OBJECTIVE. To calculate the prevalence of developmental coordination disorder at 7 years of age by using Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition criteria in a large UK birth cohort.
METHODS. Cases of developmental coordination disorder were defined by using data from the Avon Longitudinal Study of Parents and Children, a UK birth cohort. The motor coordination of >7000 children was assessed by using tests that measured manual dexterity, ball skills, and balance. The 5th percentile of the derived Avon Longitudinal Study of Parents and Children coordination impairment score was used to define severe motor coordination difficulties. Data from national handwriting tests and an activities-of-daily-living scale quantified the impact of poor coordination on daily life. Children with known neurologic conditions or an IQ of <70 were excluded.
RESULTS. Complete data were available from 6990 children aged 7 to 8 years who attended the coordination session and completed the writing test or activities-of-daily-living scale. One hundred twenty-three children met criteria for developmental coordination disorder, resulting in a prevalence of 18 of 1000 children at a mean age of 7.5 years (SD: 2.9 months). An additional 223 children were considered as having “probable developmental coordination disorder” by using broader cut-offs for coordination testing and activities of daily living. There was an increased risk of developmental coordination disorder in families from lower socioeconomic backgrounds, in children with a birth weight of <2500 g, and those born before 37 weeks' gestation.
CONCLUSIONS. This is the first study to use strict criteria to define the prevalence of developmental coordination disorder in a representative cohort of UK children. A prevalence of 1.8% is lower than studies that have not taken into account the impact of poor motor coordination on daily living but indicates that poor coordination is an important, and often hidden, cause of disability in childhood.
Developmental coordination disorder (DCD) is defined, using the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), as a condition marked by a significant impairment in the development of motor coordination, which interferes with academic achievement and/or activities of daily living (ADL). These difficulties are not due to a general medical condition (eg, cerebral palsy) and are in excess of any learning difficulties if present.1,2
The calculated prevalence of DCD ranges from 1.4% to 19%, depending on case definition.3–5 However, high prevalence figures reflect the number of children who fail a standardized test of motor coordination (usually defined as the 15th percentile of the population), rather than the number of children with severe coordination difficulties who have functional impairment in their ADL.
The aim of this study was to calculate the prevalence of DCD in children, at 7 years of age, in a large UK birth cohort using strict inclusion and exclusion criteria derived from the DSM-IV definition.
The prevalence sample was drawn from the Avon Longitudinal Study of Parents and Children (ALSPAC), an ongoing population-based cohort study designed to investigate the interaction of environment and genotype on the health and development of children.6 The cohort recruited 14 541 pregnant women in the geographically defined area of Avon, Southwest England, with an expected date of delivery between April 1, 1991, and December 31, 1992, resulting in 14 062 live births.6 The total sample of live births was increased to 14 610 children by recruiting eligible children, at 7 years of age, who missed enrollment in the original cohort.
Mothers of infants in the ALSPAC were comparable with the rest of the United Kingdom at the 1991 census.6 Data have been prospectively collected on a variety of developmental traits using parental self-completion questionnaires, hands-on assessments, and linked education and health records. Antenatal and postnatal demographic data have also been collected.
Ethical approval for the ALSPAC was obtained from the research ethics committees of United Bristol, Southmead, and Frenchay Health Care Trusts, and the study was monitored by the ALSPAC Ethics and Law Advisory Committee.
At the age of 7.5 years, ALSPAC children were invited to attend an assessment clinic, called the Focus@7. This assessment included a motor coordination testing session designed for children aged 7 and 8 years. In total, 7256 7- and 8-year-old children started the motor coordination session (3684 boys, 3572 girls); the mean age was 7.5 years (SD: 2.8 months) (Fig 1).
The 7256 7- and 8-year-old children who started the coordination session were compared with those who did not attend or were missing from the cohort of live births (n = 7211). Compared with the birth cohort, the Focus@7 population underrepresented ethnic minorities and families from lower socioeconomic backgrounds (Appendix 1).
Case Definition: Inclusion and Exclusion Criteria
DCD cases were defined by using inclusion (criteria A and B) and exclusion criteria (criteria C and D) derived from the DSM-IV and adapted for research using the 2006 Leeds Consensus Statement, which provides internationally agreed cutoffs to define DCD for clinical and research purposes.1,2,7 The ALSPAC has collected data that fulfill each of these criteria.
