Estimated Mean per Capita Energy Requirements for Planning Emergency Food Aid Rations (1995)

◆ ◆ ◆ ◆ ◆

In the past several decades numerous examples of famine and forced dislocations have required the implementation of massive food aid programs. In large feeding programs such as refugee feeding, planning for food procurement and distribution is critical. Estimates of initial food needs are essential to begin the process of procurement. The most critical factors in arriving at these initial estimates are the size of the population to be fed and an estimate of the average daily energy need. Clearly, nutrients other than energy are important considerations, but for planning purposes energy need is accepted as the principal concern.

There has been considerable debate regarding the appropriate estimated energy value to be used in planning initial food needs. Prior to 1988, the value of 1,500 kcal (6.3 MJ) per capita was used. This value was recognized to be less than the actual energy requirement, but was deemed to be adequate to support survival. In 1989, the World Health Organization (WHO) convened a meeting on “Nutrition in Times of Disaster” at which it was suggested that, for planning purposes, a value of 1,900 kcal per person per day be used for longer term feeding situations and that 1,500 kcal per day be employed in acute survival situations. The values were derived for a typical population using estimated basal metabolic rate (BMR) and assumed activity levels of 7 percent above BMR (i.e., BMR × 1.07) for the 1,500 kcal per day estimate and 45 percent above (BMR × 1.45) for the 1,900 kcal estimate. The value of 1,900 kcal was generally accepted as the benchmark for initial planning purposes until very recently (Rivers and Seaman, 1989).

In 1993, Dr. Graeme Clugston of WHO presented a rationale for using a value of 2,100 kcal (8.8 MJ) rather than the benchmark value of 1,900 kcal (Clugston, 1993). The increase of 200 kcal (837 kJ) per day was accounted for largely by (1) an increase in the activity increment to 55 percent above BMR (from 40 percent), assuming that a physical activity level (PAL) of 1.4 was incompatible with long-term health; (2) some adjustments to the population's composition; and (3) an increase in the proportion of pregnant and lactating women in the population. While 200 kcal per day represents only a 10 to 11 percent increase, it has significant implications for the amount of food required to support food aid programs.

As will also be evident in this report, the major variable influencing estimates of the mean energy “requirement” for planning purposes is the activity increment. It is difficult, and in many cases impossible, to initially establish the activity level of a population. Therefore, not surprisingly, the various estimates of the mean per capita requirement are not well supported by detailed studies. This acknowledgment reinforces the need to develop and rapidly apply a standardized protocol to monitor physical activity levels in order to refine initial estimates of food needs.

The underlying assumptions and the methods available for estimating the energy needs of individuals, classes of people, and of populations are described in the FAO/WHO/UNU report, “Energy and Protein Requirements ” (1985). As defined in that report:

the energy requirement of an individual is the level of energy intake from food that will balance energy expenditure when the individual has a size and body composition, and level of physical activity, consistent with long-term good health; and that will allow for the maintenance of economically necessary and socially desirable physical activity. In children and pregnant or lactating women the energy requirement includes the energy needs associated with the deposition of tissues or the secretion of milk at rates consistent with good health (p. 12).

There is only one level of intake that balances the energy requirement of an individual. If energy intake is below this requirement, there will be a reduction in body weight and fat, and in lean body mass and activity if the deficit is sufficiently severe or prolonged. If intake is more than required, the individual will gain primarily fat. When sufficient food energy is available, physiological mechanisms control appetite so that individuals can regulate their intake to maintain energy balance. While energy balance and health can be maintained over a range of body weights for the same individual, severe or prolonged weight loss

will have adverse effects on human function (such as ability to sustain physical activity) and health. There is little evidence of an increase in the efficiency of energy utilization in individuals deprived of dietary energy; a reduction in basal metabolic rate in this situation primarily reflects a loss of lean body mass.

