Motor, cognitive, and behavioural disorders in children born very preterm

Transcription

1 Motor, cognitive, and behavioural disorders in children born very preterm LA Foulder-Hughes MEd Dip COT PhD, Research Fellow; RWI Cooke* MD FRCP FRCPCH, Professor of Neonatal Medicine, Department of Child Health, University of Liverpool, Royal Liverpool Children s Hospital (Alder Hey), Liverpool L12 2AP, UK. *Correspondence to second author at Neonatal Unit, Liverpool Women s Hospital, Liverpool L8 7SS, UK. mc19@liv.ac.uk Children born preterm have been shown to exhibit poor motor function and behaviour that is associated with school failure in the presence of average intelligence. A geographically determined cohort of two-hundred and eighty preterm children (151 males, 129 females) born before 32 weeks gestation and attending mainstream schools were examined at 7 to 8 years of age together with 210 (112 males, 98 females) age- and sex-matched control participants were tested for motor, cognitive, and behavioural problems. Tests applied were the Movement Assessment Battery for Children (MABC), Clinical Observations of Motor and Postural Skills (COMPS), Developmental Test of Visual-Motor Integration (VMI), Wechsler Intelligence Scale for Children, and Connors Teacher Rating Scale for attention-deficit hyperactivity disorder (ADHD). Control children scored significantly better than the preterm group on all motor, cognitive, and behavioural measures. The lowest birthweight and most preterm individuals tended to score the lowest. Motor impairment was diagnosed in 86 (30.7%) of the preterm group and 14 (6.7%) of the control children using the MABC; 97 (42.7%) and 18 (10.2%) using the COMPS; and 68 (24.3%) and 17 (8.1%) respectively using the VMI. Each test of motor function identified different children with disability, although 23 preterm children were identified as having motor disability by all three tests. Preterm children were more likely to have signs of inattention and impulsivity and have a diagnosis of ADHD. Minor motor disabilities persist in survivors of preterm birth despite improvements in care and are not confined to the smallest or most preterm infants. They may exist independently of cognitive and behavioural deficits, although they often co-exist. The condition is heterogeneous and may require more than one test to identify all children with potential learning problems. Improved survival of very preterm infants in the past two decades has led to an increased interest in their long-term neurodevelopmental outcome. Reported series of such infants show a major disability rate of 10 to 15%, largely due to cerebral palsy (CP), and visual and hearing deficits (Cooke 1994). Most of the children without disability attend mainstream schools but it has become apparent that one-third to one-half of these children will experience learning and behavioural difficulties, often associated with minor motor disability (Powls et al. 1995). In the absence of major sensory, physical, or intellectual disability such children have been referred to as clumsy (Losse et al. 1991) or as having developmental dyspraxia (Ayres et al. 1985), perceptual motor dysfunction (Lazlo and Sainsbury 1993, Jongmans et al. 1998), deficits in attention, motor control, and perception (DAMP; Landgren et al. 1998), and developmental coordination disorder (DCD; Wilson et al. 1992). DCD appears to be the preferred term at present and is defined as impairment of motor performance sufficient to produce functional performance not explicable by the child s age, intellect, or other diagnosable neurological or psychiatric disorders (Polatajko et al. 1995, p 310). Children born preterm may have co-existing conditions, such as CP or low intelligence, which would preclude the diagnosis of DCD. This disorder is also well recognized to be comorbid with attention-deficit hyperactivity disorder (ADHD) in up to 50% of such children (Parry 1996). There is no single diagnostic test for DCD and a wide variety of tests has been used. The Movement Assessment Battery for Children (MABC; Henderson and Sugden 1992) or its predecessor, the Test of Motor Impairment (TOMI; Stott et al. 1984), has proved popular. A score at less than the 5th centile on the MABC has been used as indicating motor impairment consistent with a diagnosis of DCD. The test has been standardized in a large cohort of children in the UK and Canada. The Developmental Test of Visual-Motor Integration (VMI; Beery 1997) is a test for use with children from a variety of cultural and social backgrounds and educational experiences and is claimed to be a reliable tool to assess the integration of visual and motor abilities. It has been used previously to assess preterm children. The Clinical Observations of Motor Postural Skills test (COMPS; Wilson et al. 1994) is intended as a screening tool for DCD in children without major physical or intellectual disabilities. The weighted results take the child s age into account, which few standard neurological examinations do. It is widely used by occupational therapists. Of the many published case series of low-birthweight and preterm infants most have consisted of hospital-based cohorts, usually selected by birthweight, and have not always included term control participants (Table I). Many of these published series predate relatively recent treatment advances. Extensive antenatal steroid prophylaxis, exogenous pulmonary surfactants, and improved postnatal nutrition introduced in the past decade might be expected to improve long-term outcome by reducing the severity of initial respiratory failure and improving rates of postnatal brain growth. This study examined the rate of motor impairment and associated behavioural and cognitive disabilities in a geographically determined cohort of 7- to 8-year-old children born before 32 completed weeks gestation from 1991 to All were attending mainstream schools; comparison with a term classmate control group was made. Developmental Medicine & Child Neurology 2003, 45:

2 Method PARTICIPANTS All preterm infants born before 32 completed weeks gestation from 1991 to 1992 in eight hospitals within the Liverpool, UK postal districts were ascertained. Those who died before discharge from hospital or whose mothers were not resident within a Liverpool postal district at the time of birth were excluded. Initial contact was made with the family doctor via the child s hospital paediatrician to ascertain current health status and school placement. Parents of those children who were alive and attending mainstream schools were approached for consent to take part in the study. The individual children s schools were then contacted to arrange assessment visits and to request that the class teacher choose a child of the same sex and first language in the class whose birthday was closest to that of the preterm child. The parents of that control child were approached with information about the study and consent for their child to participate was sought. The preterm group comprised 280 children (151 males, 129 females; mean age at testing 89.8 months; mean gestational age 29.8 weeks). The term control group comprised 210 children (112 males, 98 females; mean age at testing 89.9 months). Most children were tested at their schools, although some were tested at the Institute of Child Health, Royal Liverpool Children s Hospital at their parents or teacher s request. The study protocol was approved by the local research ethics committee. Parents gave written consent to their children being tested. TEST INSTRUMENTS Fine and gross motor skills were assessed using age band 2 of the MABC (Henderson and Sugden 1992). The test comprises eight items, two in each of four subsections: manual dexterity, ball skills, static balance, and dynamic balance. The scoring system for each item is from ranges from 0, no impairment to 5, severe impairment. The scores for each item are added and converted to centiles. A score between the 5th and 15th centile Table I: Some recent studies of motor impairments in children born preterm Study Participants Birthweight or Hospital or Age at Major Control Tests used n gestation geographically testing impairment based cohort y:m excluded Roberts et al <1252g Hospital 6:0 Yes Yes TOMI Marlow et al TOMI Powls et al