SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusions
  7. What this paper adds
  8. Acknowledgements
  9. References

Aim  The aim of this study was to determine the effects of early childhood stunting (height for age 2SD or more below reference values) and interventions on fine motor abilities at 11 to 12 years, and the relationship between fine motor abilities and school achievement and intelligence.

Method  A cohort of stunted children who had participated in a randomized trial of psychosocial stimulation and/or nutritional supplementation in early childhood was compared with a group of non-stunted children. Fine motor abilities were assessed in 116 stunted (67 males, 49 females) and 80 non-stunted children (43 males, 37 females) at a mean age of 11 years 8 months (SD 4.3mo) and 11 years 9 months (SD 3.8mo) respectively. Testers were blind to the children’s group assignment.

Results  Two fine motor factors were derived: rapid sequential continuous movements (RSCM) and dexterity. No effect of the early intervention was found. RSCM scores were lower in the stunted group than in the non-stunted group (p=0.01), but differences in dexterity were not significant (p=0.18) after adjusting for social background. Among stunted children, the RSCM score was significantly associated with IQ (p=0.04) and school achievement (all p<0.05).

Interpretation  Stunting in early childhood is associated with poor scores on tests of rapid sequential continuous hand movements in later childhood. Children with poorer scores are at greater risk for low IQs and low levels of school achievement.

List of Abbreviations
PPVT

Peabody Picture Vocabulary Test

RSCM

Rapid sequential continuous movements

Moderate stunting (height for age 2SD or more below references values) is an indicator of chronic undernutrition and is estimated to affect 156 million children under 5 years of age in developing countries.1 Several studies have shown that children who are stunted in early childhood have poorer cognition, school achievement, and psychosocial function in later childhood.2,3

Animal research suggests that the cerebellum is particularly vulnerable to postnatal undernutrition4 owing to its relatively late development in neuro-ontogeny. The cerebellum is linked to motor coordination and fine adjustments to muscle tone.5 There is limited evidence that children’s fine motor ability is affected by undernutrition.6,7 In a study of Indian school-aged males, low weight for age was associated with poorer performance on motor tasks; out of the 16 males with the worst scores, 15 had abnormal electroencephalogram patterns, mostly in the frontal lobes.8 In another study, 10-year-old Indian children who were underweight and stunted in early childhood had deficits in timed coordinated tasks and balance,9 but not in dexterity.10 However, undernutrition is usually associated with poverty, and socio-economic status was not controlled for in these studies.

In 1986, we conducted a 2-year intervention trial of nutritional supplementation and psychosocial stimulation with stunted children aged 9 to 24 months. A group of non-stunted children were recruited for comparison. At enrolment, the stunted children had significantly lower developmental quotients and lower scores on all measured subscales (locomotor, hand and eye coordination, hearing and speech, and performance) of the Griffiths Scales of Mental Development11 than the non-stunted children, and the deficit increased over the 2 years. Stimulation significantly benefited the children’s developmental quotients and scores on each subscale, whereas supplementation benefited their growth, developmental quotients, and scores on the performance and locomotor subscales. In the group that received both interventions, the effects were additive and in this group scores were not significantly different from those of non-stunted children.12

At the first follow-up, at age 7 to 8 years, several tests were completed and two factors were derived from the scores: perceptual motor function and cognitive function. Stimulation had significant benefits on perceptual motor function, and supplementation had no benefit on either factor. The non-stunted group had significantly better scores on both factors than the stunted children who did not receive intervention.13

As previously reported,14 participants were reassessed at age 11 to 12 years; the groups receiving stimulation had significantly higher IQs on the Wechsler Intelligence Scales for Children – Revised and higher scores on the Raven’s Progressive Matrices and Peabody Picture Vocabulary Test than the control stunted group. There was no significant effect of supplementation on any test, and neither intervention benefited school achievement or behaviour. Compared with the non-stunted children, the stunted groups had significantly lower scores in IQ and school achievement, and more conduct problems were reported by their mothers.14,15

The aims of this study were (1) to examine children’s fine motor functions at age 11 to 12 years and determine whether early stimulation and supplementation has any effect on these, (2) to compare fine motor functions in stunted and non-stunted children, and (3) to examine the relationship between fine motor abilities and school achievement and intelligence.