Criterion A: Marked Impairment in the Development of Motor Coordination1
The 5th percentile of an individually administered motor coordination test is the standard measure of severe motor coordination difficulties. Those children scoring between the 5th and 15th percentile are considered at risk of probable difficulties.7–9
The ALSPAC coordination test consisted of 3 subtests derived from the Movement Assessment Battery for Children (MABC).9 The subtests were chosen to test the 3 realms of coordination: balance (heel-to-toe walking), manual dexterity (placing pegs task), and ball skills (throwing bean bag into box). Principal component analysis with varimax rotation of the original standardization data set of the MABC (n = 264, within age band 2), obtained from Professor Henderson, DPE, BA, MA, PhD, AcSS, identified 3 groups of variables:
heel-to-toe walking, stork balance, and flower trail;
placing pegs and threading lace; and
throwing bean bag into a box and 1-hand catch.
Heel-to-toe walking, placing pegs, and throwing bean bag into a box best represented each group in turn (Appendix 2) (S. E. Henderson, DPE, BA, MA, PhD, AcSS, A. Barnett, PhD, MABC standardization data set, personal communication, 2008).
Trained ALSPAC examiners demonstrated each task; the children were then given a practice test to ensure understanding and to correct procedural errors. Test administration differed from the MABC manual in that only children who failed to understand the test or did something that resulted in a poor score (eg, dropped a peg and stopped to look for it) were allowed to repeat a trial.
In the heel- to-toe walking test, the child was asked to walk, heel to toe, along a straight line. The tester recorded the largest number of consecutive correct steps taken to a maximum of 15. In the placing pegs task, the time taken for the child to insert 12 pegs into a peg board 1 at a time, using first their preferred then their nonpreferred hand, was recorded. The testing of ball skills involved the child attempting to throw 10 single bean bags underarm into a box at a distance of 6 feet. The number of bean bags thrown into the box was recorded.
Age adjusted ALSPAC coordination scores for each subtest were derived in 6-month bands by comparing the scores of each child to raw scores of the cohort itself (>7000 7- and 8-year-old children).
Criterion B: The Disturbance in Criterion A Significantly Interferes With Academic Achievement or ADL1
Because of the complex relationship between motor coordination and academic achievement, the Leeds consensus statement recommends that handwriting be measured when considering the functional academic consequences for the diagnosis of DCD.7 Under UK law, state-funded schools are obliged to subscribe to a national curriculum that includes regular academic testing at “key stages.”10 The ALSPAC has linked educational authority data from literacy and numeracy testing at the end of key stage 1 (aged 7 years). As part of the literacy testing, there is a writing task, which is scored 1 to 4, with 4 being the best. A nationally set level of 2 is expected at this age. Children not achieving level 2 in their writing task were considered to have significant difficulties with handwriting.
Activities of Daily Living
A 23-item measure of ADL was derived from the ALSPAC parent-completed questionnaire administered when the child was 6 years and 9 months of age. This scale included questions on key areas in which children with DCD struggle: self-care skills, play skills, and gross and fine motor skills including drawing and copying.11 The questions, derived from the Schedule of Growing Skills II and the Denver Developmental Screening Test II, were developmentally age appropriate and represented skills the child would be expected to have achieved by the age of 81 months.12,13 The ADL scale is presented in Appendix 3.
Parents were asked to answer each question: “yes can do” (2 marks); “can do but not well” (1 mark); “can not do as yet” (0 marks); or “has not had the opportunity to do.” A score was produced by adding the individual item scores (0 to 2) and dividing by the total number of items (23) minus the number of items the child had not had a chance to attempt. Age adjustment was made by stratifying by the child's age when questionnaires were completed. Previous screening tests for DCD have used the lowest 15th percentile to define impairment.14,15 We chose a more stringent cutoff of the 10th percentile to define significant difficulties in ADL, because the ADL questions described age-appropriate tasks that children were expected to have achieved.