MABC Levene et al <1500g Hospital 5:0 No Yes TOMI Mutch et al <1750g Geographical 4:6 No No TOMI Hall et al :0 MABC Jongmans et al <35wk Hospital 6/7:0 Yes Yes MABC, VMI Goyen et al <1500g Hospital 5:0 Yes No PDFM, VMI Luoma et al <32wk Hospital 5:0 Yes Yes VMI, motor battery TOMI, Test of Motor Impairment (Stott et al. 1984); MABC, Movement Assessment Battery for Children (Henderson and Sugden 1992); PDFM, Peabody Developmental Fine Motor Scale; VMI, The Developmental Test of Visual-Motor Integration (Beery 1997). Table II: Movement Assessment Battery for Children scores, median (interquartile range) Preterm Control p Bean bag catch 1.00(0 4) 0(0 2) <0.001 Bounce and catch a 0.75 (0 3) 0 (0 1.25) <0.001 Flower trail 1 (0 3) 0 (0 1) <0.001 Heel-to-toe walking 0 (0 1) 0 (0 0) <0.001 Jumping in squares 0 (0 1) 0 (0 0) <0.001 Placing pegs 1 (0 2) 0 (0 1) <0.001 Stork balance 1 (0 3) 0 (0 3) <0.001 Threading lace 0 (0 3) 0 (0 1) <0.001 Total raw score 8.50 ( ) 3.50 ( ) <0.001 Total raw score (IQ>70) 8 ( ) 3.50 (1 7.00) <0.001 Total raw score (IQ>70 and excluding ADHD) 8.50 ( ) 4 (1 7.50) <0.001 a Average for both hands. Movement Assessment Battery for Children (Henderson and Sugden 1992) is an impairment score with a range of 0 to 40, the higher the score the greater the motor impairment. Scores can be standardized in age bands to produce a percentile score, but raw scores are used here. 98 Developmental Medicine & Child Neurology 2003, 45:

3 for age is considered borderline impairment, and 5th centile definitely impaired. In this study, a score at or below the 5th centile was taken to indicate motor difficulties consistent with DCD. A behavioural checklist also forms part of the MABC and records the presence of 12 behavioural difficulties which may arise during testing. Motor difficulties were also assessed using the COMPS (Wilson et al. 1994). This provides a more objective test of soft neurological signs and is thought to reflect cerebellar function, motor coordination, and postural stability. It can also detect the presence of motor problems with a postural element. The test comprises six items, each with a score of 0 to 12. The six items are slow (ramp) movements, rapid forearm rotations (diadochokinesis), finger nose touching, prone extension posture, asymmetrical tonic neck reflex, and supine flexion posture. Item scores are converted to total weighted scores. A total weighted score <0 suggests difficulties with motor postural skills. It is also useful to use the scores of individual items to assess whether specific areas of difficulty exist. Integration of visual and motor abilities was assessed using the VMI (Beery 1997). It consists of 27 geometric forms which increase in complexity and are in a developmental sequence. There is a 1-point scoring system for each shape. Scores are then standardized for age. Standard scores have a mean of 100 (SD 15). General intelligence was measured using the short form of the Wechsler Intelligence Scale for Children UK, 3rd edition (WISC-III UK; Wechsler 1992). Total, Verbal, and Performance scores were calculated. Teachers were asked to complete the Connors Teacher Rating Scale (1987) to evaluate attentiondeficit hyperactivity disorder (ADHD). A score 70th centile for age is considered equivalent to the DSM-IV s (American Psychiatric Association 1994) definition of ADHD. A perinatal clinical dataset was completed from the clinical records where available by a medical secretary. The MABC, COMPS, and VMI were administered by an occupational therapist, and IQ tests were administered by three graduate research assistants trained in their use. The research assistants were not blind to whether the child being examined was a preterm child or a control participant. Statistical analysis of data was carried out using SPSS software (version 10) using parametric or non-parametric tests, depending on the distribution of data. Statistical significance was set at p<0.05. Results Of a potential cohort of 382 preterm children identified, 280 children were tested (33 children had moved out of the area or could not be traced, 18 had died, 29 children were attending special