The CIN recognizes that for a given class of individuals, or population group, there will be a distribution of energy requirements among those in the group. As suggested by FAO/WHO/UNU (1985), however, the appropriate single descriptor of the energy needs for a defined group of individuals is “the average of the individual requirements, without specific provision for the known individual variation in requirement” (p. 14). Hence, for the present purposes it is entirely justifiable to derive an estimated mean per capita energy requirement (EMPCER), as described below. While there will be individuals for whom this value is insufficient, there will also be individuals for whom it is in excess, and it is assumed for the present that the prevailing circumstances will permit individuals to make selections according to whether their needs either exceed or fall below the EMPCER. It is the CIN's view that, in reference to emergency situations, it is not an appropriate strategy to add into the values, derived below, an additional safety factor to cover the needs for persons with an above-average requirement, since this would result in food wastage and unavailability in other instances. Also, any further discussion of this issue would need to involve a large number of considerations that are more appropriately programmatic in nature and, thus, beyond the scope of the committee's charge.

The two main determinants of energy requirements are basal metabolic rate (comprising about 60 percent of the total) and physical activity (almost all of the remainder, which also includes 5–10 percent expended in dietary-induced thermogenesis).

The BMR is the energy cost of metabolism in an individual who is completely at rest, in a thermoneutral environment, and fasting. Because this measurement is impractical in most situations, equations have been developed from which BMR can be estimated from the sex, age category, and weight of an individual (FAO/WHO/UNU, 1985, p. 171). For example, for a man age 30–60 years, energy requirements (in kcal/d) for BMR are 11.6 × body weight (kg) + 879. The inclusion of data on height in addition to weight does not significantly improve the reliability of the BMR estimate.

These equations were developed from data on well-nourished, Western individuals and are theoretically valid only for individuals who fall within the

“desirable” weight range for their height. Information is available on the lower, average and upper limits of this desirable weight range, for an individual of a specified height (FAO/WHO/UNU, 1985, p. 183). To estimate the BMR of an individual who is underweight for their height, either their actual weight or the mean desirable weight for their height, can be used in the estimation equations. In the latter situation height must be measured, or can be estimated using locally available or published data (James and Schofield, 1990, pp. 106–115) for the region or location. If the decision is made to use desirable weights, estimated energy requirements for BMR will be slightly higher for the underweight individual than if actual weights had been used. Using the higher value will allow for some recovery of body weight although this would be very slow given that a gain of 1 kg will require about 5,000 kcal (20.9 MJ) of additional energy. As discussed below, a substantial energy increment would be needed to achieve a reasonably rapid recovery of weight lost by a population that now has a low BMI.

Because energy expended in physical activity is the most variable component of total energy requirements, it is important to make reasonable estimates of usual activity levels. For practical purposes, in the nonpregnant, nonlactating individual, energy requirements can be calculated from:

total energy requirements = BMR × PAL,

where PAL is the physical activity level. The PAL values are expressed as multiples of BMR, setting the value of BMR as 1; in this way a correction for age and sex is already incorporated into the BMR. Published values are available for a wide variety of activities (FAO/WHO/UNU, 1985; James and Schofield, 1990).

Rough estimates of habitual energy expenditure can be obtained by classifying the activity level of an individual or group as light, moderate, or heavy (see Table 2-1). Examples of the mean PAL for different types of activities are provided in Table 2-2 and Table 2-3. PAL figures for these categories have been proposed by FAO/WHO/UNU, assuming that detailed information on activity is not available (1985, p. 78). These PAL values were derived primarily from indirect calorimetry, and may be revised in the future based on new information obtained by the doubly-labeled water method of measuring energy expenditure (Schulz and Schoeller, 1994; Torun et al., 1994).

The PAL values in Table 2-1 differ for men and women, especially at moderate and high levels of activity, based on the assumption that they are engaged in different activities. They should be adjusted for specific situations if information is available, or can be deduced, about how people are actually spending their time.

The 1988 WHO calculations that arrived at a mean value of 1,900 calories assumed an average PAL of approximately 1.4 for adult males and females (Rivers and Seaman, 1989). An energy expenditure of 1.4 times the BMR is considered to be the minimum for individuals who are not engaged in any occupational or discretionary activities (FAO/WHO/UNU, 1985, p. 73). It is only slightly above the PAL of 1.27 established as a “survival requirement” of practical value only in conditions of crisis, for estimating the short-term needs of “totally inactive, dependent people” (FAO/WHO/UNU, 1985, p. 73). A PAL of 1.27 allows for minimal movement, is incompatible with long-term health, and allows no energy expenditure in food preparation.

