it has been suggested that energy metabolism is more variable and adaptable than was previously thought and that this should be taken into account in setting energy requirements. For example, a letter to Nature entitled ‘How much food does man require?’ by four eminent nutritionists begins ‘We believe that the energy requirements.…… of man are not known’ (2). More recently Edmundson has echoed this point in an article entitled ‘Adaptation to undernutrition: how much food does man require’ (3). He states ‘physiological variability is a basic human characteristic and it appears that the efficiency of food utilisation and the amount of work an individual is able to produce from a given quantity of food depends to a great extent on who is eating it’. Further support can be found in the recent work on the phenomenum of luxus-konsumption and dietary-induced thermogenesis stimulated by the reports of Miller, Mumford and Stock (4) that individuals could maintain body weight during overfeeding by wastefully disposing of the excess energy intakes. Finally, the U.N. University publication on ‘Protein-energy requirements under conditions prevailing in developing countries: current knowledge and needs’, states ‘It is clear that there are wide variations in the efficiency of energy utilisation both among and within individuals, and, therefore, in their dietary energy requirements’ (5).

The evidence adduced for variation and adaptation in energy metabolism can be summarised as follows:

  1. large numbers of apparently healthy, active adults exist on very low energy intakes (< 6MJ (1500 kcal) d-1),

  2. the efficiency of work is raised on low energy intakes and lowered on high energy intakes,

  3. in any group of 20 or more similar individuals, food intake can vary by as much as two fold,

  4. homeostatic mechanisms may reduce weight loss in undernutrition and weight gain in overnutrition and energy balance may be re-established.

Some of these observations may be explained by differences or changes in physical activity, body size, etc. but what is implied or explicitly stated is that metabolic efficiency and mechanical work efficiency are variable and show adaptation, particularly to varying levels of energy intake.

This paper considers the evidence for variation and adaptation in efficiency with varying levels of energy intake in order to assess their relevance to energy requirements. Adaptation is defined as a difference or change that reduces the physiological strain produced by a stressful component of the environment and variation as a difference or change that may or may not have physiological benefits.


The energy requirement of a man pursuing light activity is 11.3 MJ d-1 and for a woman 8.4 MJ d-1 (1). Groups with low intakes, <8.4 MJ d-1 in men and <6.3 MJ d-1 in women, include New Guineans, Ethiopians, Indonesians, Jamaicans, Indians and even some Americans (3, 6). Such intakes might have been dismissed as inaccurate in the past. How may they be explained and do they provide evidence of adaptive increases in efficiency?

Firstly, there is the conventional explanation of the effects of body size and physical activity. Most of the populations with low energy intakes have lower body weights than those observed in European populations. A New Guinean man of 55 kg with a pattern of light activity may require only 9.5 MJ d-1 which seems low but requires no novel explanation. Few dietary intake studies include measurements of energy expenditure and this is particularly so in those groups with low intakes. Level of physical activity is assessed subjectively which often results in an over-estimation of differences in total daily energy expenditure (7). New Guineans and many other populations are capable of and do heavy physical work as we can testify from working amongst them (8) but they also spend many hours and days in sedentary activities. People entirely dependent on their own expenditure of energy to produce food, build houses, engage in social activities, etc., may not need to be more than light to moderately active.

A second possibility is that the energy intake results are wrong. There may be a failure to record every item consumed or the measurement period may have been atypical of the normal intake pattern. Many studies show a decrease in body weight during the hungry periods of the year (9). This is unlikely to be the cause in every case and certainly not in the Ethiopians and New Guineans who were studied over most of the year.

Ashworth (10) found that dietary surveys gave a correct picture of the normal dietary pattern in only 5 out of 10 Jamaican subjects with low intakes. When fed their measured domestic intakes (mean of 68% of FAO/WHO requirements) in a metabolic ward for 7 days, 5 of the subjects lost weight despite being less active in the ward than at home. Durnin (6) selected 10 subjects who had relatively the lowest intakes out of a group of 33 Ethiopian market labourers and found that only 3 maintained weight on their normal intake in the laboratory. Thus, in some cases intake data are wrong. Ashworth found no clear cut evidence of an adaptive change in basal metabolic rate (BMR) and this was confirmed by Durnin. She did find that the energy cost of a standardised stepping task, was significantly lower in the low intake subjects compared to control subjects. Ashworth concluded that the remarkably low intakes represented partly an error in dietary survey, in part reduced physical activity and in part an increased efficiency of physical work.