schools, 16 parents refused permission for their Table III: Behavioural checklist for MABC for preterm and control infants Behaviour Preterm Control p Overactive 186/33/61 186/8/16 <0.001 Passive 200/31/49 166/15/ Timid 259/13/8 194/12/ Tense 197/46/37 174/20/ Impulsive 186/37/57 177/21/12 <0.001 Distractible 189/42/49 182/16/12 <0.001 Disorganized 207/37/36 187/13/10 <0.001 Overestimates ability 228/28/24 197/9/4 <0.001 Underestimates ability 259/12/9 201/3/ Lacks persistence 217/19/44 184/11/ Upset by failure 257/13/10 201/5/ No pleasure from success 225/14/41 172/11/ Numbers refer to categories: Rarely / Occasionally / Often. MABC, Movement Assessment Battery for Children (Henderson and Sugden 1992). Table IV: Summary table for multiple regression analysis with total score on MABC as dependent variable (preterm only) Behaviour ß Standard error of ß T p Constant <0.001 Overactive Tense Impulsive Distractible Disorganized Overestimates Lacks persistence MABC, Movement Assessment Battery for Children (Henderson and Sugden 1992). Table V: COMPS subsection scores, median (interquartile range) Subsection Preterm Control p Asymetrical tonic neck reflex 12 (8 12) 12 (12 12) <0.001 Finger nose touching 8 (5 11) 12 (9 12) <0.001 Prone extension 8 (4 10) 10 (6 12) <0.001 Forearm rotations 12 (10 12) 12 (12 12) <0.001 Slow motion 12 (10 12) 12 (12 12) <0.001 Supine flexion 8 (4 12) 10 (8 12) <0.001 Total weighted score 0.95 ( ) 1.89 ( ) <0.001 Total weighted score (IQ>70) 1.02 ( ) 1.70 ( ) <0.001 Total weighted score (IQ>70 and excluding ADHD) 1.02 ( ) 1.56 ( ) <0.001 Clinical Observation of Motor Postural Skills (COMPS; Wilson et al. 1994) consists of six items, each with a possible score of 0 to 12. Score of 12 indicates optimal performance. Item scores are converted to total weighted scores. Total weighted score of <0 suggests overall difficulties with motor postural skills. Motor Impairment in Preterm Children L A Foulder-Hughes and RWI Cooke 99

4 children to be tested, and six missed appointments). In addition to the 280 preterm children, 210 term age- and sexmatched control children were recruited. In 70 cases a control child was not found, because the preterm child was tested (1) out of school at the request of teacher or parent, (2) during school holidays, or (3) because the parents of the chosen control child did not agree to testing. Time constraints at school meant that a few children did not complete all the tests. Mean gestational age for the preterm group (n=151) was 29.8 weeks (range 23 to 32) and mean birthweight was 1467g (SD 424; range 512 to 2860). Of this group 21.4% were of 28 weeks gestation, and 3.6% 24 weeks gestation; 50.4% of the group were <1500g birthweight and 14.6% 1000g. There were 215 singleton births, 56 twins, and nine triplets. Mean age at testing was 89.8 months (SD 5.3; range 82 to 101) for the preterm children and 89.9 months (SD 6.1; range 72 to 107) for the control children. Of the 266 preterm children from whom perinatal data were Table VI: Standardized scores, mean (SD), for VMI for preterm and term control children Preterm Control p VMI standardized score 90.5 (9.2) 96.9 (7.8) <0.001 VMI score (IQ>70) 91.9 (8.1) 96.9 (7.7) <0.001 VMI score (IQ>70 and excluding ADHD) 92.4 (8.2) 96.8 (8.1) <0.001 Development Test of Visual-Motor Integration (VMI; Beery 1997) consists of 27 geometric forms of increasing complexity and which are in developmental sequence. There is a one point score for satisfactorily copying each shape. Scores are then standardized for age. Standard scores have a mean of 100 (SD 15). Table VII: Subsection scores for WISC-III for preterm and term control children Subsection Preterm Control p Verbal Comprehension 91 (13.7) 99 (13.1) <0.001 Perceptual Organization 87 (15.6) 99 (15.4) <0.001 Freedom from Distraction 97 (14.2) 106 (13.4) <0.001 Processing Speed 94 (16.4) 104 (15.7) <0.001 Information 8.6 (2.8) 10.0 (2.8) <0.001 Similarities 9.1 (3.1) 10.7 (2.9) <0.001 Arithmetic 10.3 (3.2) 11.4 (2.8) <0.001 Vocabulary 7.7 (2.9) 8.9 (2.9) <0.001 Comprehension 8.3 (3.5) 9.8 (3.1) <0.001 Digit Span 8.6 (2.7) 10.0 (3.0) <0.001 Picture Completion 9.0 (3.1) 