Chapter 5 Functional consequences of low BMI in adults
Several attempts have been made to determine the effects of CED on work efficiency and work output. Physical work capacity, which often is measured and expressed as the body's maximal capacity to consume oxygen is largely determined by muscle mass. In physically strenuous work, positive correlations have been found between work capacity measured in this way and work performance. Taller individuals with larger body and muscle masses have been consistently shown to have a higher work capacity and work performance than short individuals. However, when work capacity is expressed per kilo body weight or per kilo active tissue mass, the evidence is less clear.
Undernutrition in young adults in Guatemala reduced their maximal work capacity. However, the effect of undernutrition on work output at submaximal work levels or under normal work situations has not been assessed. When groups of young Colombian adults who were described as having CED of varying degrees of severity were studied, the results showed that only the "severely malnourished" group had a marked reduction in VO2 max1, expressed in total litres of oxygen per minute (Spurr, 1987). They also had a lower aerobic capacity when this was expressed as VO2 max corrected for body weight. This severely undernourished group was the only group with a mean BMI < 18.5, the other two groups had BMIs > 18.5. A BMI below 18.5 appeared to impair work capacity. There seemed to be a gradient in aerobic capacity with a change in BMI, with a decrease in aerobic capacity in those with lower BMIs.
[ 1. VO2 max or max VO2 is the point at which the oxygen consumption plateaus and shows no further increase with an additional work load such as increasing the elevation of a treadmill or increasing the resistance to pedalling a bicycle ergometer. It is generally assumed that this represents an individual's capacity for aerobic metabolism and provides a quantitative statement of an individual's physiological work capacity. ]
Detailed studies of migrant agricultural labourers have shown an association of poor nutritional status, identified by anthropometry, with lowered work capacity (Desai, 1989). Among Brazilian adolescent males who were migrants and considered to be undernourished, physical work capacity was assessed in physiological tests using a bicycle ergometer. They had a mean BMI of less than 17.0 compared with the control group of well-to-do Brazilian adolescent males who had a mean BMI of about 20.0. The undernourished adolescents had similar rates of oxygen consumption and gross exercise efficiency when compared to the well-nourished adolescents in south Brazil (Desai et al., 1984). However, it was apparent that the undernourished adolescents achieved this work rate using a higher percentage of their maximum work capacity. Thus, they had a significantly higher heart rate for the same level of oxygen consumption. The lactic acid levels in the blood were also higher during exercise suggesting that the available muscle mass was under greater stress as the young men tried to accomplish the same task.
General findings from work physiology would suggest that these features indicate a probable limit to the endurance of the undernourished males if they had to work over longer time periods or with higher work loads. Similar data are emerging from India, where good correlations were found between body weight, BMI and work capacity in undernourished individuals in Hyderabad (Satyanarayana et al., 1989). In the Indian study, the undernourished adults with BMIs <18.5 demonstrated similar work capacities expressed in terms of their body weight as did the urban Indians with a mean BMI >20.0. This means that their total work capacity was reduced as a result of their low BMI status. However, they had similar increments in oxygen uptake with increasing work loads which is an expression of no change in mechanical efficiency of the body.
Assessing the link between poor nutritional status and diminished work productivity is complicated by several other variables, such as motivation, the wages paid and the health status of the individual. Low BMI and poor nutritional status may also limit productivity indirectly through their effects on absenteeism and reduced opportunities for recruitment by potential employers who prefer larger individuals. Agricultural labour productivity and wages have been linked with household energy intakes in Sierra Leone (Strauss, 1986) and in Sri Lanka. Labour productivity in South India, measured in terms of wage and farm output, was related to a weight-height index (Deolalikar, 1988), while in the Philippines higher agricultural wages appear to be achieved by taller individuals (Haddad & Bouis, 1991). People with low BMIs from India have been shown to have lower productivity in urban industrial situations as well (Satyanarayana et al., 1977). Low BMIs also seemed to predict lower productivity in industrial tasks. The productivity, on a per working day basis, of individuals with a low BMI was compared with that of people of the same stature but within a normal range of BMI. In this case, men with the lowest mean BMI (16.5) were the least productive group (Satyanarayana et al., 1989).
Studies from Guatemala compared the agricultural work output of two groups of men who were matched for age and had similar ethnic-cultural backgrounds (Town et al., 1989). When the two groups of men were asked to do heavy, standardized agricultural tasks for 3 to 6 days, including wood-cutting, clearing a stretch of land with a machete and working with a hoe and sickle, the sub-group with significantly less lean body mass and muscle mass were able to do the same amount of agricultural work as their well-nourished counterparts. However, the group with less muscle mass took considerably more time to complete the same amount of work (397 minutes compared with 235 minutes per day) and were able only to perform the task at a less intense level (4.6 versus 5.1 kcal per minute). In this study, the mean BMIs of both groups were over 20, (i.e. means of 20.1 and 23.2). This may indicate that these features of stamina and rate of working relate to body size even within the range of BMIs now classified as normal. The capacity for work may be progressively lower as body weight declines. Choosing a cut-off point for CED is then somewhat arbitrary, but the evidence so far obtained does not suggest that the cut-off point needs to be set any lower.
It could be speculated that a low BMI will alter physical work capacity from a metabolic point of view. However, the physiological changes seen are only proportionate to the reduction in muscle tissue associated with the low BMI states and the state of muscle activity does not seem to have been impaired. Nevertheless, the physical and economic disadvantages of a low body weight seem to be clear and harmful on their own.