Edmundson (3, 11, 12) has emphasised the role of adaptation in the resting and work metabolism of low intake groups. He calculated the ‘relative efficiency’ (ratio of daily energy expenditure to daily energy intake) in 54 Indonesians. Low intake individuals had high ratios. As energy expenditures were based on average energy costs of tasks, i.e. excluding the possibility of variation in efficiency, the ratio reflects the short term imbalance of energy intake to expenditure, not efficiency. Efficiency of work is more clearly and usually defined as either gross mechanical efficiency (GME) = work performed ÷ energy expenditure or net mechanical efficiency (NME) = work performed ÷ (energy expenditure - BMR).

Edmundson also measured the BMR and energy cost of cycling of 5 high and 6 low intake Indonesians matched for height and weight (3, 12). BMR was 50% lower in the low intake group, 2.6 versus 5.5 kJ min-1 which are 60% and 125% of European standards (1) and appear to be extreme variations. The energy cost of light work, cycling at 50W, was 18% lower and of medium work, 100W, 30% lower in the low intake group. The differences in BMR and at 100W were statistically significant. The NME are typical of values found in Europeans, except in the high energy group at 100W where the value of 18% is at the lower limit of the expected range, 22% ± 4% (13). Thus, the significant difference in work metabolism appears to arise from a low efficiency in the high intake group. Spurr and co-workers found no significant difference in GME of treadmill walking in groups of Columbians classified as normal, mild, intermediate and severely nutritionally compromised, on the basis of anthropometric and biochemical indices (14).

Although it is difficult to compare the energy cost of everyday freestyle activities as the pace at which they are pursued may differ, Viteri (15) found that the energy cost of work in over 50 agricultural activities was similar in groups of either very poor or nutritionally supplemented (i.e. presumably low and normal intake) Guatemalan peons and that they were identical to values reported in the African and European worker. He regarded this as evidence for a constancy of work efficiency in different populations. In our studies on New Guineans (8), the energy cost of resting activities did not differ from European values in a consistent manner. Energy expenditures while sitting were higher but while standing they were lower.

Thus, the evidence for adaptive increases in efficiency at rest and during work in low intake groups is tenuous. Nor is there evidence in the literature for increased mechanical efficiencies in various population groups (including those that might be expected to have low intakes) compared to European values. This is true for New Guineans, Ethiopians, East and Southern Africans, arctic and high altitude peoples. Indeed, in most cases, efficiencies are lower, which may be due to the nature of the standardised work tasks.


Widdowson (16) reported that in every yearly age group containing 20 or more boys or girls aged 1–18 years there was a two-fold difference in energy intake between the lowest and highest intake. This was also observed in adults and persisted when intakes were expressed per kg body weight. Some of the difference might be explained by diverse levels of physical activity. Within any occupation group, there is a 1.5 – 1.7 fold difference in the energy expenditure of the individuals with the lowest and highest expenditures (17).

Rose and Williams (18) found no difference in BMR or the energy cost of resting and standardised work activities of 6 large eaters (19 MJ d-1) and 6 small eaters (10 MJ d-1. The pace of free-style activities was 10–15% greater in the large eaters which is noteworthy as there exists differences in intakes but no apparent avenue of utilisation unless there are differences in activity. However, there was no direct evidence for differences in efficiency at work or rest.

To overcome the problems that arise in the interpretation of comparisons of different individuals, the effects of varying the energy intakes in the same individual should be examined.


In the classical Minnesota study of undernutrition (19) the energy intakes of 32 young men were approximately halved for 24 weeks. Body weights fell by 24% but by the end of the experiment had stabilised. The subjects had adapted by reducing energy expenditure as a result of reduced body size, physical activity and metabolic activity. BMR fell by 39%, which was more than the fall in body weight. In terms of per kg active cell mass, they were 16% lower which represents a true metabolic adaptation. One third of the fall in BMR could be explained by reduced metabolic activity and the remainder by reduced body size.

The falls in BMR made up a third of the total reduction in energy expenditure. The rest arose from decreased energy expenditure due to physical activity, of which 60% could be explained by reduced volitional activity and 40% by the reduced energy cost of tasks due to the smaller body size. There was no change in the efficiency of physical work. The energy cost of treadmill walking expressed per kg body weight was constant throughout the experiment.

However, the contribution of metabolic adaptation varies according to the duration of undernutrition. Grande and co-workers (20) found that 70% of the reduction in BMR resulting from undernutrition of 2–3 weeks duration could be attributed to reduced metabolic activity and only 30% to reduced cell mass. In chronic undernutrition, all of the reduction in BMR can be attributed to reduced cell mass (21).