10.4 (2.7) <0.001 Coding 9.1 (3.7) 10.4 (3.6) <0.001 Picture Arrangement 8.3 (3.8) 10.8 (3.9) <0.001 Block Design 7.3 (3.3) 8.9 (3.3) <0.001 Object Assembly 7.5 (3.1) 9.3 (2.9) <0.001 Symbol Search 8.7 (3.2) 10.9 (3.3) <0.001 Mazes 7.4 (3.3) 9.1 (3.2) <0.001 WISC-III, Wechsler Intelligence Scale for Children, 3rd edn. (Wechsler 1992). Table VIII: Correlations between motor tests and Total, Performance, and Verbal IQ Test COMPS VMI Total IQ Performance IQ Verbal IQ Connors Total weighted Teacher Rating Scale MABC a a a a a a COMPS Total weighted a a a a a VMI a a a a Total IQ a a a Performance IQ a a Verbal IQ a a p<0.01 Spearman s rho. COMPS, Clinical Observation of Motor Postural Skills; VMI, Development Test of Visual- Motor Integration; MABC, Movement Assessment Battery for Children. 100 Developmental Medicine & Child Neurology 2003, 45:

5 obtained, 140 (53%) were ventilated for a median of 5 days (range 1 to 70) and 153 (57.5%) received added oxygen for a median of 6 days (range 1 to 150). Eleven of the preterm children (4%) had a clinical seizure. Cerebral ultrasound showed subependymal haemorrhage in 16 children (6%), intraventricular haemorrhage in three children (1%), and parenchymal haemorrhage in two children (0.7%) on the right side; and 15 (5.6%), 16 (6%), and 1 (0.4%) respectively on the left side. Late scans showed 15 children (5.6%) to have persistent ventricular dilatation, three children (1%) had cystic periventricular leukomalacia, and one child had shunted hydrocephalus. The control children as a group scored significantly better than the preterm children on all measures of motor function and intelligence. Median total score on the MABC was 3.50 (range 1 to 6.63) for control children and 8.5 (3.13 to 15.38) for the preterm group (p<0.001, Mann Whitney U test; Table II). Using the 5th centile to define a DCD group, 86 (30.7%) of the preterm group fulfilled this criterion. Of the control group, 14 children (6.7%) (expected 5%) also fulfilled this criterion suggesting that they were indeed normal. No significant difference in the frequency of DCD was seen between the sexes in preterm nor control children. A significant negative correlation between gestation and birthweight and score on the MABC was found in the preterm group (Spearman s rho, 0.19, p=0.001 for gestation; 0.18, p=0.003 for birthweight). However, a gestational age of 28 weeks (p=0.24) or birthweight <1500g (p=0.056) was not associated with DCD. Using the MABC checklist, seven behaviours were seen significantly more often in the preterm than term group (Table III). When these variables for the preterm group only were entered in a multiple regression analysis with total score on the MABC as the dependent variable, overactive, tense, distractible, and lacks persistence were independently associated with poorer performance on the MABC (Table IV). The COMPS test was carried out on 227 preterm and 176 control children. Ninety-seven (42.7%) preterm and 18 (10.2%) control children displayed subtle motor coordination difficulties with total weighted scores <0 (p=0.001, Mann Whitney U test). Sub-section scores are shown in Table V. Statistically significant correlations between the COMPS total weighted score and gestational age and birthweight were found (Spearman s rho 0.180, p=0.007; 0.159, p=0.017). Control children performed significantly better on the Table IX: Some comparisons between centile scores obtained on MABC and VMI in preterm infants in this and other recent studies (percent) MABC <15th centile <10th centile <5th centile This study Jongmans Powls Hall Mutch VMI <85 This study 23.7 Goyen Jongmans MABC, Movement Assessment Battery for Children (Henderson and Sugden 1992); VMI, The Developmental Test of Visual-Motor Integration (Beery 1997). Figure 1: Venn diagram showing overlap of disability identified by three motor tests in preterm children. 128/227 children were considered to have a disability by one or more test(s); 99/227 children were not considered to have a disability by any test. MABC, Movement Assessment Battery for Children; VMI, Developmental Test of Visual-Motor Integration; COMPS, Clinical Observation of Motor Postural Skills. MABC (<5th centile; n=74) VMI (score <85; n=63) COMPS (score <0; n=94) Motor Impairment in Preterm Children L A Foulder-Hughes and RWI Cooke 101