The committee was not aware of any empirical studies of activity patterns among refugee and other populations requiring food aid. However, based on minimal information, such as the fact that the majority of food is distributed as dry rations, which must be carried and prepared by household members, the committee examined the appropriateness of using the 1985 FAO/WHO/UNU PALs for light activity (1.55 for males; 1.56 for females), rather than the PAL of 1.4 used as a basis of the 1,900 kcal per day “survival requirement.” To accomplish this, the committee estimated a daily activity pattern for adult males and females in refugee camp situations; these estimates are presented in Table 2-2 and Table 2-3.*

SOURCE: FAO/WHO/UNU (1985).

Assumptions for Female Activity Levels  Time use studies of rural women in developing countries find that, on average, women work 10 to 12 hours per day, about evenly divided between household maintenance activities and agricultural or income-generating work (see Leslie, 1989, and Leslie et al., 1988). The committee assumed that women do not continue their agricultural/income-generating work in a refugee situation, but that they do continue many of their household maintenance tasks such as cooking, carrying (e.g., fetching water or fuel, carrying dry rations) and child care. For the balance of their nonsleeping time, the committee estimated a very sedentary activity level (e.g., lying or sitting quietly), except for 2 hours of more active discretionary activities such as socializing or strolling around. Making these assumptions, the committee estimated an average PAL of 1.53 for adult women in refugee camp situations (Table 2-2).

The estimates of activity level in Table 2-2 may be conservative. The inclusion of 8 hours a day of sitting or lying quietly is probably an outside limit of the amount of time a person who is awake, even in a refugee camp situation, would spend essentially not moving. If later research or field observations suggest that women are spending a considerable amount of time each day walking, or in more energy intensive household maintenance activities such as washing clothes, pounding grain, etc., or in any agricultural activities, assumptions about average PAL should be increased. For example, the substitution of 4 hours of agricultural work at a PAL of 2.8 for 4 hours of lying or sitting quietly, would increase the average PAL for women to 1.8.

Assumptions for Male Activity Levels  Time use studies of rural men in developing countries find more variability in total work time and in the energy intensity of activities than is the case for rural women, but on average total work hours for rural men are usually found to be slightly lower than total work hours

for rural women (see Leslie, 1988, and Leslie et al., 1989). The committee assumed that this would continue to be the case in a refugee camp situation, and in particular, that men would not be engaged in the seasonal, high-energy-expenditure agricultural activities that are typical of subsistence farmers. Therefore, the committee assumed only 3 hours a day of “work,” 1 hour of which is carrying loads (e.g., fuel, water, dry rations, and household goods), and 2 hours of which are sitting tasks, which might include carving, helping with cooking, and sharpening tools. As for women, the committee estimated that 8 hours a day of standing and sitting quietly is a maximum, and that the rest of the men's waking time would be spent in sedentary recreation or in walking around or strolling. Making these assumptions, the committee estimated an average PAL of 1.53 for adult males in refugee camp situations (Table 2-3), the same value as for adult females.

The estimates of activity level in Table 2-3, like those in Table 2-2, may be conservative. If information indicates that men are significantly more active than is suggested by the time expenditure pattern in Table 2-3, for example, because they have long distances to walk, or are assigned manual labor in the camps or agricultural work, then adjustments should be made in the average physical activity level. As an example, if 4 hours of energy-intensive agricultural work or manual labor at a PAL of 3.5 is substituted for 4 hours of lying or sitting quietly, this would increase the average PAL for men to 1.9.

Based on the PAL estimates in Table 2-2 and Table 2-3, the committee concluded that a PAL of 1.4 probably underestimates the activity levels of adult males and females in refugee situations and that it is reasonable to apply the FAO/WHO/UNU PALs for light activity (1.55 for males and 1.56 for females) to populations in these situations. Therefore, most subsequent analyses in this report use the FAO/WHO/UNU PAL values. Populations in other emergency situations such as earthquakes, floods, or sieges may have higher levels of

activity, so that the committee also provides examples of how the mean per capita energy requirement would be altered in situations of this type.

