Previous - Next
A. FERRO-LUZZI, G.
PASTORE and S. SETTE*
* Unit of Human Nutrition, National Institute of Nutrition, Via Ardeatina 546, Rome, Italy.
1. Introduction
2. Reasons of seasonal variations in energy metabolism
3. Seasonal body weight fluctuations
4. Seasonal fluctuations of energy expenditure
5. Conclusions
6. Research priorities
References
The phenomenon of
seasonality has recently attracted a lively interest, as
testified by the rapidly expanding body of literature on this
topic. In spite of this, the evidence gathered so far is
inconclusive, often incomplete or contradictory or anecdotal.
In reviewing the literature on seasonality, it has been difficult in several cases to separate occasional from regularly recurring climatic or biological fluctuations, the two merging almost insensibly. The lack of distinction between the two situations might lead to unwarranted generalizations regarding the severity, frequency and geographic distribution of seasonal cycling of energy metabolism. In the context of this paper, seasonality refers to a regularly recurring set of conditions leading to approximately annual alternations of restricted and unrestricted access to food energy, often coinciding with periods of variable demand for physical labour. With this definition in mind, situations resulting either from chronic food deprivation or from acute famine have been left out.
The first
part of this paper deals with the environmental characteristics
conducive to seasonal cycling in food availability and workloads.
The second part is a review of the literature on the actual
impact of seasonality on energy metabolism and has been divided
into two sections. The first comprises studies describing the
occurrence of body weight changes, taken as a proxy of energy
imbalance. The second section includes studies describing
seasonal fluctuations of BMR and energy expenditure, with or
without concomitant body weight changes. Studies on the
fluctuation in food availability and energy intake have been
omitted, unless evidence is presented that these functions
resulted in energy imbalance.
Climatic-environmental
conditions have an enormous influence on the vegetative cycle of
the food and cash crops on which the whole economy of rural
households in Less Developed Countries (LDC) depends. The same
factors that regulate the agricultural cycle can therefore be
expected to impinge on the energy metabolism of the individuals.
Essential factors are: temperature, water availability, nature
and structure of the soil, solar radiation, and variation in the
photoperiod.
In equatorial or tropical areas, where temperature, solar radiation and photoperiod do not undergo substantial seasonal fluctuations, the only factor that limits vegetal development is water availability. For example, even sorghum, a drought-resistant vegetal species, needs about 250,000 L of water per ton of total dry matter produced. Since artificial irrigation is rarely possible in LDC, rain and other factors which limit or increase the loss of water through percolation, evaporation and/or transpiration (i.e., the nature of the ground) are critical. Therefore, the best indicators of agricultural production potential in equatorial and tropical areas are absolute and relative seasonality, derived from pluviometric data.
Absolute seasonality is the proportion of dry months among all months of a year. Dryness is not an absolute value depending only on pluviometric data; it is corrected for the evapotranspiration in relation to the culture, temperature, exposure to winds, relative humidity, nature and granulometrical structure of the soil, direct and indirect radiation, etc. The range of absolute seasonality extends from 0 (sufficient rains all year long) to 1 (total lack of a period favouring the optimal growth of plants).
Relative seasonality is the ratio between the amount of rain in different months and annual pluviosity. This index ranges from 0 (no seasonality) to 1.8 (extreme seasonality).
By combining data on annual rainfall with values of absolute and relative seasonality, we have constructed a rough, global index; it describes the likelihood and intensity of seasonal fluctuations on simultaneous demand and availability of food energy (Figure 1). Desert areas, characterized by permanent shortage, represent a peculiar and extreme situation and have not been included in this analysis. The remaining areas are subdivided into three categories.
In the very low seasonality category are areas with annual rainfall of more than 1000 mm, reasonably distributed throughout the year, and with an index of absolute seasonality of less than 0.40. No seasonal interference with the energy balance of the population is expected.