Method

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusions
  7. What this paper adds
  8. Acknowledgements
  9. References

Initial study

A house-to-house survey was conducted in poor neighbourhoods of Kingston, Jamaica, between 1986 and 1987, and 127 stunted children (height for age 2SD or more below the National Center for Health Statistics references16) aged 9 to 24 months were enrolled. They were then systematically assigned to one of four groups: control, supplementation, stimulation, or both treatments. The initial order of group assignment was ascertained randomly using random tables. Thirty-two non-stunted children (height for age >−1SD) were also enrolled and were matched for age (≤16mo or >16mo) and neighbourhood with the control stunted group. The sample size was calculated to have 80% power at a significance level of 5% to detect a benefit of 0.5SD from stimulation or supplementation.

The intervention lasted 2 years.12 Milk-based formula (1kg) was delivered weekly to the homes for the supplemented children, with 0.9kg each of cornmeal and skimmed milk powder for other family members. Six months and 15 months after beginning the intervention, dietary intakes were assessed with two 24-hour recalls, and the supplemented children were found to be consuming a mean of 440kJ/day and 234kJ/day respectively, more than the non-supplemented groups. They also exhibited a small increase in linear growth.17

Stimulation comprised weekly home visits by a paraprofessional and involved both the child and mother. Based on a detailed curriculum, each visit took the form of a structured play session with homemade toys, songs, and games and lasted approximately 1 hour.

At ages 7 to 8 years, the children were reassessed, and an additional 52 non-stunted children were included to increase the statistical power of detected differences between the stunted and non-stunted children. These children had been identified in the original survey and came from the same neighbourhood, thus fulfilled the same selection criteria as the original non-stunted children.13

Follow-up at age 11 to 12 years

Eight years after the end of intervention, at age 11 to 12 years, 116 stunted and 80 non-stunted children were assessed (92% of original sample). The children lost to follow-up were not significantly different on enrolment from the children who were assessed in terms of developmental quotient, anthropometry, and social background.14Figure 1 shows the sample from enrolment through to age 11 years and reasons for loss. The children’s school achievement and cognitive and fine motor abilities were measured at the University of the West Indies, Jamaica.

image

Figure 1.  Study sample from enrolment through to age 11 years.

Download figure to PowerPoint

Fine motor measurements at 11 to 12 years

Fine motor abilities were assessed using the Grooved Pegboard18 and two subtests of the Bruninks–Oseretsky Test of Motor Proficiency19– drawing vertical lines and making dots in rows of circles. The time(s) taken to complete the Grooved Pegboard task for the right and left hand and making dots in four rows of circles with the dominant hand were measured. For drawing vertical lines, the average number of straight lines drawn between two horizontal lines in two 30-second periods was recorded. In addition, the hand pronation–supination and the finger successive opposition tests were used from the Neurological Examination of Soft Signs.20 Timed tests were selected because they are more reliable than ratings.21 Modifications were made to allow better visibility of the participants’ hands and to ensure good tester reliability. For the finger successive opposition test, instead of tapping each finger against the thumb, each finger, beginning with the thumb, was successively tapped on a table for six repetitions. Because of this modification, we refer to this test as successive finger tapping. For the hand pronation–supination test, alternating the hand 20 times (i.e. palm then back of hand) was carried out on a table instead of in the participant’s lap. The time taken to execute each test with the right and left hand was recorded. Faster times or lower scores indicated better performance in all tests except for drawing vertical lines.

Before the study, test–retest reliabilities were assessed 1 week apart. All intraclass correlations were greater than 0.78 (n=18), except for hand pronation–supination for the right hand, which was 0.65. All fine motor measurements were administered by a single tester who was blind to the children’s group assignment. Interobserver agreements between the trainer and tester were assessed during the study, and these were greater than 0.96 (n=7).