The earlier study of undernutrition by Benedict (22) shows similar findings to the Minnesota study but as body composition measurements were unavailable the reasons for the changes in energy expenditure cannot be identified. Benedict concluded, however, that undernutrition caused a decreased net energy cost of walking. As no initial measurements were made the significance of the results depends on to what group they are compared.

Thus, undernutrition of previously normal men can result in a considerable reduction in energy expenditure, mainly by a reduced body size and reduced physical activity. Increased metabolic efficiency makes only a small contribution in medium or long-term undernutrition and there is no firm evidence of an increased mechanical efficiency of work.


There has been considerable interest in the possibility that some or most individuals can overeat without gaining weight or energy stores by an adaptive response that wastefully disposes of excess energy intakes. As energy metabolism obeys the laws of thermodynamics, if excess energy is not deposited, it must appear as heat. It has become common to group together all increases in heat production related to diet under the single term dietary-induced thermogenesis (DIT), which includes the familiar specific dynamic action (SDA) of foods. This is justifiable as it is technically difficult to identify the actual cause of the increased heat production other than it being associated with the diet. However, this description of increases in heat production does not disclose variations in metabolic efficiency. Increases in heat production may arise from normal processes without changing efficiency and without involving non-conservative mechanisms. For example, overfeeding is invariably associated with weight gain to some degree which increases resting and working energy expenditure and heat production. This is by definition DIT but it is a normal response although it is certainly an adaptive response in that weight gain is reduced. Similarly, the efficiency of energy storage is not 100%, a point rarely considered in drawing up energy balances in overfeeding (23). Normal energy storage raises heat production; it causes DIT. Therefore, evidence for DIT is not evidence for metabolic adaptation. Only when DIT is greater than the effects of weight plus the energy cost of storage plus a component comparable to SDA need non-conservative mechanisms bringing about an adaptive wasteful disposal of energy be invoked.

We have found that although metabolic rate (MR) does increase with overfeeding at rest and work, the increase is similar to that of body weight (23, 24). Six young men over-ate a diet of normal composition for 6 weeks, increasing their body weight by 10%. Metabolic rates lying in bed before rising increased by 12% and by similar amounts while sitting and treadmill walking during the day. Expressed as kJ kg-1 min-1, the rates were not significantly altered by overfeeding.

There is now much evidence that during overfeeding the increases in MR at rest, whether measured as BMR, resting metabolic rate (RMR) or 24 hr metabolic rate are about 10–15% (25–30). Greater increases are proportional to the weight gain. Possible exceptions to this are the studies of Miller (4) and Sims (31). Miller calculated that 24 hr MR rose to the same extent as overfeeding, i.e. excess intakes were disposed of wastefully but it is not possible to reconcile this with other results presented in the same paper (23). In Sims' study, the energy intake required to maintain the gained weight increased 50% but no measurements of MR were made to support this.

The mechanical efficiency of work during overfeeding has been considered to be unchanged (27, 32–34) but not by Miller (4) and Apfelbaum (25). Goldman reported that the ratio of the observed to predicted MR increased with overfeeding (35).

It has been suggested that MR increases may appear only after several days overfeeding (28) or a critical overload (34) but interestingly Dauncey (30) found that one day of overfeeding increased 24 hr MR by 10%, i.e. to the same extent as several days or weeks of overfeeding. The effect was greater during the night and RMR remained elevated 14 hr after the last meal. It may represent the SDA of the larger meals; 12% of the excess intake was converted to heat.


This paper has considered variations and adaptation to varying levels of energy intake and little has been said of the ‘normal’ variation found in all individuals and populations. Lack of space prevents the detailed consideration required as some areas, e.g. differences in BMR due to climate, are controversial. Normal variation can be shown, arithmetically, to be important. On a fixed energy intake, variation might result in some individuals becoming obese and some becoming under-nourished. However, there is still much to be learnt before its significance can be assessed. There is little information on whether variation is related or random, whether if an individual or population has a high energy expenditure during sitting they will also have a high expenditure while standing or walking. If individuals or populations are not on a fixed intake, the reason why they become obese or under-nourished may depend on why they fail to control energy intake or why food is not available and not consumed rather than on variations in metabolic rate.