In the moderate seasonality category are areas with an annual rainfall between 500 and 1000 mm, an absolute seasonality between 0.40 and 0.75, and a relative seasonality between 0.8 and 1. The high absolute pluviosity or a bimodal distribution of rains ensures that the vegetative cycle is prolonged. In these areas, the likelihood of seasonal fluctuations of energy metabolism of individuals is very limited. Since agricultural work is spread throughout the year and food crops are harvested on a semicontinuous basis, energy is available at all times, without occurrence of a "hungry season", except in particular situations such as rain failure in unusually dry years.
In the severe seasonality category are areas with an annual pluviosity of less than 500 mm, an absolute seasonality index higher than 0.75, and values of relative seasonality higher than 1. In these areas, the vegetative period is very brief, and farmers are forced to concentrate all their physical efforts on a time usually coinciding with depleted food stores. Furthermore, the shortage of water imposes wide spacing between furrows, and the brevity of the vegetative period creates the need to utilize seed cultivars which have a short cycle and are less productive than those with a longer cycle. All this leads to a reduced potential for food production in these areas. BAYLISS-SMITH (1981) calculated that, in one year, a hectare of soil in an area with an absolute seasonality index below 0.5 and a relative seasonality index below 0.6 will theoretically produce 40 to 44 tons of total dry matter. An area where the two indexes are above 0.8 can, in the same period and for the same unit of surface, produce at most a theoretical quantity of 10 tons of total dry matter.
Figure 1. Grading of climatic seasonality of some areas of the world (see text for explanation).
The conclusions are obvious:
1. In LDC areas with an essentially agricultural economy, climatic seasonality is undoubtedly the overriding cause of seasonal energy imbalance of the population, either through its impact on energy expenditure or on food energy availability, or on both.
2. The statement that seasonality is "... the rule among adults in rural areas of developing countries" (TEOKUL, PAYNE and DUGDALE, 1986) represents an unwarranted generalization.
3. In areas with non-seasonal or bimodal climatic conditions, major fluctuations of human energy balance are not to be expected.
4. As climatic seasonality increases, physical labour demand - and thereby energy expenditure - tends to concentrate in a brief annual bout. Food availability is limited in a similar seasonal manner. The two fluctuations are often asynchronous, compound each other's effect and cause marked seasonal imbalances of energy metabolism.
Profound and long-term
physiological fluctuations of energy metabolism occur in several
animal species: premigratory body fat deposition in birds,
fattening in bears and marmots for hibernation, tail fat or hump
fat reserves in certain sheep, zebus and camels provide examples
of a purposeful accumulation of energy in the body in
anticipation of regularly recurring energy deficits. In man,
short-term fluctuations of energy balance probably occur all the
time, in circadian rhythm, and his body appears to be
well-equipped to mobilize stored energy. Body fat therefore may
be legitimately regarded as a specialized organ, capable of
correcting energy imbalance in the short term by mobilization of
body energy stores to meet immediate needs as one of the early,
if not the earliest, wholly physiological responses of the body.
Serial data on body weight therefore should be an excellent
operational indicator of the dynamics of energy metabolism and of
adaptation mechanisms at work, in the context of seasonality.
Since children are the most
vulnerable and sensitive members of the community, it is obvious
to look at them in search of early symptoms of seasonal impact.
Evidently, crude weight changes can not be a suitable indicator
during childhood, but growth rates can represent an acceptable
model for investigating seasonal pressures on the energy
metabolism of the child, while birth weight can be an indicator
of maternal and foetal energy status combined.
Seasonal differences in mean birth weight, ranging from 150 to 450 9, have been reported in several countries where marked seasonality in food availability and/or in work demand exists, such as rural Taiwan (ADAGE, 1984) and rural Gambia (PRENTICE et al., 1983) (Figure 2). The picture is complicated, however, by the unexpected fact that seasonal differences in birth weight have been reported also in areas where climatic and agricultural seasonality are small, if at all present (e.g., in New Guinea; CRITTENDEN and BAINES, 1986). To explain this observation, special external pressures were thought to operate, disrupting the fragile New Guinean ecosystem, and the precarious agricultural and socioeconomic system. This disruption supposedly resulted in a great amplification of the originally very modest seasonal cycling.