Criterion C: The Disturbance Is Not a Result of a General Medical Condition (eg, Cerebral Palsy or Muscular Dystrophy) and Does Not Meet Criteria for a Pervasive Developmental Disorder1
Neurologic and developmental conditions in the ALSPAC were identified from hospital and community health service notes and educational records from the Avon area between 1993 and 2003. Diagnoses were confirmed by clinical researchers using the International Statistical Classification of Diseases, 10th Revision classification system and anonymously linked to ALSPAC identifiers.16,17 Children with known visual, developmental, or neurologic conditions were thus excluded from prevalence figures. The exclusion of autism spectrum disorders (pervasive developmental disorder) is controversial, because there is often an overlap between these conditions and DCD.18 In accordance with the Leeds consensus statement, children with autism have not been excluded but are highlighted.7
Criterion D: If Mental Retardation Is Present, the Motor Difficulties Are in Excess of Those Usually Associated With It1
IQ was measured when children were 8.5 years of age by using a validated, shortened form of the Wechsler Intelligence Scale for Children–III.19–21 Children with an IQ of <70, as advised in the Leeds consensus, were excluded from the case definition of DCD.7
Demographic factors of the “DCD or probable DCD” and “non-DCD” groups were compared by using either unpaired Student's t tests or χ2 tests, as appropriate.
Inclusion Criteria A: Poor Motor Coordination
Of the 7256 children aged 7 and 8 years who started the Focus@7 coordination session, 7058 (97%) children completed the 3 subsections of the ALSPAC coordination test.
Age-adjusted item scores (0–5) were allocated using empirical percentiles of the raw test scores for the cohort in 6-month bands. Cutoffs were derived in the same way as the original standardization of the MABC, with the 2% of children scoring the worst for each individual subtest given an item score of 5, the next 3% given an item score of 4, etc (Table 1). These age-adjusted scores for each of the 3 subtests were summed giving an ALSPAC coordination impairment score out of 15.
Empirical percentiles of the ALSPAC coordination impairment scores allowed us to define the 5% of children with the highest impairment scores (severe coordination difficulties) as those scoring ≥7 (n = 324). Nine hundred seventy-five children had a coordination impairment score of between 4 and 6 (scores between the 5th and 15th percentile); these children were defined as having “probable coordination difficulties.” As scores are categorical, these cutoffs equated to 4.6% and 18.4% of the population, respectively (Fig 2).
Inclusion Criteria B: Handwriting and ADL
Data were available from the key stage 1 writing task for 11 710 children. By the end of key stage 1, 1924 (16.4%) children had not achieved a writing level of 2. Data from 7422 ADL scales were available (Fig 3).
Seven hundred twenty children (the bottom 10%) were defined as having significant impairment in their ADL. This equates to not being able to do 7 items well. Children scoring between the 10th and 15th percentile were considered to have probable impairment in ADL. There was no significant difference in mean ADL test scores in children who did compared with those who did not attend the Focus@7 motor coordination session (mean score difference: 0.001 [95% confidence interval (CI): −0.006 to 0.009]); P = .75; Appendix 1).
Exclusion Criteria C and D: Known Neurologic Condition and IQ
In total, 110 children had visual, developmental, or neurologic diagnoses. Of the 7354 children who took the IQ test, 136 had an IQ of <70.
One hundred forty-nine children met DSM-IV inclusion criteria for DCD, that is, had severe motor coordination difficulties (the lowest 5th percentile of the ALSPAC coordination impairment score), that interfered with either ADL (lowest 10% ADL scale) or academic achievement (not meeting national standards for handwriting). We excluded those 26 children with a known neurologic diagnosis or an IQ of <70, leaving 123 children who met full DSM-IV criteria for DCD. Our denominator comprised the 6990 children, aged 7 and 8 years, who started the Focus@7 coordination session and had results for the ADL questionnaire or writing test (Fig 1). The prevalence of DCD was thus 1.8%, 18 per 1000 children at a mean age of 7.5 years (SD: 2.9 months). The gender ratio was 1.9:1 males to females.
Exclusion of those children with an IQ of <70 or a known neurologic diagnosis decreases the denominator to 6867 but does not alter the prevalence of DCD.