The evidence for increases in efficiency of energy utilisation at rest and work in populations with low energy intakes is tenuous. Nor does it seem that the metabolic or mechanical efficiency of individuals with high or low intakes in a population differ to any marked degree. The principal effects of reducing energy intake are to reduce physical activity and body size. Decreases in RMR representing metabolic adaptation occur particularly in the first weeks but this becomes less important in chronic undernutrition. When energy intakes are raised above normal levels, increases in metabolic rate greater than the increased body size have been found but few studies have taken into account the energy cost of storage. The increases are about 10 – 15% of the RMR and the excess intake ( 27–29 ). This seems to be an increase not much greater than the effect of feeding in normal individuals, i.e. the SDA of foods, and may reflect the energy cost of storage.

There are many phenomena still to be explained, which adaptive efficiency could do, e.g. the difficulty in reducing the obese. Also, it is a common finding in overfeeding studies that the fate of much of the excess intakes cannot be explained. Neither energy stores nor MR are increased to the extent expected, which suggests methodological inaccuracies. A number of mechanisms of adaptive efficiency have been considered, e.g. protein turnover and ionic pumping, perhaps under the influence of altered thyroid metabolism. Recently, considerable attention has been given to brown adipose tissue (BAT) although its role in man is not clear (36). Direct evidence to implicate BAT as a heat producer in adult man has not yet been obtained (37).

Since the 1973 report was published, a joint FAO/WHO informal gathering of experts has made further recommendations on some aspects of the report (38, 39) but did not comment on variation and adaptation. Neither have the U.K. and U.S.A. national committees revising requirements considered them.

If minimum energy requirements are being set, the metabolic adaptation in resting metabolism during undernutrition, may need to be taken into account. The causes of the variation in energy metabolism during overfeeding need to be identified before such variation can be considered as adaptive changes in efficiency, although this will probably test the limits of the method. To adjust requirements for variations in efficiency of energy utilisation beyond those due to normal variation, effects of activity and size, etc. appears premature.


1. F.A.O./W.H.O. (1973). Energy and protein requirements. F.A.O. Nutrition Meetings. Rep. Series No. 51, FAO, Rome.

2. Durnin, J.V.G.A., Edholm, O.G., Miller, D.S. & Waterlow, J.C. (1973). How much food does man require? Nature, 242, 418.

3. Edmundson, W. (1980). Adaptation to undernutrition: how much food does man need? Soc. Sci. & Med. 14D, 119–126.

4. Miller, D.S., Mumford, P. & Stock, M.J. (1967). Gluttony 2. Thermogenesis in overeating man. Am. J. Clin. Nutr. 20, 1223–1229.

5. United Nations University. (1979). Protein-energy Requirements under Conditions Prevailing in Developing countries: Current Knowledge and Research Needs. Food & Nutr. Bull. Supplement 1, U.N.U., Tokyo.

6. Durnin, J.V.G.A. (1979). Energy balance in man with particular reference to low intakes. Biblthca Nutr. Dieta, 27, 1–10.

7. Norgan, N.G. & Ferro-Luzzi, A. (1978). Nutrition, Physical Activity, and Physical Fitness in Contrasting Environments. In: Nutrition, Physical Fitness and Health. (Parizkova, J. & Rogozkin, V.A., eds), pp 167–193. University Park Press, Baltimore.

8. Norgan, N.G., Ferro-Luzzi, A. & Durnin, J.V.G.A. (1974). The energy and nutrient intake and the energy expenditure of 204 New Guinean adults. Phil. Trans. R. Soc. Lond. B. 268, 309–348.

9. Brun, T., Bleiberg, F. & Goihman, S. (1981). Energy expenditure of male farmers in dry and rainy seasons in Upper-Volta. Brit. J. Nutr. 45, 67–75.

10. Ashworth, A. (1968). An investigation of very low calorie intakes reported in Jamaica. Brit. J. Nutr. 22, 341–355.

11. Edmundson, W. (1977). Individual variations in work output per unit energy intake in East Java. Ecol. Food Nutr. 6, 147–151.

12. Edmundson, W. (1979). Individual variations in basal metabolic rate and mechanical work efficiency in East Java. Ecol. Food Nutr. 8, 189–195.

13. Lange Anderson, K., Masironi, R., Rutenfranz, J. & Seliger, V. (1978). Habitual Physical Activity and Health. W.H.O., Copenhagen.

14. Spurr, G.B., Barac-Nieto, M. & Maksud, M.G. (1979). Functional assessment of nutritional status: heart rate response to submaximal work. Am. J. Clin. Nutr. 32, 767–778.

15. Viteri, F.E. (1968). Considerations on the effect of nutrition on the body composition and physical working capacity of young Guatemalan adults. In: Amino-acid Fortification of Protein Foods (Scrimshaw, N.S. & Aetschul, A.M., eds) pp 350–375, M.I.T. Press, Cambridge.