Marked seasonal differences in weight and in skinfold thick-nesses have been described in children of Sahelian pastoralist populations, the Wodaabe (LOUTAN and LAMOTTE, 1984) and the Ferlo (BENEFICE, CHEVASSUS-AGNES and BARRAL, 1984). AS can be seen in Figure 3, the impact on Ferlo children depends on age. Note that the growth disturbances occurred at different times of the year in the two ethnic groups, despite their shared geographic and climatic ecozone. This lack of synchronism highlights the complexity of the interaction between environmental seasonality and sociocultural, physiological and pathological variables. Upper Volta children living in a country with high seasonality index did not show the expected seasonal changes in their growth rates (SERRE, 1955). New Guinean children from the Nembi plateau (CRITTENDEN and BAINES, 1986) were also peculiar as they exhibited seasonal patterns of growth, in spite of the absence of climatic and agricultural seasonality (Figure 4). The picture in this case was further complicated by the observation that the children born during one part of the year with low birth weights, had a better postnatal growth performance than the children born in the "better" part of the year with higher birth weights. Seasonal patterns in growth performance were found to occur also in urban areas (TOMKINS et al., 1986), closely reproducing the rural pattern (ROWLAND, COLE and WHITEHEAD, 1977) with lowest increments during the rainy season, and compensatory catch-up growth during the dry months (Figure 5). These findings were unexpected because of the cash economy of the urban environment and the steady demand for women labour throughout the year.
In trying to understand these findings, several ideas have been put forward. Infectious diseases may play a major role in the urban Gambian environment (TOMKINS et al., 1986). For the New Guinean children of the Nembi plateau, the initial benefit of a higher birth weight during the "good" part of the year may be counterbalanced by an earlier exposure to weaning stress and the accompanying loss of passive immunity, because mothers resumed their field work when prevalence of infectious diseases was the highest of the year (CRITTENDEN and BAINES. 1986).
Figure 3. Seasonal growth performance of two Sahelian pastoralist populations:
- Wodaabe (LOUTAN and LAMOTTE, 1984): weight changes (in absolute values) of children aged 1 to 5 years.
- Ferlo (BENEFICE, CHEVASSUS AGNES and BARRAL, 1984): mean weight-for-height changes in different age groups.
In conclusion, seasonal fluctuations in birth weight and in infant and child growth performances have been reported from various parts of the world; they have been considered to reflect the seasonal occurrence of a powerful synergism between mother's workload and food shortages. However, the occurrence of fluctuations in growth velocities in areas of the world without seasonality suggests that factors other than primary energy imbalances may play an important role, namely the close interrelation between environmental seasonality, infectious diseases and sociocultural characteristics. Furthermore, it is impossible to evaluate the extent to which the observed growth retardation represents the direct expression of a primary perturbation of the energy metabolism of the child, rather than being an unspecific reactive body response to a multiplicity of environmental stress-ors. It is important to remember that seasonal fluctuations in weight and height gains have been described also in children of affluent societies, even when controlling for infectious diseases. This should be kept in mind when examining seasonal growth disturbances in developing countries, as a significant component of it "may be found to be due to factors other than lack of food or incidence of disease" (PAYNE, in press).
Adults differ from children
insofar as they are relatively less prone to infectious diseases,
have a better command over food intake, and a larger component of
energy expenditure is dictated by obligatory physical activity in
relation to work and social duties. Therefore, it is justified to
regard seasonal changes in body weight of the adult population as
an excellent proxy for energy imbalances, uncomplicated by
interference of unspecific stress-ors and confounding variables.