The behavioural adaptation in physical activity patterns that accompanies low BMI states is mainly related to the individual's allocation of time and energy to different productive and leisure activities and to the estimation of the biological and economic consequences of these altered behavioural patterns. When there is both a fall in energy intakes and an increased demand for energy expenditure at work, e.g. during some points in agricultural seasonal activities, individuals will adjust the time they allocate to different tasks: more time is given to work activities and less time and energy are expended for pleasure and dealing with productive tasks at home (Immink, 1987). This is an important form of behavioural adaptation and the pattern of such adaptation may be different for CED individuals than for those who are well-nourished.
The earliest, well-documented study of behavioural alteration during long-term energy deficiency was that by Keys and his colleagues (1950) in the classic Minnesota semi-starvation studies. After 24 weeks of semi-starvation, the Minnesota study participants had BMIs of about 16.5 and showed severe inactivity with a marked reduction in spontaneous activity and clear lack of motivation to undertake much activity. The dramatic reduction in physical activity was an important survival mechanism during this study because the low food intake was fixed. In Asia and Africa, however, there seem to be adults who manage to maintain reasonable levels of daily activity despite having low BMIs equivalent to the Minnesota volunteers. This suggests that the Asians and Africans have a higher energy intake since no ethnic differences in energy metabolism and efficiency have been discovered. The severe weight loss of the tall American volunteers who had never before experienced food deprivation may have been crucial to their behavioural change. In contrast, in Africa or India the adults probably had been deprived of food over long periods of time and had never achieved a normal BMI. They had maintained their levels of activity at the expense of weight gain.
An analysis of the pattern of physical activity during a voluntary reduction in food intake has shown that the behavioural response to a deficient intake is that the individual changes his or her activity pattern (Gorsky & Calloway, 1983). The loss in body weight associated with the deficient intake was associated with a distinct change in activity patterns: lower effort discretionary activities replaced any which needed greater effort. The obligatory activities were not, however, affected. Gorsky and Calloway (1983) considered that observations such as those of Viteri (1982) and Satyanarayana et al., (1977) may not have provided the same evidence on behavioural adaptation as they had found. Individuals who are motivated to perform activities of highest priority do select and adjust their behaviour accordingly.
Researchers of the Institute of Nutrition of Central America (INCAP) arrived at similar conclusions (Town et al., 1989). The rural Guatemalan men in their studies who had lower muscle masses were able to carry out the specific agricultural task allocated to them but took a much longer time to do it. Additional interesting changes in their activity behaviour were then observed: these individuals took a significantly longer time to walk home after work and they spent approximately 3 hours each day taking a nap or sitting, playing cards or doing other sedentary activities. In contrast, the better nourished age-matched males did not sleep during the day and proved to be active at home in addition to playing soccer for their recreation. The latter group, therefore, remained physically active for a significantly greater proportion of the day. Marginally undernourished individuals, therefore, tend to become more sedentary at the expense of social interactions and discretionary non-salaried activities.
In Kenya, a study of 220 male roadworkers, also found that there was a highly significant relationship between weight-for-height and the time taken to complete a task (Latham, 1989). Men with low weight-for-height took significantly longer to complete the same task than men with a higher weight-for-height. In this study, anaemia was not a confounding factor. It was concluded that energy deficient individuals will be forced over a period of time to limit their activities in order to conserve energy. Whereas some do so consciously and wilfully, others do it unconsciously.
Careful analysis (Ferro-Luzzi et al., 1992) of more recent data based on two concurrent studies (Norgan et al., 1993; Branca et al., 1993) on seasonal variations in activity patterns of men and women in rural India and Ethiopia showed that individuals with varying degrees of CED spent fewer hours per day working than did individuals in the same socio-cultural milieu whose BMI was > 18.5. Analysis of the hours of productive activity of normal and CED Ethiopian women also revealed that if the women had BMIs less than 18.5 a smaller amount of time was spent on productive activities in a day. Data collated from the National Food Consumption and Household Budget Survey carried out in Rwanda also provided information on the physical activity index of 1,100 women in rural areas (François, 1990). When the physical activity index is expressed as a multiple of the basal metabolic rate, computed for all days on a yearly basis, it shows a gradient in the levels of physical activity. Those individuals with a BMI < 17.6 have average physical activity levels which are significantly lower and they have increased levels of rest time per day (See Figure 5.1).
Figure 5.1 - Effect of body mass index on physical activities of Rwandese women, 1982
Source: François, P., FAO, 1990, unpublished data.
These carefully analysed studies support the impression that adults with a low BMI are forced to adapt their behaviour in terms of spontaneous, free-living physical activity. Low body weights seem to limit an adults' work output, productivity and income-generating ability. Restricting discretionary or obligatory physical activity becomes an important short-term survival strategy for low-weight individuals but this may reduce their earning power and jeopardize their long-term survival. They may be less able to respond to stressful conditions when they suddenly face greater demands.
All of these analyses point to the same view regarding low BMI which is different than that previously held. The previously accepted view that a low body weight in an adult is not of much consequence now seems not to be true. Low BMIs are associated with a consistent behavioural pattern demonstrable in Asia, Africa, South and Central America. In all communities, there is a clear decline in work output and the ability to sustain productivity throughout the day. The time taken in work has to be extended to the detriment of socially desirable leisure activities. These observations justify the emphasis given to the energy needs at work and at home by the FAO/WHO/UNU 1985 Report and as advocated by FAO for a number of years (Périssé & Kamoun, 1987) and suggest that far greater attention should be paid to adult BMIs. The data also reinforce the view that a cutoff point of 18.5 is not unreasonable. Finally, the BMI seems to be a reasonably sensitive index of function and its monitoring is necessary if development projects depend on the physical activity of the community.