Other measurements at 11 to 12 years

A questionnaire pertaining to maternal education, marital status and employment, and the number of people, rooms, and possessions in the house was administered to the mothers, and the quality of sanitation and water supply was rated on a 6-point scale. The level of stimulation in the home was recorded, including number of reading materials, toys, and games available in the home, availability of adults to help with homework and to play with, visits to a library, and trips.14 The mother’s receptive vocabulary was assessed using the Peabody Picture Vocabulary Test (PPVT).22

The study was approved by the University of the West Indies Ethics Committee, and written consent was obtained from all mothers or primary carers of the children.

Statistical analysis

The fine motor scores were examined for deviations from normalcy using tests of skewedness and kurtosis, and by observing their histograms. All test scores were log transformed except for the drawing vertical lines test. In order to reduce the number of variables and to identify the underlying constructs, factor analysis of the eight fine motor test scores was conducted using principal component analysis with varimax rotation. This produced two factors (see Results).

Factor analysis was also used to generate housing and home stimulation factors. The housing factor was calculated from the sanitation and water ratings, number of possessions, and crowding (people per room), and was used as a socio-economic indicator. Factor analysis of 11 home stimulation questions resulted in three derived factors – homework (has table and chair to do homework, adults help with homework, visits public library, and goes on trips); toys and games (number of toys and games, adult plays indoor and outdoor games with child); and reading materials (number of books, child reads newspapers). Higher scores on the factors represent higher levels of stimulation. Further details of these factors have been published elsewhere.14

Analyses of variance (ANOVA) were used to compare the fine motor factor scores of the stunted groups with stimulation (yes/no) and supplementation (yes/no) as between-group factors. Differences between the stunted groups and the non-stunted group were determined by ANOVA, using a contrast to compare the mean stunted groups against the non-stunted group. Analysis of covariance (ANCOVA) was used to adjust for age and background variables.

Within the stunted group, multiple regression analyses were used to examine the relationship between fine motor ability and school achievement and intelligence, adjusted for covariates previously shown to be significantly related to achievement and intelligence (age at test, sex, caregiver’s education and PPVT score, housing factor, homework and reading materials factor).15

Results

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusions
  7. What this paper adds
  8. Acknowledgements
  9. References

Background characteristics

There were no significant differences among the stunted groups in anthropometric measures, housing, and maternal characteristics. However, children who had received stimulation had higher factor scores for reading materials (p<0.01) and lower scores for toys and games (p<0.05) than non-stimulated children. Children who received supplementation had lower homework scores than non-supplemented children (p<0.01; Table I).

Table I.   Current anthropometry and social characteristics of all groups
 Stunted groupsaNon-stunted group (n=80)p valueb
Control (n=31)Supplemented (n=30)Stimulated (n=27)Both interventions (n=28)
  1. Values are mean (SD) except where otherwise stated. aANOVA of stunted groups with stimulation and supplementation as between group factors. bANOVA comparing the mean of the four stunted groups with the non-stunted group. cCompleted grades 9–11, log-linear analysis. PPVT, Peabody Picture Vocabulary Test.

Height, cm141.6 (5.9)142.1 (6.1)140.7 (6.2)140.8 (6.1)153.4 (6.6)0.001
Body mass index, kg/m216.56 (2.00)16.53 (2.55)16.76 (1.90)15.75 (1.43)17.74 (2.51)0.001
Head circumference, cm52.5 (1.5)52.7 (1.4)52.3 (1.4)52.5 (1.2)54.1 (1.3)0.001
Housing factor score−0.20 (0.87)−0.38 (1.20)−0.19 (0.90)0.20 (0.96)0.21 (0.97)0.018
Stimulation factor
 Homework0.11 (0.95)−0.57 (0.80)0.00 (1.16)−0.26 (0.87)0.26 (0.98)0.002
 Toys and games0.11 (1.00)0.25 (0.85)−0.35 (1.08)−0.15 (0.92)0.04 (1.04)0.602
 Reading materials−0.41 (0.89)−0.35 (0.89)−0.14 (1.01)0.38 (1.02)0.21 (1.00)0.018
Caregiver’s PPVT score83.1 (19.5)89.7 (22.5)83.8 (21.5)89.3 (23.7)93.4 (27.9)0.054
Caregiver’s education,cn (%)4 (12.9)7 (23.3)4 (14.8)10 (35.7)24 (30.4)0.001

The stunted groups were significantly shorter and thinner and had smaller head circumference than the non-stunted group (p<0.001). They also had significantly lower housing (p<0.05), homework (p<0.05), and reading materials scores (p<0.05). Their caregivers were less likely to have attended secondary schools (log-linear analysis, p<0.001) and tended to have lower PPVT scores (p=0.054) compared with the non-stunted group (Table I).