Seasonal body weight fluctuations of adults have been recorded in several developing countries where a decrease in body weight has been observed during the preharvest "hungry" period, with a few notable exceptions to be described later. The postharvest "glut" generally led to a rapid recovery of the lost weight, and up to the highest yearly value. Rural populations, mostly farmers, but also some nomadic groups, have been studied more intensively. Most studies have been undertaken in Africa. We could not find any study of seasonal body weight changes of urban adults in developing countries. Only very few studies report not having found seasonal fluctuations of body weight, but it is not clear whether this is due to a lack of interest in reporting a negative finding rather than reflecting an actual rarity of such situations. Absence of seasonal weight losses have been reported by BIDINGER, NAG and BABU (1986) in a semi-arid tropical area with a high seasonality index in India, and by NORGAN and colleagues (1974) in New Guinea. The first case was tentatively interpreted as due to a drought which prevented recovery from previous weight losses (undocumented), the second can be explained by the low seasonality index of the study area.
Figure 6. Seasonal changes in body weight of adult men and women:
1) Upper Volta, farmers (ANCEY, 1974);
2) Upper Volta, farmers (BRUN, BLEIBERG and GOIHMAN, 1981; BLEIBERG, BRUN and GOIHMAN, 1980);
3) Niger, pastoralists (LOUTAN and LAMOTTE, 1984);
4) Senegal, farmers (GESSAIN, 1978);
5) Senegal, pastoralists (BENEFICE and CHEVASSUS-AGNES, 1984);
6) Senegal, farmers (ROSETTA, 1986);
7) Mali, farmers (BENEFICE and CHEVASSUS-AGNES, 1985);
8) Gambia, farmers (BILLEVICZ, 1981);
9) Gambia, peasants (PRENTICE et al., 1981);
10) Benin, peasants (SCHULTINK et al., in preparation);
11) Kenya, non-sugar farmers (COGILL, 1987);
12) Kenya, sugar farmers (COGILL, 1987);
13) Bangladesh, farmers (landless) (CHEN, CHOWDHURY and HOFFMAN, 1979);
14) Bangladesh, farmers (landowners) (CHEN, CHOWDHURY and HOFFMAN, 1979);
15) New Guinea, horticulturalists (CRITTENDEN and BAINES, 1986);
16) New Guinea, horticulturalists (SPENCER and HEYWOOD, 1983);
17) Burma, farmers (TIN-MAY-THAN and BA-AYE, 1985);
18) Ethiopia, peasants (FERRO-LUZZI, in preparation);
19) Zaire, farmers (PAGEZY, 1982);
20) Zaire, farmers (PAGEZY, 1984).
A crude summary of maximum yearly body weight fluctuations is presented in Figure 6. Seasonal studies in LDC are not easy to conduct, and, in spite of the apparent simplicity, obtaining reliable serial measures of weight in free-living adults throughout one year, is a task the difficulty of which should not be underestimated. This partially explains why, despite the growing concern of various disciplines and the recognition of the central position of weight changes as the best indicator of impact on human communities, there are so few comprehensive studies of seasonality.
Nomadic pastoralists of Senegal (BENEFICE, CHEVASSUS-AGNES and BARRAL, 1984), and Niger (LOUTAN and LAMOTTE, 1984) were found to lose 2.1 to 2.4 kg in women, and 2.7 to 3.1 kg in men. This represents about 4 to 5% of their body weight. African farmers show a similar variation, with losses ranging from 0.7 kg (equal to 1.4% of body weight; COGILL, 1987) to 3.8 kg (or 6% of body weight; ANCEY, 1974). Little serial body weight data is available for countries outside Africa: Burmese male farmers (TIN-MAY-THAN and BA-AYE, 1985) and Bangladeshi women (CHEN, CHOWDHURY and HOFFMAN, 1979) have been found to lose 5 and 3% respectively of their body weight.