16. Widdowson, E.M. (1962). Nutritional individuality. Proc. Nutr. Soc. 21, 121–128.

17. Durnin, J.V.G.A. & Passmore, R. (1967). Energy, Work and Leisure. pp 31–39, Heinneman, London.

18. Rose, G.A. & Williams, R.T. (1961). Metabolic studies on large and small eaters. Brit. J. Nutr. 15, 1–9.

19. Keys, A., Brozek, J., Henschel, A., Mickelson, O. & Taylor, H.L. (1950). The Biology of Human Starvation. University of Minnesota Press, Minneapolis, Minn.

20. Grande, F., Anderson, J.T. & Keys, A. (1958). Changes of basal metabolic rate in man in starvation and refeeding. J. Appl. Physiol. 12, 230–238.

21. Garrow, J.S. (1974). Energy Balance and Obesity in Man. North-Holland Publishing Co., Amsterdam.

22. Benedict, F.G., Miles, W.R., Roth, P. & Smith, M. (1919). Human vitality and efficiency under prolonged restricted diet. Publication No. 280, Carnegie Institute of Washington, Washington, D.C.

23. Norgan, N.G. & Durnin, J.V.G.A. (1980). The effect of 6 weeks of overfeeding on the body weight, body composition, and energy metabolism of young men. Am. J. Clin. Nutr. 33, 978–988.

24. Durnin, J.V.G.A. & Norgan, N.G. (1969). Variations in total body metabolism during ‘overfeeding’ in man. J. Physiol. Lond. 202, 106P.

25. Apfelbaum, M., Bostarron, J. & Lactis, J. (1971). Effect of caloric restriction and excessive caloric intake on energy expenditure. Am. J. Clin. Nutr. 24, 1405–1409.

26. Bray, G.A., Whipp, B.J. & Koyal, S.N. (1974). The acute effects of food intake on energy expenditure during cycle ergometry. Am. J. Clin. Nutr. 27, 254–259.

27. Sims, E.A.H. (1976). Experimental obesity, dietary-induced thermogenesis, and their clinical implications. Clinics Endocrin. Met. 5, 377–395.

28. Danforth Jnr., E., Burger, A.G., Goldman, R.F. & Sims, E.A.H. (1978). Thermogenesis during weight gain. In: Recent Advances in Obesity Research II (Bray, G.A., ed), pp 229–236. Newman, London.

29. Dauncey, M.J. (1979). Energy metabolism in man and the influence of diet and temperature: a review. J. Hum. Nutr. 33, 259–269.

30. Dauncey, M.J. (1980). Metabolic effects of altering the 24 h energy intake in man, using direct and indirect calorimetry. Brit. J. Nutr. 43, 257–269.

31. Sims, E.A.H., Danforth, Jnr., E., Horton, E.S., Bray, G.A., Glennon, J.A. & Salans, L.B. (1973). Endocrine and metabolic effects of experimental obesity in man. Rec. Prog. Horm. Res. 29, 457–487.

32. Whipp, B.J., Bray, G.A. & Koyal, S.N. (1973). Exercise energetics in normal man following acute weight gain. Am. J. Clin. Nutr. 26, 1284–1286.

33. Hanson, J.S. (1973). Exercise responses following production of experimental obesity. J. Appl. Physiol. 35, 587–591.

34. Garrow, J.S. (1978). The regulation of energy expenditure in man. In: Recent Advances in Obesity Research II (Bray, C.A., ed), pp. 200–210. Newman, London.

35. Goldman, R.F., Haisman, M.F., Bynum, G., Horton, E.S. & Sims, E.A.H. (1975). Experimental obesity in man: metabolic rate in relation to dietary intake. In: Obesity in Perspective (Bray, G.A., ed), pp 165–186. U.S. Govt. Ptg. Office, Washington, D.C.

36. Rothwell, N.J. & Stock, M.J. (1979). A role for brown adipose tissue in diet-induced thermogenesis. Nature, 281, 31–35.

37. Trayhurn, P. & James, W.P.T. (1981). Thermogenesis - a control mechanism in obesity? 1: Brown fat may play an important part. Nutr. Bull. Lond., 6, 15–22.

38. F.A.O./W.H.O. (1975). Energy and protein requirements: recommendations by a joint FAO/WHO informal gathering of experts. Food & Nutr. 1, 11–19.

39. F.A.O./W.H.O. (1979). Protein and energy requirements: a joint FAO/WHO Memorandum. Bull. Wrld. Hlth. Org. 57, 65–79.

Top of Page