Broader inclusion criteria can be used to compare ALSPAC to other studies and consider those children with “probable DCD.”4,9 An additional 223 children were considered to have probable DCD: coordination impairment score between the 5th and 15th percentile, scores between the 10th and 15th percentile on the ADL scale, or failed the writing test. As such, 346 children, 4.9% of 7- to 8-year-olds, could be considered to have DCD or probable DCD (gender ratio 1.7:1 males to females). Ten of the 123 children with DCD and 9 of the 222 children with probable DCD had an autism spectrum disorder diagnosis.
One hundred ninety-eight children attended the coordination session but refused testing. One hundred eighty-seven of these children were included in the denominator as they started the session and had data from their writing tests or ADL (Fig 1). Potentially some of these “test refusers” had poor coordination. Fifty-seven of these 187 children scored below the 15th percentile on the ADL scale or failed their writing test. However, 15 of these 57 met our exclusion criteria C and D. If we consider the remaining 42 children as having probable DCD, the prevalence of DCD or probable DCD changes from 4.9% to 5.5%.
There was an increased risk of DCD or probable DCD with rented housing tenure and lower maternal socioeconomic group (SEG). There was also an association with lower birth weight (<2500 g) and younger gestation (<37 weeks) (Table 2).
Our study is the first to use strict DSM-IV criteria to define a population of children with DCD within a cohort of >7000 children. We have used the best available current guidelines and a multistage approach to identify children with DCD. The strength of the study is the prospectively collected data and the linkage of coordination testing to ADL, education, and health records. The main limitation, as with all longitudinal studies, is the missing data because of differential dropout.
Our prevalence level of 1.8%, at 7.5 years, is comparable to the DCD prevalence of 1.4% to 4% described by Wright and Sugden,3 and 2.7% described by van Dellen et al.22,23 Both authors used a 2-step procedure where a screening test was followed by a motor coordination test. However, in both studies, motor coordination testing was only performed on children who failed the screening test. IQ was measured by van Dellen et al but not by Wright and Sugden. It is unclear if children with an IQ of <70 were excluded from prevalence figures.3,23
Previous large studies have in general ignored the disability caused by motor coordination delay and thus defined clumsiness rather than DCD. Using nonstandardized neurologic testing in the Collaborative Perinatal Project, 8.1% of the 30 000 US children studied were defined as having “poor coordination”24; 13.5% of Swedish 7-year-old children were defined as having either moderate or severe DCD, again dependent solely on motor coordination ability.4 All children defined as having DCD were said to have functional impairment, but this was not quantified.
There is an established link between children born prematurely and DCD.25,26 Our finding of an increased risk of DCD with lower gestation and birth weight supports this link and adds credence to our case allocation methods.
Because of time constraints with the large number of children assessed, the ALSPAC coordination test consisted of subtests rather than the entire MABC. However, each of the selected subtests has been shown to have concurrent validity with other similar coordination tests (P < .001) and represents the 3 domains of coordination using principal component analysis.10,27 With the large sample size available, we were able to use the raw scores of these tests within the cohort, in 6-month age bands, to classify children as having severe or probable coordination difficulties as below the 5th and between the 5th and 15th percentile of the cohort, respectively. The ability of the ALSPAC coordination test to predict children with severe coordination impairment and thus fulfill criteria A of the DSM-IV is supported by the significant association of the ALSPAC coordination impairment score with the ADL score (P < .001). However, because the full MABC was not performed on the ALSPAC cohort, direct comparison between the ALSPAC coordination test and the full MABC, a test that has a well-established predictive validity, cannot be made.
Interference with ADL is not quantified in the DSM-IV criteria for DCD.2 It is, however, fundamental to diagnosis, because it implies functional difficulties and ensures case allocation does not rely solely on cutoffs from motor testing. In their review of research criteria used in DCD, Geuze et al8 found that only 50% of DCD studies mentioned interference with ADL, mainly as referral to a clinical service, and 9% mentioned interference with school life. We have created a novel ADL scale incorporating age-appropriate functional motor activities and defined severe impairment by using a stringent cutoff to account for potential parental reporting bias and the age-appropriate nature of the tasks.