A rather modest gender difference in body weight losses emerges from the analysis, women tending to have slightly smaller weight losses (1.4 to 4.6%) than men, who lose 2.3 to 6.4% of their body weight. New Guinea is the only place where women have been found to lose more weight than men; they also had a bimodal yearly weight change, while men followed a unimodal pattern (SPENCER and HEYWOOD, 1983) (Figure 7).
An interesting aspect of most of the reported values of seasonal body weight changes is that they are characterized by a very large interindividual variability. This suggests either large differences in exposure to seasonal stress within the same community, or alternatively the existence of non-uniform coping strategies. Two studies provide some insight into this issue, showing that the amount of weight loss might be related to initial body fat stores. In a rural group of women in Ntomba, Zaire, the more corpulent women were found by PAGEZY (1984) to tend to lose more weight than the leaner women. The correlation was not significant, but the observed tendency was supported by a similar correlation between skinfold thicknesses and amount of fat lost from the same site. Similar results were obtained in Benin, where the women with the highest BMI were the only ones to lose significant amounts of body weight (SCHULTINK et al., in preparation). Very recent data, describing the weight changes in a group of about 400 rural non-pregnant women in a seasonally bimodal area in Kenya, provide conflicting results (COGILL, 1987). Mean weight changes were very modest and the lean women (lowest quartile of BMI) had significantly greater changes (about 2 kg) than heavier women (highest quartile of BMI). Ethiopian women lost weight in quantities totally independent of their degree of fatness (FERRO-LUZZI, SCACCINI and DENTE, in preparation). No such evidence exists for men. However, most authors have reported that a small percentage of their study groups experienced a much more marked weight loss than the rest of the group. Unfortunately, no information is available as to whether these are the individuals with higher body fat reserves.
In
conclusion, while there is no doubt that body weight changes
occur, especially but not necessarily in areas of the world
characterized by a high seasonality index, these changes appear
to be rather modest, rarely exceeding 5% of the maximum yearly
value of body weight. The functional significance of weight
losses of this size is far from obvious, and it is legitimate to
question their impact on work capacity, on functional competence,
and on their power to initiate adaptive responses.
4.1. Basal metabolic rate (BMR)
4.2. Physical activity and energy expenditure
A decrease in BMR,
persisting even after adjusting for body weight and fat-free mass
(FFM) losses, represents one of the earliest and best known
adaptive responses to energy deficit, and has been extensively
described under experimental and controlled laboratory
conditions. However, the seasonal impact on BMR has not been
systematically investigated, and little can be found on this
topic in the literature.
A study conducted in rural Gambia showed that 21 unsupplemented women, who lost 5 kg of body weight during the rainy period (Figure 8), lowered their BMR by a maximum of 50 kcal/d (LAWRENCE et al., in press). This drop is statistically significant (p<0.01), but obviously can have only a very doubtful energy-saving value. Furthermore, the drop appears to have slightly preceded the loss of body weight. Since there was a concomitant reduction of energy intake and thus an energy deficit, these findings suggest that BMR was as early a response to energy imbalance as the loss of body weight.
Women in Benin lost 1.5 kg between April and June, at the end of the rainy season, while their BMR did not change (SCHULTINK et al., in preparation). Ethiopian women lost about 1 kg and experienced a simultaneous decrease of energy intake. Their BMR appeared to follow energy intake changes more closely than body weight changes (FERRO-LUZZI, SCACCINI and DENTE, in preparation).
In
conclusion, the changes in BMR are neither consistent throughout
the various reports, nor biologically significant. Our
understanding of their meaning is limited, and it is hardly
possible to draw firm conclusions from these scarce and
conflicting data. Studies conducted by SUZUKY (1959) in Japan
revealed a significant seasonal fluctuation in BMR, closely
related to average temperature. There were also fluctuations in
energy intake, although it would be rather unlikely, given the
social status of the study subjects, that there were external
limitations to food availability. Therefore, as for growth
performance in children, it could well be that at least part of
the observed seasonal fluctuations of BMR in LDC may be explained
by factors other than energy imbalance.