6 VMI with a mean score of 96.9 (SD 7.8) compared with the preterm group mean score of 90.5 (SD 9.2), a difference of 6.4; 95% CI 4.84 to 7.95 (Table VI). Again, significant correlations were observed between the VMI standard score and gestational age at birth and birthweight (Spearman s rho 0.181, p=0.002; 0.151, p=0.011). Mean (SD) Total, Verbal, and Performance IQ scores in control children were (13.7), (12.7), 99.6 (15.8) respectively, compared with 89.4 (14.2), 92.9 (13.9), 87.8 (15.6) respectively for the preterm group (p<0.001, independent t-tests). Subscores are shown in Table VII. Total IQ was correlated significantly with gestational age and birth weight (Spearman s rho 0.264, p<0.001; p<0.001). Further significant correlations were seen between each of the three tests of motor function and Total, Performance and Verbal IQ (Table VIII). The scores for the MABC, COMPS, and VMI were recalculated when children with a total IQ of <70 were excluded, but they did not differ significantly from those of the groups as a whole (Tables II, V, and VI). The overlap of disability as identified by the three tests is shown in Figure 1. Only children who completed all tests are shown (n=227) of whom 74 scored below the 5th centile on the MABC, 94 had a total weighted score of <0 on the COMPS, and 63 had a standardized VMI score of <85. Only 23 children performed poorly on all three tests. These children were of lower gestation, birthweight, and IQ, and had a greater discrepancy between Verbal and Performance scores when compared with the children identified by only one or two of the tests. Median (interquartile range) centile scores on the Connors Teacher Rating Scale were 53 (range 45 to 62) for inattention, 49 (range 45 to 58) impulsivity, and 52 (range 46 to 60) total ADHD for the preterm group and 45 (range 44 to 51), 46 (range 44 to 51), and 46 (range 44 to 51) respectively for the control group (p<0.001). Eighteen (8.9%) preterm and three (2.1%) term children s scores met the criteria for ADHD (p=0.01). Excluding children with a total IQ <70 left 12 preterm and three term children meeting the criteria (p=0.04). The total ADHD score on the Connors Scale was significantly correlated with all measures of motor and cognitive ability (see Table VIII). However, when children with both an IQ <70 and those meeting the criteria for ADHD were omitted, the differences in motor impairments between the preterm and control groups remained essentially unchanged (see Tables II, VI, and VII). Discussion Consistent with earlier work with very-low-birthweight children, participants born very preterm in this study showed a high incidence of minor motor impairment on all three tests (MABC, VMI, COMP) when compared with age-matched control children born at term (Elliman et al. 1991, Levene et al. 1992, Mutch 1992, Marlow et al. 1993, Hall et al. 1995, Powls et al. 1995, Goyen et al. 1998, Jongmans et al. 1998, Luoma et al. 1998; Table IX). This impairment has been shown to have an impact on overall school performance, despite general intelligence within the normal range. Motor impairment at six years predicted school failure (performing in the bottom 10% of the class in mental arithmetic and/or spelling and/or reading) at eight years (Marlow et al. 1993). As expected, there was a tendency for children born with the lowest gestational ages and birthweights to perform significantly less well on all the tests, although the correlations were weak. Indeed, those performing the least well on the motor tests were distributed through the gestation and weight range, suggesting that the relation was not a simple one. The proportion of children affected in this area was very similar to that in previous reports of outcome in cohorts born one and two decades ago, and does not appear to have been altered by improvements in perinatal care, unlike in major neurodisability (Cooke 1999). This study has a number of methodological limitations. The population