For children, there is an additional energy requirement for growth. For all practical purposes, this is a very small percentage of total requirements after the first few months of life. Recommended increments throughout pregnancy are 285 kcal (1.2 MJ)/d (FAO/WHO/UNU, 1985). It has been suggested that future pregnancy allowances should be based on: (1) the BMR of the nonpregnant, nonlactating woman (i.e., on prepregnancy weight) multiplied by an average PAL appropriate to her usual activity level (using the same PAL values for each type of activity as for the nonpregnant woman), and (2) a daily increment throughout all three trimesters of pregnancy that averages about 250 kcal (1.0 MJ)/d (Prentice et al., 1994). Estimates based on this newer approach are only slightly lower than the 285 kcal/d increment proposed by FAO/WHO/UNU (1985). Even though there appear to be adaptive mechanisms that reduce the usual pregnancy BMR increment in undernourished women, these are not considered to be a desirable means of meeting the energy costs of reproduction (Prentice et al., 1994).

The additional energy requirements for lactation were set by FAO/WHO/UNU at 500 kcal (2.1 MJ)/d, assuming that in addition to this, 200 kcal are available from maternal fat stored during pregnancy. Even in well-nourished women little of this fat seems to be used during breastfeeding because appetite is increased, and in populations where dietary energy is inadequate, it is not appropriate to assume that the use of fat stores for lactation is feasible or desirable. As for pregnancy, it has been proposed that future estimates of the energy requirements for lactation should be based on the BMR of the nonpregnant, nonlactating woman, times a PAL that depends on the duration and type of activities performed, plus an increment for lactation that does not assume maternal fat loss and that falls in the second 6 months, assuming that breastmilk is not the sole source of food for the infant (Prentice et al., 1994).

Individual energy requirements can be used to plan the food energy needs of a population, using the following three sets of information: (1) the proportion of individuals in the total population that belong to specific subclasses defined in

terms of age and physiological status (i.e., whether pregnant or lactating); (2) the average actual, or average desirable, body size (hence BMR) of the people in each subclass; and (3) the population 's average level of physical activity. If this information is not available for a specific country, certain assumptions can be made, as described below. As noted earlier, to expedite decisions about the immediate purchasing and shipping of emergency food rations to refugee and other populations requiring food aid, the population 's energy needs can be expressed as a single value: the EMPCER. The EMPCER is calculated as the weighted average of the energy needs of each subgroup of the population. It can be modified, as necessary, to allow for special factors affecting a particular population. As discussed below, it may be desirable to increase the amount of energy allotted to adjust for rehabilitation of preexisting undernutrition or cold environmental conditions.

The first EMPCER calculated by the committee—2,076 kcal (8.7 MJ)/d—estimates the energy needs of a “typical” population in a developing country (see Table 2-4). The numbers presented in Table 2-4 are based on the following assumptions: (1) the population is distributed as indicated in the World Population Profile 1994 report (Jamison and Hobbs, 1994) for developing countries; (2) the average height of adult males is 170 cm and of adult females is 155 cm, which are the approximate heights of average males and females in sub-Saharan Africa and slightly greater than those of adults in South and Southeast Asia (James and Schofield, 1990); (3) the weights of these adults are at the median for U.S. adults of the stated heights (FAO/WHO/UNU, 1985); and (4) the total energy expenditure of the adults is 1.55 and 1.56 times the BMR for both males and females, respectively. The median weight-for-height of U.S. males was chosen because there is no international reference for adult weight-for-height, and the choice of the U.S. population would make for a conservative estimate of the EMPCER for populations in most developing countries. The PALs are those suggested for a light level of physical activity by FAO/WHO/UNU (1985), which were found by the CIN to be the minimum reasonable for both men and women in emergency feeding situations (Table 2-2 and 2-3), and which will need to be increased where individuals are not confined or are engaged in nonsurvival activities.

For individuals below 18 years of age, the values in Table 2-4 are based on data from affluent populations. Although the individuals from whom these data were derived are larger (and therefore have a greater BMR) than many children and adolescents from low-income countries, this “extra” allotment for children in developing countries is deemed appropriate because they could presumably benefit from the additional food for compensatory growth. The increments in

energy requirements for pregnant and lactating women were assumed to be 285 and 500 kcal per day (FAO/WHO/UNU, 1985).