The other potentially very
powerful mechanism for saving energy (and thus likely to become
operative under the stressful conditions of seasonal energy
imbalances) is commonly held to consist in the reduction of
physical activity. This mechanism seems so obvious that it is
surprising how little scientific evidence exists that it actually
becomes operative under real life conditions. And if it did, we
do not know if it is voluntary or subconscious, if savings are
made through reducing the tempo, decreasing the intensity, or
increasing the mechanical efficiency, or if certain types of
activities are reduced, and if so, which? Unfortunately, good
survey material on the topic is too scanty to allow any firm
conclusions, and we can only make some speculations.
General apathy, lassitude and torpidity have been described as characteristic of the spontaneous behaviour of the experimental subjects of the Minnesota Starvation Study (KEYS et al., 1950), but the special tightly controlled conditions under which the experiment was conducted do not allow extrapolation to free-living, real-life conditions. Anecdotal evidence has been reported in a few energy supplementation studies; all of it points towards an improvement in physical performance following energy supplementation (VITERI, 1971).
Figure 9. Extra energy spent (kcal/d) during peak agricultural season:
1) North Nigeria, labour sellers (LONGHURST, 1984);
2) North Nigeria, labour buyers (LONGHURST, 1984);
3) Upper Volta, farmers (BRUN, BLEIBERG and GOIHMAN, 1981; BLEIBERG, BRUN and GOIHMAN, 1980);
4) Upper Volta, farmers (ANCEY, 1974);
5) The Gambia, peasants (LAWRENCE et al., in press);
6) Machiguenga, peasants (MONTGOMERY and JOHNSON, 1977);
7) Burma, farmers (TIN-MAY-THAN and BA-AYE, 1985).
One crucial piece of information, required to understand the impact of seasonality on energy metabolism, is to establish the likelihood that adaptive mechanisms would be switched on by such modest weight losses as observed in the seasonality studies reviewed above. The fact that weight losses have usually been found to occur at the same time of the year when, due to the agricultural workload, physical activity is supposed to be most intense, implies the priority of maintaining intact the level of energy expenditure over the mobilization of body energy stores. That an increasing proportion of the day is devoted to tasks with high energy requirements at those times of the year when body weight losses have been more frequently observed, is documented by several time-allocation studies (KUMAR, 1987; TRIPP, 1982). The validity of these data is somewhat questionable because time-allocation surveys provide at best a crude estimate of energy expenditure, since they refer to duration rather than intensity of work or effort. A large margin of uncertainty is therefore attached to the translation of time-allocation data into energy expenditure.
Fortunately, there are a small number of studies where energy expenditure in different seasons has been actually measured. These results, shown in Figure 9, confirm that energy expenditure rises during the peak agricultural season, usually the rainy season, at the time body weight decreases. The peak increase in expenditure over minimum values varies between 320 and 1050 kcal/d. If we assume an average discrepancy of 600 kcal, i.e., an imbalance of ± 300 kcal/d, this would lead, over an arbitrary, but reasonable period of three months, to a cumulative imbalance of ± 27,000 kcal. The observed seasonal changes in body weight, ranging from 2 to 4 kg, appear commensurate with this imbalance, i.e., there is no need to invoke the intervention of adaptive energy-sparing mechanisms.
A closer scrutiny of the energy expenditure data reveals some tenuous evidence that a high workload may play a leading role in seasonal energy imbalance; the Gambian data, for example, suggest that the body weight starts to decrease immediately after, or even at the same time as energy expenditure rises (LAWRENCE et al., in press; Figure 8). Similar synchronous shifts of energy expenditure and body weight have also been observed in rural Upper Volta (ANCEY, 1974; Figure 10). It is possible to argue that a lower energy intake, due to depletion of food stocks, may be partly responsible for the negative energy balance. Obviously, such a coincidence may indeed exist, but there is evidence that seasonal weight losses, coinciding with higher loads of physical work, may also occur in the absence of a restriction in food availability, as in the case of New Guinean women (SPENCER and HEYWOOD, 1983). The Gambian women, too, were reported to start losing weight before food shortage had "started to bite" (LAWRENCE et al., in press).
