This is the old United Nations University website. Visit the new site at http://unu.edu
Abstract
Introduction
Methods
Results
Discussion
Acknowledgements
References
Ernesto Pollitt, Jia-Fen Huang, and Abas Jahari
The authors are affiliated with the Section of Pediatric Nutrition in the Department of Pediatrics at the University of California in Davis, California, USA.
Mention of the names of firms and commercial products does not imply endorsement by the United Nations University.
Learning about the dynamics of early childhood development in the context of poverty and malnutrition is an effective way to identify the periods of highest psychosocial vulnerability and the role of timing in the differential responsivity of young children to interventions. Motor activity under natural conditions is one area of development that requires further study because of its biological and psychological developmental importance. This paper presents an estimate of the development function of motor activity among poorly nourished children whose ages ranged from 12 to 30 months. It also focuses on the longitudinal relationship between the nature, level of motor development, and intensity of motor actions, on the one hand, and physical growth, dietary intake, and motor activity, on the other.
Two cohorts (12 and 18 months old) of rural West Javanese children were studied. Motor activity, energy intake, physical growth, and motor development were measured every two months. A motor activity score was calculated based on the estimated energy cost and intensity of each activity and their frequency and duration during each of the periods of observation. Intensity was classified according to a three-point scale (high = 1.25, moderate = 1.0, low = 0.75). The Bayley Scale of Motor Development and a scale of motor milestones were used to assess motor development (e.g., crawling, creeping, walking, and running).
The developmental function of activity was curvilinear. It showed a moderately fast acceleration from 12 to about 18 to 20 months; at that point the curve reached a plateau. The functions of growth and energy intake were linear. The average motor behaviour of these children was characterized by its low level of intensity. Although the relationship between motor activity and motor development was relatively close up to about 20 months, thereafter the motor activity and motor development scores were independent of each other. The changes in activity level were primarily determined by a significant decline of involvement in activities having a low energy cost and an increase in activities having moderate or high energy cost. The decline of motor activity observed at 18 months is reminiscent of the decline in growth velocity observed during the first six months of life among malnourished children. It is plausible that the pattern of activity observed could be explained by either contextual or cultural factors.
Early childhood development programmes aim to prevent and ameliorate the psychosocial developmental delays observed among children from economically impoverished populations. In low-income countries, large-scale (national) programmes are justified because they allegedly foster a child’s competence to meet the demands of primary education. In many of those countries, the governments face heavy economic losses associated with the failure of children to meet formal educational demands by dropping out of school early or having to repeat grades. For example, children finish primary school in 8 to 10 years instead of 6 years as it is programmed. These disappointing outcomes often follow early childhood malnutrition, poor health, and limited educational opportunities.
The recent attention that international organizations and interested governments have given to early developmental delays calls for a comprehensive understanding of the patterns of development among children who are developmentally at risk in low-income countries to identify sensitive points for intervention. Good intentions, by themselves, will not bring about salutary developmental changes. Little is known about child-care practices and the social, emotional, motivational, motor, and mental development of such children. Learning about the course of their development and its environmental determinants in the context of poverty and malnutrition will help identify the periods of highest vulnerability and the differential responsivity to interventions as a function of timing.
Physical growth among low-income populations in low-income countries illustrates the importance of drawing the trajectory of a particular growth or developmental outcome. The linear growth of such populations is close to the median of the reference standards of the World Health Organization (WHO) up to about four months of age. This period is followed by a period of growth faltering up to about 36 months [1-3]. The most timely period of intervention to prevent, in part, growth faltering is during the periods when growth deficits are steep [4].
The reduction of energy intake leads to a reduction of energy expenditure, including lowering motor activity [5, 6]. It is theoretically plausible that this adaptation [6] places a heavy toll on development, since motor activity is an important antecedent of cognitive function and social-emotional regulation [7-10]. In the context of developmental psychology, motor activity is also recognized as a key indicator of infant temperament. Interindividual differences in activity have been associated with differences in the intensity with which children respond to environmental stimuli [11,12].
The theoretical recognition of the biological and psychological importance of motor activity has not been accompanied by strong field studies on the motor activity of nutritionally at-risk children under natural conditions [13]. There is a large body of knowledge and normative data on the timing and characteristics of physical growth of well-nourished and poorly nourished children, but this is not so for motor activity. For example, it is not known whether the deceleration in growth that starts at about four months among poorly nourished children is associated with a similar shift of motor activity. Likewise, we know little about the behavioural mechanisms that mediate the reduction of motor activity under conditions of deficient diets or increased morbidity. Moreover, only recently has it been recognized that besides energy deficiency, deficiencies of zinc and iron also reduce motor activity [14, 15].