To examine how different assumptions might influence the committee 's EMPCER, we explored the effects of using varied population distributions, different assumed body sizes, four distinct levels of physical activity, and different proportions of infants who are breastfeeding. The impact of each of these assumptions on the EMPCER is discussed below.

The effect of different population distributions was studied by substituting the usual demographic distribution of industrialized countries for that of developing countries (Jamison and Hobbs, 1994; see Table 2-5). All other assumptions regarding adult body size and level of physical activity were held constant as described for Table 2-4. As indicated in Table 2-5, this change in demographic distribution had relatively little effect on the EMPCER, changing the estimate from 2,076 kcal/d to 2,105 kcal (8.8 MJ)/d.

The committee further examined how the EMPCER might change with unusual population distributions, such as might occur in a situation in which women and children migrate, but most men do not. For this exercise, the committee assumed that 50 percent of the males aged 10–17 years and 75 percent of the males from 18–59 years from the original population distribution for developing countries were no longer present in the index population (a situation that might conceivably occur in certain refugee situations). As indicated in Table 2-5, this modification in the assumed demographic distribution reduced the EMPCER to 1,936 kcal (8.1 MJ)/d. Because population distribution had a

minor impact on the EMPCER (except in the case of unusual migration patterns), the decision was made by the committee to use the typical demographic distribution of developing countries in all subsequent analyses. When information is available on unusual demographic distributions due to differential migration of subclasses in a population, however, the committee suggests that the EMPCER be revised accordingly.

We next examined the effects of different body sizes on the average BMR of adults and on the population's EMPCER. For these analyses, the average heights (James and Schofield, 1990) of three sets of men and women from the following areas were compared: South and Southeast Asia, sub-Saharan Africa, and the United States of America. The average heights of each set of individuals and the U.S. mean weight for those heights (FAO/WHO/UNU, 1985) are shown in Table 2-6. The table also presents the EMPCER for each of these hypothetical populations. Although there was relatively little difference in the EMPCERs calculated for the two developing country populations, the estimated population requirement for individuals in the United States was approximately 120–150 kcal/d greater. Thus, when the body size of the recipient population is close to that in developed countries, the EMPCER will be about 100 kcal/d higher, or closer to 2,200 kcal (9.2 MJ)/d.

The next analyses compared the effects of different assumed physical activity levels (PALs). The typical population distribution of developing countries and the reported average heights of adults in these settings were used, as described in the foregoing paragraphs.

As indicated in Table 2-7, the level of assumed physical activity had a relatively marked impact on the EMPCER, which ranged from 1,908 kcal (8.0 MJ)/d in populations with a minimal (“survival”) level of total energy

expenditure of 1.40 BMR, to 2,535 kcal (10.6 MJ)/d in populations with a high level of energy expenditure (1.96 BMR). Because it is difficult to predict the true level of physical activity without specific information from the field, the committee decided that it would be appropriate to use the FAO/WHO/UNU PAL values for light activity, i.e., 1.55 for males and 1.56 for females until further field data in a given situation become available.

A possible refinement of the original analysis presented in Table 2-4 is the separate consideration of energy needs for young infants who are breastfeeding. Because these infants have a reduced need for energy supplied from food sources, it may be desirable to take into account the proportion of infants who are breastfeeding. The calculations in Table 2-8 assume that all children from birth to 6 months of age are predominantly breastfeeding and that they will have no additional energy needs. This analysis further assumes that half the infants from 6–11 months of age are still breastfeeding and that breastmilk will provide half their energy needs. Although these corrections are theoretically appropriate to avoid “double counting” of the energy needs of these children, the impact on the EMPCER is minimal, reducing the EMPCER from 2,076 special correction for rates of breastfeeding was necessary.

Although the EMPCER for the total population does not need to be adjusted for breastfeeding, it should be recognized that special foods, such as breastmilk substitutes or processed complementary foods, may be required for children less than 2 years of age in situations where either (1) their mothers are unable or unavailable to breastfeed or (2) the relief foods that are provided for older children and adults cannot be prepared in a form easily consumed by young children. In these cases, the population EMPCER must be calculated separately for the specific subgroups of young children to determine the amounts of these special

foods that are needed. As with the EMPCER for the general population, the EMPCER for the subgroup(s) would be the product of the number of individuals in the subgroup multiplied by their average total daily energy requirement. In the case of complementary foods, the estimated amount of energy provided by breast milk would be subtracted from the children's average total daily energy requirement.