was defined geographically in that only children of mothers living at an address within the Liverpool postcodes area were recruited from the eight hospitals in the area. Children of mothers who delivered outside the area, however, were not recruited. Tracing the whole cohort was made difficult as it was a retrospective exercise and there were many changes of name and address. Nevertheless, 91% of the cohort was traced, and only six traced children who could have been tested were not tested. Some schools had very little space for testing which limited their cooperation. Under- ascertainment has been shown to bias previous follow-up studies, but usually because children with severe impairments have not been included. Such children were excluded in this study as they were not in mainstream schools. Not all children completed every test. This was usually due to time constraints, although some teachers declined to complete the questionnaires. Motor impairments seen in children born preterm have been described as DCD, although a strict definition of the term excludes those with an IQ below 70. In this study, exclusion of children with cognitive impairment did not change the average test scores for preterm and control groups, indicating that the excess of motor impairment in the preterm group could not be explained by low IQ. There appears to be a link between DCD and ADHD (Landgren et al. 1996) and prevalence rates of DCD in children with ADHD have been found to range between 8 and 52% (Parry 1996). In this study the rate was 50% using <5th centile MABC as the criterion for DCD. When children diagnosed as having ADHD were excluded, as well as those with low IQ, again there was no significant change in the average motor impairment scores for the preterm and control groups, suggesting that the motor impairment seen in the preterm group was also independent of ADHD in the majority of those affected. However, the preterm children showed an excess of adverse behavioural characteristics while performing the MABC tasks, and some of these independently predicted the MABC score. Similarly, reported signs of inattention and impulsivity from the Connors ratings were also seen significantly more often in the preterm group and were associated with poorer performances on all motor and cognitive tests. There has been debate as to the best test to use to identify the presence of DCD in children and agreement between tests of less than 80% has been shown (Crawford et al. 2001). This was the case in this study. The three tests of motor function overlapped in their identification of motor disability, but imperfectly. Of the preterm children completing all three motor tests, 130 (57.3%) were identified as having problems by at least one, 78 (34.4%) by two, and 23 (10.1%) by all three tests. This is not surprising as they are testing different aspects of motor function: the VMI, pencil and paper and visual perceptional skills; the COMPS, posture, tone, and coordination; and the MABC, a combination of all these. Motor 102 Developmental Medicine & Child Neurology 2003, 45:

7 disability in these children may also be a heterogeneous condition. Also, the ability to perform well in the VMI may not only require visual perceptional and fine motor skill but also the ability to achieve and hold overall body posture as measured by the COMPS. The MABC identified the least number of children not identified by one of the other two tests, and a combination of the COMPS and VMI identified the most children with motor difficulty. Both the MABC and the VMI were significantly correlated with IQ, and in particular Performance IQ, presumably as they contain some similar tasks. It is clear that despite considerable improvements in perinatal care and survival of preterm infants, motor difficulties occur at school age in these children with a similar frequency as that of a decade ago. Also, the problems are not confined to the smallest or most immature infants. Identification of poor function in these children may require the use of several tests rather than a single test in view of the heterogeneous nature of the condition. DOI: /S Accepted for