Field studies of the motor activity of young children from different populations will help us understand the nature and causes of intra-individual, interindividual, and intergroup variability of activity levels. The enormously complex task of standardization of methods to collect normative data may be prohibitively expensive and is not likely to generate interpretable information if the ecocultural context is not accounted for.
This paper addresses the following questions: What is the developmental function of motor activity among poorly nourished children whose age ranges from 12 to 30 months? How does this developmental function compare with the weight gain and linear growth curves of the same children? Is there an association between changes in the shape of the curve for the developmental function of motor activity and changes in energy intake? What is the longitudinal relationship between the nature, level of motor development, and intensity of motor behaviour, on the one hand, and motor activity, on the other hand?
Research design
Location
Communities and day-care centres
Subjects
Variables and measurements
This study is part of a larger project on early supplementary feeding, motor development, behaviour, and cognition in West Java, which included two cohorts of children (12 and 18 months old) enrolled in 24 day-care centres in six tea plantations. There were three different dietary interventions: energy plus micronutrients (E), micronutrients (M), and skim milk (SM). The dietary interventions were randomly assigned to the day-care centres. For the present study, the effects of treatment were statistically controlled, and within each cohort the subjects from the three groups were pooled together.
The six tea plantations were located in the subdistrict of Pangalengan, 50 km south of Bandung, the capital of West Java. Pangalengan is at an elevation of 1,500 to 1,800 m above sea level and has a temperature that ranges from 12° to 17°C. The six plantations encompassed 32 communities of workers and their families, with little in- or out-migration. In 1992 the average wage for picking tea leaves was approximately Rp 72,000 per month (Rp 2000 = US$1.00). The population of each community ranged from 272 to 2,399 (average, 1,047), with an average household size of 4.2 members.
Twenty-four communities in the six tea plantations met two inclusion criteria: presence of a day-care centre and at least two infants or toddlers classified as wasted and stunted. All houses had electricity and potable water. Public bathrooms with toilets were available in separate quarters. Primary schools and medical services were available in all communities.
The day-care centres were usually located in the residential area of each community and were open from 6 a.m. until 2 p.m. Most had one multipurpose room for playing and napping, a kitchen, a restroom, and a small vegetable garden. The children enrolled in the day-care centres ranged in age from 1 to 72 months, and the number of children per day-care centre ranged from 5 to 35.
All children 12 (±15 days) and 18 (±15 days) months of age living in the communities and enrolled in the day-care centres were potential participants. Among these children, we selected and invited to enroll those children who met two anthropometric criteria: wasting and stunting. Wasting was defined as a weight-for-length between one and two standard deviations below WHO standards. Stunting was defined as a length-for-age below one standard deviation of the respective median of the same reference. There were 53 children in the 12-month cohort and 61 children in the 18-month cohort.
Motor activity
Every two months, three fieldworkers carried out four hours of observation of motor activity when the children were awake: two hours in the morning at the day-care centres and two hours in the afternoon at home. The second session usually took place after the mother returned home from working in the field.
The list of activities used is a modification of the lists used by others [13,16] in studies on malnutrition and motor activity (table 1). Continuous observations were conducted with the aid of computer notebooks (Acer Model No. K386S) and custom-tailored software. Observation and recording of the data were done simultaneously.