Fine motor factors

Factor analysis of the eight fine motor tasks generated two factors with eigenvalues >1 (Table II). Hand pronation–supination and successive finger tapping tests for both hands were loaded for the first factor and were named ‘rapid sequential continuous movements’ (RSCM). The Grooved Pegboard test (both hands) and drawing vertical lines and dots in circles loaded on the second factor. These tests involved the manipulation of objects (pencil or pegs), and the factor was called ‘dexterity’. Higher factor scores indicate slower times or poorer performance.

Table II.   Factor loading of fine motor tasks (transformed variables) for all participants
 Factor loadingEigen-valueVariance (%)
Factor 1 – Rapid sequential continuous movements (RSCM) 4.56157.01
 Hand pronation–supination (left hand)0.851  
 Hand pronation–supination (right hand)0.768  
 Finger tapping (left hand)0.749  
 Finger tapping (right hand)0.745  
Factor 2 – Dexterity 1.00612.57
 Grooved pegboard (dominant hand)0.829  
 Vertical lines−0.802  
 Grooved pegboard (non-dominant hand)0.770  
 Dot in circle0.658  

Intervention effects

The means and SDs of individual fine motor tests and the two factors of the intervention groups and non-stunted group are shown in Table III. No significant intervention effects (effect sizes in SDs: RSCM – supplementation 0.08, stimulation 0.10; dexterity – supplementation 0.04, stimulation 0.05) were found for the fine motor factors, nor were there any significant interactions between the interventions.

Table III.   Means (SDs) of individual fine motor tasksa and two fine motor factors of intervention and nutritional groups
 Stunted groupsbNon-stunted group (n=80)p valuec
Control (n=31)Supplemented (n=30)Stimulated (n=27)Both interventions (n=28)
  1. aAll scores in seconds except vertical lines (number of lines). For vertical lines, stunted groups drew fewer lines per minute than the non-stunted group. bANOVA of stunted groups with stimulation and supplementation as between-group factors, all non-significant. cANOVA comparing the mean of the four stunted groups with the non-stunted group. Stunted groups performed significantly slower than the non-stunted group.

Dot in circle21.27 (4.53)21.13 (7.10)20.39 (5.64)19.52 (3.48)18.26 (3.71)0.001
Grooved Pegboard (dominant hand)68.50 (8.11)69.53 (15.65)66.54 (8.50)69.89 (12.21)64.46 (8.86)0.007
Grooved Pegboard (non-dominant hand)89.32 (22.41)85.50 (24.03)81.78 (13.21)89.34 (17.33)82.81 (14.32)0.196
Hand pronation–supination (left hand)15.05 (4.04)14.73 (3.01)14.93 (3.83)14.46 (3.79)13.78 (3.53)0.037
Hand pronation–supination (right hand)12.98 (3.38)13.03 (3.92)12.96 (2.85)12.00 (2.07)11.89 (2.85)0.039
Successive finger tapping (left hand)18.74 (5.79)20.03 (11.22)18.44 (6.41)19.02 (6.23)16.11 (5.14)0.001
Successive finger tapping (right hand)18.81 (6.35)18.77 (9.79)17.79 (6.09)18.75 (7.96)15.22 (4.73)0.001
Vertical lines43.55 (10.77)45.60 (13.75)44.00 (9.35)45.00 (10.94)48.95 (10.61)0.007
Rapid sequential continuous movements0.20 (1.06)0.20 (1.01)0.19 (0.92)0.01 (0.94)−0.22 (1.00)0.011
Dexterity0.26 (1.03)0.08 (1.35)−0.02 (0.77)0.26 (1.09)−0.22 (0.84)0.013

Differences between stunted and non-stunted

The stunted groups (control, stimulated, supplemented, stimulated and supplemented) had significantly poorer scores than the non-stunted group on all individual tests, except one (Grooved Pegboard non-dominant hand), and on the RSCM and dexterity factors (Table III).