publication 4th October References Ayres AJ Mailloux Z Wendler C. (1985) Developmental dyspraxia: is it a unitary function? Occup Ther J Res 7: Beery KE. (1997) The Beery-Buktenica Developmental Test of Visual-Motor Integration. Administration and Scoring Manual 4th edn. Parsipanny, NJ: Modern Curriculum Press. Connors CK. (1987) Connors Teacher Rating Scale Revised (L). North Tonawanda, NY: Multi-Health Systems Inc. Cooke RW. (1994) Annual audit of three year outcome in very low birth weight infants. Arch Dis Child 63: Cooke RW. (1999) Trends in incidence of cranial ultrasound lesions and cerebral palsy in very low birthweight infants, Arch Dis Child 80: F Crawford SG, Wilson BN, Dewey D. (2001) Identifying developmental coordination disorder: consistency between tests. Phys Occup Ther Pediatr 20: Elliman AM, Bryan EM, Elliman AD, Walker J, Harvey DR. (1991) Co-ordination in low birth weight seven year olds. Acta Paediatr 80: Goyen TA, Lui K, Woods R. (1998)Visual motor, visual-perceptual and fine motor outcomes in very-low-birthweight children at 5 years. Dev Med Child Neurol 40: Hall A, McLeod A, Counsell C, Thomson L, Mutch L. (1995) School attainment, cognitive ability and motor function in a total Scottish very-low-birthweight population at eight years: a controlled study. Dev Med Child Neurol 37: Henderson SE, Sugden DA. (1992) Movement MABC. London: Psychological Corporation Harcourt Brace and Co. Jongmans MJ, Mercuri E, Dubowitz LMS, Henderson SE. (1998) Perceptual-motor difficulties and their concomitants in six-year old children born prematurely. Hum Mov Sci 17: Landgren M, Pettersson R, Kjellman B, Gillberg C. (1996) ADHD, DAMP and other neurodevelopmental/psychiatric disorders in 6-year-old children: epidemiology and comorbidity. Dev Med Child Neurol 38: Landgren M, Kjellman B, Gillberg C. (1998) Attention deficit disorder with developmental co-ordination disorder. Arch Dis Child 79: Laszlo JI, Sainsbury KM. (1993) Perceptual-motor development and prevention of clumsiness. Psychol Res 55: Levene M, Dowling S, Graham M, Fogelman K, Glaton M, Philips M. (1992) Impaired motor function (clumsiness) in five year old children: correlation with neonatal ultrasound scans. Arch Dis Child 67: Losse A, Henderson SE, Elliman D, Hall D, Knight E, Jongmans M. (1991) Clumsiness in children do they grow out of it? A 10-year follow-up study. Dev Med Child Neurol 33: Luoma L, Herrgard E, Martikainen A. (1998) Neuropsychological analysis of the visuomotor problems in children born preterm at 32 weeks of gestation: a 5-year prospective follow up. Dev Med Child Neurol 40: Marlow N, Roberts BL, Cooke RW. (1993) Outcome at 8 years for children with birthweights of 1250g or less. Arch Dis Child 68: Mutch L for the Scottish Study Group. (1992) The Scottish Low Birthweight Study: 1. Survival, growth, neuromotor and sensory impairment. Arch Dis Child 67: Parry T. (1996) Multiple stimuli disorganisation syndrome: treatment and management of children with attentional disorders. Aust Educ Dev Psychol 13: Polatajko HJ, Macnab JJ, Anstett B, Malloy-Miller T, Murphy K, Noh S. (1995) A clinical trial of the process-oriented treatment approach for children with developmental co-ordination disorder. Dev Med Child Neurol 37: Powls A, Botting N, Cooke RW, Marlow N. (1995) Motor impairment in children years old with a birthweight of less than 1250g. Arch Dis Child 72: F62 6. Roberts BL, Marlow N, Cooke, RWI. (1989) Motor problems among children of very-low-birthweight. Br J Occup Ther 52: Stott DH, Moyes FA, Henderson SE. (1984) Test of Motor Impairment (Henderson Revision). Guelph, Ontario: Brook International. Wechsler D. (1992) Wechsler Intelligence Scale for Children. 3rd edn. London: Psychological Corporation, Harcourt Brace and Co. Wilson BN, Pollock N, Kaplan BJ, Law M, Faris P. (1992) Reliability and construct validity of the Clinical Observations of Motor and Postural Skills. Am J Occup Therapy 46: Wilson BN, Pollock N, Kaplan BJ, Law M. (1994) Clinical Observations of Motor and Postural Skills. San Antonio, TX: Therapy Skills Builders. Motor Impairment in Preterm Children L A Foulder-Hughes and RWI Cooke 103