TABLE 1. Motor development scale, observed activities, and activity cost
|
Motor development |
Motor activity |
||||
|
|
|
|
1 |
Lying |
1.20 |
|
1 |
Sit 1 |
Sits with support |
2 |
Sitting still |
1.20 |
|
2 |
Sit 2 |
Sits without support and the body is not upright |
|
|
|
|
3 |
Sit 3 |
Sits without support and the body is upright |
|
|
|
|
4 |
Crawl 1 |
Lying on abdomen, can raise himself or herself up by the
hands |
3 |
Crawling |
2.50 |
|
5 |
Crawl 2 |
Lying flat on the abdomen, can raise the body bearing the
weight on the hands and the tips of the toes alone |
|
|
|
|
6 |
Creep 1 |
Crawls and starts by going backwards |
4 |
Creeping |
2.00 |
|
7 |
Creep 2 |
Creeps on all fours (on hands and knees) |
|
|
|
|
|
|
|
5 |
Kneeling |
1.40 |
|
|
|
|
6 |
Carried |
1.20 |
|
|
|
|
7 |
Sitting with movement |
1.40 |
|
|
|
|
8 |
Squatting |
2.00 |
|
|
|
|
9 |
All fours |
1.20 |
|
8 |
WaIk 0 |
Walks with assistance of other; feet are not steady
yet |
10 |
Walks with assistance |
2.50 |
|
9 |
Walk 1 |
Walks with assistance of other; feet flat on the
floor |
|
|
|
|
10 |
Walk 2 |
Walks by self with support |
|
|
|
|
11 |
Stand 0 |
Learning to stand |
11 |
Stands with assistance |
1.30 |
|
12 |
Stand 1 |
Stands with support |
|
|
|
|
13 |
Stand 2 |
Stands without support |
12 |
Stands without support |
2.30 |
|
14 |
Walk 3 |
Walks a few steps without support |
13 |
Walks without assistance |
3.00 |
|
15 |
Walk 4 |
Walks alone with small steps |
|
|
|
|
16 |
Walk 5 |
Walks with large steps |
|
|
|
|
17 |
Run |
Running |
14 |
Running |
5.00 |
|
|
|
|
15 |
Climbing |
3.00 |
|
|
|
|
16 |
Pushing |
2.00 |
|
|
|
|
17 |
Jumping |
3.50 |
TMA = E x I x F x D
where
E = estimated energy cost of an activity,
I = estimated level of intensity (vigour) of an
activity,
F = frequency of an activity during four-hour period of
observation,
D = duration of each episode of each activity.
The energy cost of each activity was estimated based on the existing literature [Durnin J, personal communication, 1996]. Intensity was classified in a three-point scale (low = 0.75, middle = 1.0, high = 1.25). Because the activities were exclusive of each other, the computer calculated the duration in minutes of each activity coded (table 1).
Of particular interest were the processes involved in the variability of activity within or between children. We investigated three particular changes: changes in intensity; changes in the frequency and duration of activities having low, moderate, or high energy cost; and acquisition of new motor skills.
Dietary intake
The total energy intake of the children at the day-care centres and at home was measured for a 24-hour period at the beginning of the study and every 2 months thereafter for 12 months. These measurements included the energy from the supplements given to the children from the two cohorts and as part of the larger study. However, as noted, the effects of the nutritional intervention were controlled for in the assessment of the growth and developmental outcomes reported here.
Anthropometry
According to the same schedule as the dietary-intake assessments, a series of anthropometric measurements was obtained, including weight, recumbent length, head circumference, skinfolds, and arm circumference. This paper is concerned with only the first two of the anthropometric measurements.
Motor milestones
The main objective was to assess the sequential changes in gross motor skills (e.g., crawling and standing) leading to bipedal locomotion. To assess these changes, we developed an ordinal scale based on the classical work of Myrtle McGraw [17] on the neuromuscular maturation of the human infant and on over 100 hours of observation of infants and toddlers who attended the day-care centres. McGraw had proposed a timely sequence in the development of different phases of motor skills involved in movement, ranging from incipient propulsion in the superior region of the body to erect locomotion. A motor milestone score was obtained based on the child’s most advanced skill. These data were collected weekly.
Bayley Scale of Motor Development
This scale includes items that test both fine and gross motor movements. Palmar prehension, thumb opposition, and partial finger prehension of a pellet are examples of items that test fine motor movements. Gross motor movements are assessed by items such as sits with support, pulls to sitting position, stepping movements, and walks with help.
Developmental function of motor activity (12 to 30 months)
Energy intake and anthropometry
Processes involved in the regulation of motor actions
Figure 1 presents the mean scores of motor activity obtained every 2 months for 12 months from the two cohorts. It also includes an exponential regression estimate for the 12-month cohort (R2 = .98). After reaching a motor activity score of about 450, the curve turns flat in both the 12- and the 18-month cohort (see also table 2). This pronounced change in motor activity was unexpected, because on average children in this population are first able to walk without assistance, which is a comparatively high-energy expenditure activity, at about 14 months [18].
Figure 2 presents the curve estimated from the means of the total energy intake at each time point for the two cohorts. It also includes the results from the respective regression equations. For both cohorts, the best fit to the data were linear regressions; for the 12-month cohort, R2 = .34 (p < .0001), whereas for the older cohort, R2 = .32 (p < .0001). The differences in the shape of the curves that best fit the intake (linear) and activity (curvilinear) data show that the precipitous change in activity level at about 18 to 20 months was not preceded or accompanied by a change in the intake of energy.
Figures 3 and 4 present the weight and the recumbent length curves and the results from the regression equations. Here again these functions do not resemble the curve of activity level.