Associations between the two fine motor factors and the child’s age, caregiver’s education and PPVT score, housing factor, and three home stimulation factors were examined with Pearson’s product–moment correlations. Child’s age (r=−0.27, p<0.001), housing (r=−0.20, p<0.01), homework (r=−0.22, p<0.01), and reading materials (r=−0.19, p<0.01) were negatively correlated with the dexterity factor. None of these variables was correlated with the RSCM factor. There was no sex difference in either fine motor factor.

An ANCOVA was conducted to determine the effects of stunting on the dexterity factor adjusted for age and social background. Age, housing, homework, and reading materials factors were entered as covariates, and group contrasts were examined as before. There was no longer a significant effect of stunting after adjusting for these variables (adjusted difference −0.18; 95% confidence interval [CI] −0.46 to 0.10; p=0.21; effect size 0.18SD). The dexterity factor was therefore not considered in further analyses. Age, homework, and reading material factors were significantly associated with dexterity. Therefore we examined interactions between these variables and stunting. These interactions were not significant. Analysis of the RSCM factor with group only showed a difference between the non-stunted and stunted groups of −0.37 (95% CI −0.66 to −0.09; p=0.01), giving an effect size of 0.37SD.

RSCM factor with school achievement and IQ

We previously reported that the stunted children had poorer school achievement and IQ scores than the non-stunted children.15 Multiple regressions of the school achievement test scores and IQ, controlling for variables previously shown to be associated with the outcome variables, showed that children’s scores on the RSCM factor were significantly related to arithmetic (p=0.039), spelling (p=0.006), word reading (p=0.009), reading comprehension (p=0.014), and IQ (p=0.036) (Table IV).

Table IV.   Regression coefficient for rapid sequential continuous movements (RSCM) factor from multiple regression of school achievement test scores and IQ in the combined stunted groups (n=116)
 Adjusted R2,%Regression coefficient, BSEp value
  1. Variables entered: RSCM factor, age at test, sex, carer’s education and Peabody Picture Vocabulary Test score, housing factor, and homework and reading materials factor. WISC, Wechsler Intelligence Scales for Children – Revised.

Arithmetic16−1.500.720.039
Spelling (log)24−0.060.020.006
Word reading (√)20−0.360.140.009
Reading comprehension27−2.701.080.014
Full-scale IQ (WISC-R)17−2.491.170.036

Discussion

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusions
  7. What this paper adds
  8. Acknowledgements
  9. References

Neither stimulation nor supplementation resulted in long-term benefits on either of the fine motor factors in stunted children. The stunted groups had poorer performance on both the RSCM and dexterity factors than the non-stunted group. However, after controlling for age and social background, no difference remained in the dexterity factor. The stunting effect on the RSCM factor was small to moderate.

All tasks measured speed of execution and had good test–retest and interobserver reliability. Speed of performance is more reliable and stable over time than ratings of quality of movements.23 The eight fine motor tasks loaded onto two distinct factors – one requiring sequential or alternating continuous movements and the other requiring fine motor manipulation or dexterity. The factors related differently to social background. The dexterity factor was related to several social background and stimulation variables. In two dexterity tests, the tasks were related to writing and may depend on practice and socioeconomic background. The RSCM factor was not related to any of the measured social background variables. The motor tasks in the RSCM factor originated from a test of soft neurological signs,20 which are often unrelated to socio-economic status.24 Similar tests are used in the assessment of minor neurological dysfunction.25

The stimulation intervention included activities related to fine motor function, such as drawing and threading of beads, and the lack of benefit is surprising. Stimulation resulted in a significant improvement in hand–eye coordination immediately after intervention12 and in Grooved Pegboard scores at the 7-year follow-up.13 However, the lack of benefit from supplementation agrees with our findings at 7 years that supplementation results in no benefit on any test.