We found an increased risk of DCD in the lowest SEGs. The Focus@7 coordination session underrepresented children who failed their key stage 1 writing test. The fact that these children are missing is likely to reflect their SEG and maternal education, known risk factors for loss to follow-up within ALSPAC. Therefore, our study may represent a minimum prevalence because of differential cohort drop out. With decreased recognition of disability and access to health care in the poorest families, DCD may add to the cumulative effect of social disadvantage on child health and represent a hidden burden of disability in the community.28
We found that 18 of 1000 7-year-olds have DCD according to strict DSM-IV criteria and that 49 of 1000 7-year-olds have DCD or probable DCD. To develop effective prevention and targeted intervention strategies, future longitudinal research should aim to identify which of these children with motor coordination difficulties are at risk of long-term psychological morbidity and educational failure.
Dr Lingam was funded by a personal award scheme, researcher development award from the United Kingdom National Institute for Health Research.
We are extremely grateful to all the families who took part in this study, the midwives for their help in recruiting them, and the whole ALSPAC team, which included interviewers, computer and laboratory technicians, clerical workers, research scientists, volunteers, managers, receptionists, and nurses. The United Kingdom Medical Research Council, the Wellcome Trust, and the University of Bristol provided core support for the ALSPAC. We are also extremely grateful to professor Sheila Henderson and Dr Anna Barnett for their advice and access to the original MABC standardization cohort.
- Accepted December 16, 2008.
- Address correspondence to Raghu Lingam, MBChB, MRCPCH, MSc, University of Bristol, Centre for Child and Adolescent Health, Department of Community-Based Medicine, Bristol B566JS, United Kingdom. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
This publication is the work of the authors, and Dr Lingam will serve as guarantor for the contents of this article.
What's Known on This Subject
There has been limited work looking at the prevalence DCD using the DSM IV criteria. High prevalence figures reflect the number of children that fail a standardised motor coordination test rather than those children who also have functional impairment.
What This Study Adds
Our study is the first to use strict DSM IV criteria to define the prevalence of DCD within a cohort of over 7000 children. We found a prevalence of 1.8% which used linkage of coordination testing to daily activities, education and health records.
- ↵American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition. Washington, DC: American Psychiatric Association; 1994
- ↵American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition Text Revision. Washington, DC: American Psychiatric Association; 2000
- ↵Sugden DA, Chambers M, Utley A. Leeds consensus statement 2006. Available at: www.dcd-uk.org/consensus.html. Accessed January 4, 2008
- ↵Henderson SE, Sugden DA. Movement Assessment Battery for Children Manual. Sidcup, United Kingdom: Psychological Corporation; 1992
- ↵HM Government: Department for Children, Schools, and Families. Higher Standards, Better Schools For All. London, United Kingdom: HM Government; 2005
- ↵Geuze RH. Motor impairment in DCD and activities of daily living. In: Sugden DA, Chambers M, eds. Children With Developmental Coordination Disorder. London, United Kingdom: Whurr; 2005:19–46
- ↵Bellman M, Lingam S, Aukett A. Schedule of Growing Skills. 2nd ed. London, United Kingdom: NFER-Nelson Publishing Co; 1997
- ↵Frankenburg WK, Dodds J, Archer P. The Denver II: a major revision and restandardization of the Denver Developmental Screening Test. Pediatrics.1992;89 (1):91– 97
- ↵World Health Organization. International Classification of Diseases, 10th Revision. Geneva, Switzerland: World Health Organization; 1992
- ↵Weschsler D, Golombok S, Rust J. Wechsler Intelligence Scale for Children. United Kingdom Manual. 3rd ed. Sidcup, United Kingdom: Psychological Corporation; 1992
- ↵van Dellen T, Vaessen W, Schoemaker M. Clumsiness: definition and selection of subjects. In: Developmental Biopsychology: Experimental and Observational Studies in Children at Risk. Ann Arbor, Michigan: University of Michigan Press; 1990:135–152
- ↵Nichols PL, Chen TC. Minimal Brain Dysfunction: A Prospective Study. Hillsdale, NJ: Lawrence Erlbaum Associates Inc; 1981
- ↵Jongmans M, Mercuri E, de Vries L, et al. Minor neurologic signs and perceptual-motor difficulties in prematurely born children. Arch Dis Child Fetal Neonatal Ed.1997;76 (1):F9– F14
- ↵Bauman LJ, Silver EJ, Stein RE. Cumulative social disadvantage and child health. Pediatrics.2006;117 (4):1321– 1328
- Copyright © 2009 by the American Academy of Pediatrics