Intensity (vigour)
The Count growth model [19] was also used to estimate the developmental function of activity with and without the intensity factor. We have shown previously that the Count model provides a close fit to growth data [2,20]. The three model parameters represent three components: initial departure point (i.e., not time dependent) a, linear increase in activity b, and deceleration c.
TABLE 2. Exponential least squares summary statistics (12-month cohort)
|
Source |
DF |
Sum of squares |
Mean square |
r2 |
|
Regression |
3 |
66,321,101 |
22,107,033.6 |
|
|
Residual |
368 |
1,275,591.3 |
3,466.28 |
|
|
Uncorrected total |
371 |
67,596,692 |
|
0.98 |
FIG. 2. Mean total energy intake according to cohort and age
The mathematical expression is written as follows:
y= a+ bx+ clog(x+ 1)
Figures 5 and 6 present the curves of the motor activity scores, with and without intensity, for the 12-and 18-month cohorts, respectively. Figure 5 shows that the Count model provides a moderate fit to the data from the 12-month cohort, with (R2 = 0.35; p < .0001) and without (R2 = 036; p < .0001) intensity. Both curves show a modest incremental change in motor activity up to about 18 to 20 months of life, followed by significant deceleration. The equations for the 18-month cohort (fig. 6) were also statistically significant, but the R2 in both equations was very modest (0.10). Both lines show a very slow increase of activity throughout the 12 months of observation.
FIG. 3. Mean weight according to cohort and age
FIG. 4. Mean recumbent length according to cohort and age
Changes in motor development
In the 12-month cohort, the non-parametric (Spearman) coefficients of correlation among the three motor variables (activity score, motor milestones, and Bayley Motor score) up to about 18 months are moderate to large in size. The coefficients range from 0.35 (motor activity and Bayley Motor score) to 0.82 (motor activity and motor milestone scores). However, after 18 months there is a large drop in the size of the two coefficients, so that by 24 months these correlations are not different from zero. Such a statistical pattern in the association of motor development and motor activity should not be surprising, because by about 20 months the variability of motor development was significantly reduced, as most children were able to walk with or without assistance.
Changes in motor behavioural profile
Children can change their motor behavioural profiles by changing the frequency and duration of activities with low or high energy cost. Figure 7 presents the longitudinal curves based on the mean values of the three sets of activities having different energy costs at each of the assessment periods of the 12- and 18-month cohorts. These curves show that during the first 6 months of observation, there was a change in the behavioural profile of the children of the 12-month cohort. Simultaneously, their involvement in low-cost activities decreased and their involvement in moderate- and high-cost activities rose. This is not obvious for either the second 6 months of the 12-month cohort or for the entire 12-month period of observation of the children of the 18-month cohort. In this older group there was a modest change in the pattern of behaviour involving the three categories of activity.
FIG. 5. Fitted regression line for total motor activity score according to age (12-month cohort)
FIG. 6. Fitted regression line for total motor activity score according to age (18-month cohort)
FIG. 7. Mean motor activity scores (MAS) according to time, cohort, and energy cost of activity
There was a downward shift in motor activity at about 18 months. This developmental function is reminiscent of the deceleration of weight and linear growth observed at about 4 to 6 months among many nutritionally at-risk infants in low-income countries. However, because the study started when the youngest children were 12 months of age, no conclusions can be drawn from this sample regarding the shape of the activity function at that earlier age. We can conclude, however, that the steep decrement in motor activity observed at 18 months was not associated with a similar decrement in the velocity of physical growth. Likewise, the change in activity was not associated with any major change in the caloric intake of the children.
Other research sources have shown that motor development is not spared in the presence of malnutrition during the first 18 months of life. Several studies have shown that early supplementary feeding of poorly nourished children improves their performance in motor development scales [21]. It would be adaptive if malnutrition were to have a larger effect on physical growth before it affected motor development, because, as we have already noted, skills such as creeping, crawling, and walking are milestones of substantive value for cognitive and social-emotional development. In particular, this type of delay of motor development can delay visual perceptual development.
The intra- and inter-individual variability of motor activity depends on the intensity of the actions and the nature and use of the motor skills that are part of the behavioural profile of children. In our sample, the children were neither vigorous nor energetic, and their behaviour was characterized primarily by involvement in activities having a moderate or low energy cost. This was true even for children whose ages ranged from 24 to 30 months.