Our finding that RSCM performance is poorer in stunted children concurs with the findings of other studies that have used tasks from the Neurological Examination of Soft Signs,7–9 and we controlled more extensively for socioeconomic differences. One study also found differences in performance on Grooved Pegboard tests,6 but we are unaware of another study with undernourished children using similar pencil and paper tasks.

The findings suggest that nutritional insults in early childhood are associated with subtle changes in brain development in areas that control certain fine motor functions, and the effects are evident several years after the period of undernutrition. A recent study26 of functional magnetic resonance imaging in children with autism and typically developing children found that sequential oppositional finger tapping was related to activity in cortical and subcortical regions associated with motor execution, including the contralateral primary sensorimotor cortex, the contralateral thalamus, ipsilateral cerebellum, and supplementary motor area. The children with autism were slower at finger tapping and showed reduced activation in the ipsilateral anterior cerebellum, and greater activation in the supplementary motor area. It is possible that the stunted children also had changes in brain function in these areas. Why one factor was associated with undernutrition and not the other is unknown. We speculate that the factors are associated with activity in different areas of the brain and that one area may be more affected by undernutrition than another.

Poor fine motor function identified stunted children at risk of low IQ and poor school achievement, especially spelling and reading. Similar associations between fine motor ability and learning have been shown in other populations.24 Fine motor disability is a consistent predictor of school achievement,27 and academically at-risk children have a slower speed in finger opposition than those not at risk.28

A limitation of this study is that relatively few tests were used, and we were not able to carry out a full neurological examination. However, it is remarkable that the few tests used showed differences between nutritional groups and predicted intellectual and academic ability.

Conclusions

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusions
  7. What this paper adds
  8. Acknowledgements
  9. References

There was no sustained benefit from either stimulation or supplementation on any measured fine motor function. Early childhood stunting was associated with poor performance in fine motor functions involving rapid sequential continuous finger and hand movements, and children with poorer function were at greater risk of lower cognitive and academic ability.

What this paper adds

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusions
  7. What this paper adds
  8. Acknowledgements
  9. References
  •  Early childhood stunting is associated with poorer fine motor abilities in later childhood.
  •  Early intervention does not improve stunted children’s fine motor ability.
  •  Previously stunted children with poorer fine motor function at 11 to 12 years are at greater risk for lower intellectual and scholastic development.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusions
  7. What this paper adds
  8. Acknowledgements
  9. References

We thank Carol Ewan-Whyte and Sharmaine Edwards for interviewing the parents and testing the children. This study was funded by Nutricia Research Foundation.