It may be desirable to adjust the committee's EMPCER for such special situations as preexisting level of undernutrition or environmental cold stress. These factors are described below.

Recovery from malnutrition  The theoretical energy needs for convalescent feeding of malnourished populations can be calculated from: (1) the magnitude of the observed weight deficit (assuming that the minimum acceptable body mass index [BMI] is 18.5 kg/m² [James et al., 1988]), (2) the energy cost of weight gain, and (3) the desired duration of the recovery period. If we assume, for example, that the average BMI of all adults is 17 kg/m² for a population that has an average height of 1.62 m, then the existing mean weight deficit is 3.937 kg. The energy cost of tissue synthesis is assumed to be 5 kcal/g (FAO/WHO/UNU, 1985); this is a value calculated for 12-month-old children recovering rapidly from malnutrition, but is probably the best estimate available. Thus, the energy required to replace a 3.937 kg weight deficit would be 3,937 g × 5 kcal/g, or nearly 20,000 kcal (83.7 MJ). To replace this deficit within a period of 30 days, an additional 667 kcal (2.8 MJ)/d would be necessary for each adult. If approximately 60 percent of the population are adults, this would imply a need to increase the EMPCER by 400 kcal (1.7 MJ)/d. Table 2-9 provides estimates of the additional energy theoretically required each day for different degrees of wasting and desired rates of nutritional repletion. In reality it is difficult to recommend an appropriate rate of refeeding. This will depend on factors such as the distribution and extent of weight loss in the population, the availability of limited rations, and the extent to which the additional food would be consumed versus sold or exchanged. In some cases it may be necessary to limit the amount of food intake during the early stages of recovery from severe undernutrition to avoid complications of malabsorption or sodium and fluid overload. When specific infectious or metabolic diseases accompany severe malnutrition, these former conditions often must be treated successfully before nutritional recovery can occur.

The committee also considered the possible need to provide additional food supplements for children with preexisting malnutrition. However, because the energy requirements described above for children were calculated according to

the body size of reference children in North America, there is already supplemental energy available for those children in the target population who are undernourished. For example, a 12-month-old child with length-for-age at the 50th percentile (76 cm) and weight-for-length at the 3rdpercentile (8.5 kg) would actually be allotted 125 kcal (523 kJ)/kg body weight/d, because the calculations of the EMPCER allow 1,050 kcal (4.4 MJ)/d for children of this age. This is substantially higher than the recommended intake for this age (101 kcal/kg). Children who are both stunted and wasted, or more wasted than in the foregoing example, would be allotted even more energy on an actual body weight basis. These calculations (and similar ones carried out for older children) indicate that the assumptions used for calculation of the EMPCER already provide sufficient energy for nutritional rehabilitation of undernourished children. Thus, special adjustments for undernourished populations are necessary only when the adults are undernourished. On the other hand, special foods with high energy and nutrient density provided in a liquid or semi-liquid form, may be necessary for the rehabilitation of young children.

Cold environments  Living in a cold environment increases energy requirements for both BMR and physical activity, especially if clothing is inadequate. Women clad in a thin cotton trouser suit expended 7 percent more energy

per day in a metabolic chamber maintained at 22^° C compared to 28^° C; at 22^° C they felt cool, chilly at times, had occasional piloerection, but did not shiver (Dauncey, 1981). In another study, even when men could select the clothing they needed to be comfortable, energy requirements in a metabolic chamber averaged about 5 percent higher at 20^° than at 28^° C (Warwick and Bushy, 1990). A 1 percent increase per degree C below 20^° C, with the temperature estimate reduced by 5^° C in windy conditions, is recommended by the Centers for Disease Control and Prevention (1992) and the ACC/SCN (Schofield, 1994). Using this recommendation, based on an EMPCER of 2,100 kcal/d, an additional 100 kcal (418 kJ)/d would be needed at 15^° C, 200 kcal (837 kJ)/d at 10^° C, and 300 kcal (1.3 MJ)/d at 5^° C, although the increment should also depend on whether the preeminent needs for clothing and shelter are met.