It is plausible that the pattern of motor activity that we observed among the infants and toddlers in Pangalengan is explained by contextual and cultural factors [22]. For example, the caretakers in the day-care centres where the children spent over six hours a day were generally concerned with pooling children into a group and placing them in particular areas of a multipurpose room. This informal policy must constrain somewhat the mobility of the children who are walking. Moreover, the children were free to move about the yard only during a relatively short recess period. In addition, in the communities the caretakers often carry infants and toddlers on their hips. Such a practice could certainly be an additional limiting factor.
This project was supported in part by a grant from the Nestle Foundation, Lausanne, Switzerland.
1. Allen LH. Nutritional influences on linear growth: a general review. In: Waterlow JC, Schürch B, Scrimshaw NS, eds. Causes and mechanisms of linear growth retardation. Eur J Clin Nutr 1994;48:S75-89.
2. Kim I, Pollitt E. Differences in the pattern of weight growth of nutritionally-at-risk and well nourished infants. Am I Clin Nutr 1987:46:31-5.
3. Martorell R, Khan LK, Schroeder DG. Reversibility of stunting: epidemiological findings in children from developing countries. In: Waterlow JC, Schürch B, Scrimshaw NS, eds. Causes and mechanisms of linear growth retardation. Eur J Clin Nutr 1994;48:S45-57.
4. Schroeder DG, Martorell R, Rivera JA, Ruel MT, Habicht JP. Age differences in the impact of nutritional supplementation on growth. J Nutr 1995;125:1051S-9S.
5. Waterlow JC. Energy-sparing mechanisms: reductions in body mass, BMR and activity: their relative importance and priority in undernourished infants and children. In: Schürch B, Scrimshaw NS, eds. Activity, energy expenditure and energy, requirements of infants and children. Lausanne, Switzerland: International Dietary Energy Consultancy Group, 1990:239-51
6. Waterlow JC. Protein energy malnutrition. London: Edward Arnold, 1992.
7. Benson J, Uzgiris IC. Effect of self-initiated locomotion on infant search activity. Dev Psychol 1985:6:923-31.
8. Wachs TD. Temperament, activity and behavioral development of infants and children. In: Scürch B, Scrimshaw NS, eds. Activity, energy expenditure and energy requirements of infants and children. Lausanne, Switzerland: International Dietary Energy Consultancy Group, 1990:297-320.
9. Thelen E, Smith LB. A dynamic systems approach to the development of cognition and action. Cambridge, Mass, USA: MIT Press, 1994.
10. Bertenthal BI, Clifton RK. Perception and action. In: Damon W, ed. Handbook of child psychology. 5th ed. New York: Wiley, 1998.
11. Escalona S. The roots of individuality. Normal pattern of development in infancy. Chicago, 111, USA: Aldine, 1968.
12. Gandour MJ. Activity level as a dimension of temperament in toddlers: its relevance for the organismic specificity hypothesis. Child Dev 1989;60:1089-92.
13. Meeks-Gardner JM, Grantham-McGregor SM, Chang SM, Himes JH, Powell CA. Activity and behavioral development in stunted and non-stunted children and response to nutritional supplementation. Child Dev 1995; 66:1785-97.
14. Glover J, Jacobs A. Activity pattern of iron-deficient rats. BMJ 1972,2:627-8.
15. Hunt JR, Zito CA, Erjavee J, Johnson LK. Severe or marginal iron deficiency affects spontaneous physical activity in rats. Am J Clin Nutr 1994;59:413-8.
16. Torun B, Chew F, Mendoza RD. Energy costs of activities of preschool children. Nutr Res 1983,3:401-6.
17. McGraw MB. The neuromuscular maturation of the human infant. New York: Hafner, 1945.
18. Pollitt E, Husaini MA, Harahap H, Halati S, Nugraheni A, Sherlock AO. Stunting and delayed motor development in rural West Java. Am J Hum Biol 1994;6:627-35.
19. Count EW. Growth patterns of the human physique: an approach to kinetic anthropometry. Hum Biol 1943; 15:1-32.
20. Wohlleb JC, Pollitt E, Mueller WH, Bigelow R. The Bacon Chow study: maternal supplementation and infant growth. Early Hum Dev 1983;9:79-91.
21. Pollitt E, Oh SY. Early supplementary feeding, child development and health policy. Food Nutr Bull 1994; 4:885-93.
22. Super CM. The cultural regulation of infant and child activities. In: Schürch B, Scrimshaw NS, eds. Activity, energy expenditure and energy requirements of infants and children. Lausanne, Switzerland: International Dietary Energy Consultancy Group 1990;321-33.