References

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusions
  7. What this paper adds
  8. Acknowledgements
  9. References
  • 1
    Standing Committee on Nutrition (SCN). Fifth report on the world nutrition situation. Nutrition for improved development outcomes. Geneva: SCN, 2004.
  • 2
    Grantham-McGregor S, Cheung YB, Cueto S, Glewwe P, Richter L, Strupp B. Developmental potential in the first 5 years for children in developing countries. Lancet 2007; 369: 6070.
  • 3
    Walker SP, Chang SM, Powell CA, Simonoff E, Grantham-McGregor SM. Early childhood stunting is associated with poor psychological functioning in late adolescence and effects are reduced by psychosocial stimulation. J Nutr 2007; 137: 24649.
  • 4
    Levitsky DA, Strupp BJ. Malnutrition and the brain: changing concepts, changing concerns. J Nutr 1995; 125: 2212S20S.
  • 5
    Gramsbergen A. Clumsiness and disturbed cerebellar development: insights from animal experiments. Neural Plast 2003; 10: 12940.
  • 6
    Galler JR, Ramsey F, Solimano G. A follow-up study of the effects of early malnutrition on subsequent development. II. Fine motor skills in adolescence. Pediatr Res 1985; 19: 5247.
  • 7
    Galler JR, Ramsey F, Solimano G, Kucharski LT, Harrison R. The influence of early malnutrition on subsequent behavioral development. IV. Soft neurologic signs. Pediatr Res 1984; 18: 82632.
  • 8
    Agarwal KN, Das D, Agarwal DK, Upadhyay SK, Mishra S. Soft neurological signs and EEG pattern in rural malnourished children. Acta Paediatr Scand 1989; 78: 8738.
  • 9
    Upadhyay SK, Agarwal DK, Shastri J, Agarwal KN. Persistence of soft neurological signs in chronic undernourished children. Nutr Res 1995; 15: 1939.
  • 10
    Agarwal KN, Agarwal DK, Upadhyay SK. Impact of chronic undernutrition on higher mental functions in Indian boys aged 10–12 years. Acta Paediatr 1995; 84: 135761.
  • 11
    Griffiths R. The abilities of babies. Thetford, UK: Association for Research in Infant and Child Development, 1967.
  • 12
    Grantham-McGregor SM, Powell CA, Walker SP, Himes JH. Nutritional supplementation, psychosocial stimulation, and mental development of stunted children: the Jamaican Study. Lancet 1991; 338: 15.
  • 13
    Grantham-McGregor SM, Walker SP, Chang SM, Powell CA. Effects of early childhood supplementation with and without stimulation on later development in stunted Jamaican children. Am J Clin Nutr 1997; 66: 24753.
  • 14
    Walker SP, Grantham-McGregor SM, Powell CA, Chang SM. Effects of growth restriction in early childhood on growth, IQ, and cognition at age 11 to 12 years and the benefits of nutritional supplementation and psychosocial stimulation. J Pediatr 2000; 137: 3641.
  • 15
    Chang SM, Walker SP, Grantham-McGregor S, Powell CA. Early childhood stunting and later behaviour and school achievement. J Child Psychol Psychiatry 2002; 43: 77583.
  • 16
    Hamill P, Drizd T, Johnson C, Reed R, Roche A. Growth curves for children, birth–18 years. Hyattsville, MD: National Center for Health Statistics, 1977.
  • 17
    Walker SP, Powell CA, Grantham-McGregor SM, Himes JH, Chang SM. Nutritional supplementation, psychosocial stimulation, and growth of stunted children: the Jamaican Study. Am J Clin Nutr 1991; 54: 6428.
  • 18
    Lafayette Instrument Company. Grooved Pegboard Test. Lafayette, IN: Lafayette Instrument Company, 1989.
  • 19
    Bruininks RH. Bruininks-Oseretsky Test of Motor Proficiency. Circle Pines, MN: American Guidance Service, 1978.
  • 20
    Denckla MB. Revised Neurological Examination for Subtle Signs (1985). Psychopharmacol Bull 1985; 21: 773800.
  • 21
    Denckla MB. Development of motor co-ordination in normal children. Dev Med Child Neurol 1974; 16: 72941.
  • 22
    Dunn LM, Dunn LM. Peabody Picture Vocabulary Test. Circle Pines, MN: American Guidance Service, 1981.
  • 23
    Vitiello B, Ricciuti AJ, Stoff DM, Behar D, Denckla MB. Reliability of subtle (soft) neurological signs in children. J Am Acad Child Adolesc Psychiatry 1989; 28: 74953.
  • 24
    Breslau N, Chilcoat HD, Johnson EO, Andreski P, Lucia VC. Neurologic soft signs and low birthweight: their association and neuropsychiatric implications. Biol Psychiatry 2000; 47: 719.
  • 25
    Hadders-Algra M. Two distinct forms of minor neurological dysfunction: perspectives emerging from a review of data of the Groningen Perinatal Project. Dev Med Child Neurol 2002; 44: 56171.
  • 26
    Mostofsky SH, Powell SK, Simmonds DJ, Goldberg MC, Caffo B, Pekar JJ. Decreased connectivity and cerebellar activity in autism during motor task performance. Brain 2009; 132(Pt 9):: 241325.
  • 27
    Batstra L, Neeleman J, Hadders-Algra M. The neurology of learning and behavioural problems in pre-adolescent children. Acta Psychiatr Scand 2003; 108: 92100.
  • 28
    Blondis TA, Snow JH, Accardo PJ. Integration of soft signs in academically normal and academically at-risk children. Pediatrics 1990; 85: 4215.