This evidence is corroborated by data reported by KUMAR (1987) on the fluctuation of energy intakes and levels of food stocks in Nigeria. It is evident that during the main planting season, when food stores appeared to be lowest and physical labour highest, energy intake was also the highest of the year (Figure 11). Note that intakes dropped to their lowest during the postharvest period when food stocks were replenished.
1) Upper Volta, farmers (REARDON and MATLAN, in press);
2) Upper Volta, farmers (REARDON and MATLAN, in press);
3) Senegal, pastoralists (BENEFICE, CHEVASSUS-AGNES and BARRAL, 1984);
4) The Gambia, breast-feeding peasant women (PRENTICE et al., 1981);
5) The Gambia, pregnant peasant women (PRENTICE et al., 1981);
6) Bangladesh, farmers (HASSAN, HUDA and AHMAD, 1985);
7) Bangladesh, farmers (ABDULLAH and Wheeler, 1985);
8) Benin, farmers (SCHULTINK et al., in preparation);
9) Burma, farmers (TIN-MAY-THAN and BA-AYE, 1985);
10) Ethiopia, peasants (FERRO-LUZZI, SCACCINI and DENTE, in preparation).
On the
other hand, data on the fluctuation of energy intakes have been
reported from areas of the world characterized by high climatic
seasonality, showing that intakes during the wet season, when
food stocks are lowest, may drop by 400 to 500 kcal/d as compared
to intakes during the dry season (Figure 12). However, the
figure also shows that there are areas where a steady and stable
food supply allows energy consumption to increase when, during
the rainy season, labour demand may be slightly in creased.
1. Seasonality in energy metabolism exists. However, it is not universal. It occurs especially, but not exclusively, in rural environments characterized, mostly but not necessarily, by marked climatic and agricultural fluctuations.
2. The overlap and interaction of several seasonal events confound the picture and may precipitate a marginal situation into imbalance in areas with low seasonality index.
3. Children appear to be affected as much as adults. Food shortage and contagious diseases are likely to be the main reasons, which distinguishes their situation from that of adults. The picture is further complicated by the fact that seasonal fluctuations of food availability and infectious diseases are tightly interlocked.
4. Seasonality of energy balance appears to cause a modest fluctuation of 2 to 5% of the body weight of adults, equivalent to the mobilization and subsequent redeposit of about 15,000 to 20,000 kcal of the body's energy stores. The evidence that adaptive mechanisms become operative in response to a seasonally negative energy balance of this size is scarce and contradictory.
5. Peaking of workloads appears to be closely associated with seasonal losses of body weight. This suggests that energy expenditure may be the critical factor in the causation of energy imbalance in adults. The preharvest food shortage may, but does not necessarily, contribute to energy imbalance.
1. Establish by unified criteria the environmental conditions conducive to regular seasonal recurrence of bottlenecks in human energy metabolism.
2. Define the extension, distribution and severity of the phenomenon in the world.
3. Investigate the causes of interindividual variance of seasonal body weight change. Does it reflect a difference in the coping strategies? If so, what is the nature of these strategies?
4. The permanent impact on energy metabolism, of the timing and sequence of seasonally recurring growth arrest and acceleration.
5. Establish the chronological sequence of body responses to seasonal energy imbalance.
6. Investigate the BMR contribution to seasonal energy-sparing: identify the nature and severity of seasonal challenges needed to elicit BMR responses.
7. Improve methodologies to explore adaptive changes in time allocation and energy expenditure.
