For the last 100 years, poor motor coordination in children has been recognized as a developmental problem.1 As early as 1937, these children were classified as “clumsy.”1 Since then, other terms such as “motorically awkward,” “motor impaired,” and “physically awkward” have been used to describe these children, and the terms “developmental apraxia” and “perceptual motor difficulties” have been used to characterize this developmental problem.2,3 Since the 1994 International Consensus Conference on Children and Clumsiness, the term “developmental coordination disorder” (DCD) has been used to describe the condition of children with motor incoordination.1,4

The purpose of this article is to provide the following information about DCD: (1) definition, (2) prevalence, (3) etiology, (4) discussion regarding the difficulties in classifying these children, (5) common characteristics, (6) long-term prognosis, and (7) brief review of treatment approaches.

Definition of DCD

Developmental coordination disorder, a chronic and usually permanent condition found in children, is characterized by motor impairment that interferes with the child's activities of daily living and academic achievement.3,5 In order for a child to be diagnosed with DCD, these motor impairments must negatively affect some other aspect of his or her life.6 Impairment alone, however, does not qualify a child for the diagnosis of DCD; the motor impairment must not be caused by or have the symptoms of an identifiable neurological problem.2,5 That is, the child must not have any disturbances of muscle tone (ataxia or spasticity), sensory loss, or involuntary movements. If mental retardation is present, the testable IQ of the child must be greater than 70 and the motor impairments must be greater than what would normally be expected for children with mental retardation.5 Finally, a child diagnosed with DCD must not meet the criteria for a diagnosis of pervasive developmental disorder.6


Developmental coordination disorder appears to be a fairly common disorder of childhood and is usually identified in children between 6 and 12 years of age. Ten years ago, researchers7,8 estimated that DCD occurred in 10% to 19% of school-aged children. With a more precise definition of DCD, the current prevalence is estimated to be between 5% and 8% of all school-aged children,5,911 with more boys than girls (2:1) being diagnosed with DCD.12 This difference may reflect higher referral rates for boys, because the behavior of boys with motor incoordination may be more difficult to manage at home and in the classroom.8 In addition, a higher incidence of DCD may be found among children with a history of prenatal or perinatal difficulties.3


Due to the heterogeneity of DCD, finding its cause has been difficult.3 Several theories speculate that the etiology of DCD is part of the continuum of cerebral palsy5,13; is secondary to prenatal, perinatal, or neonatal insult3; or is secondary to neuronal damage at the cellular level in the neurotransmitter or receptor systems.14 Hadders-Algra14 based her view that DCD is a result of damage at the cellular level on evidence that cerebral palsy is often caused by prenatal damage that cannot be identified by current diagnostic techniques.

Although both standard and nonstandard functional tests are available to identify the specific disabilities experienced by a child with DCD, relating the observed disabilities to the primary impairment(s) or any possible neuropathology is not easily accomplished. The problems experienced by children with DCD are believed to emanate from abnormalities in neurotransmitter or receptor systems rather than from damage to specific groups of neurons or brain regions.15 Children's difficulties with coordination can result from a combination of one or more impairments in proprioception, motor programming, timing, or sequencing of muscle activity. A number of theories have evolved in an attempt to shed light on the specific neuronal processing deficits that contribute to DCD. Current models used to explain the neural regulation of posture and movement during development can serve as a basis for the examination and management of individuals with DCD. Using these models, many of the deficits of motor control observed in these children can be described.

A variety of theoretical models exist to explain the role of the nervous system in motor development. Forty years ago, the primitive reflex model was a generally accepted theory used to explain how the brain regulates early motor behavior.16 As development proceeded, the higher centers exerted increasing control over the lower reflexes.16 These earlier models were based on a hierarchy of motor control in which higher centers were capable of planning and executing a motor plan without external or internal feedback from lower centers of the central nervous system (CNS).

The more recently proposed systems model suggests a more complex interaction among various levels of the CNS. In the systems model, sensory feedback is interpreted by the CNS, and the appropriate movement strategy is selected based on current experience, the state of the internal and external environment, and memory of similar movements. Edelman's neuronal group selection theory includes aspects of both of these models and proposes that functional groups of neurons exist at all levels of the CNS.17 These neuronal groups are determined by evolution, but their functional integrity is dependent on afferent information produced by movement and experience.17 In this regard, these genetically determined collections of interconnected neurons (neuronal groups) in both cortical and subcortical structures serve as an early repertoire for motor behavior or receipt of specific sensory information.14,17,18

According to neuronal group selection theory, motor development proceeds in 2 phases.15 The first, the phase of primary variability, is characterized by crude and erratic motor activity that does not require sensory information for its initiation or guidance. These self-generated movements give rise to afferent (visual, kinesthetic) inputs that reinforce more specific synaptic connections within each group. An intermediate period in which effective patterns are selected is followed by the secondary variability phase. In this phase, sensory and motor factors interact to establish the intercellular connections that produce the specific and complex muscle contraction patterns that characterize coordinated, goal-directed movement. Reciprocal connections between groups subserving movements in various body parts and representing different parts of visual space are reinforced with each repetition of a particular function. As the more efficient movement patterns are practiced, the appropriate synaptic circuits are reinforced and subsequently established.14,17,19,20

Difficulties in Classifying Children With DCD

The literature includes a wide variation in terminology and criteria to describe DCD. This variation has made studying the causes of DCD and developing treatment approaches for the child with DCD difficult. In their analysis of clinical trial data, Macnab et al21 identified 5 different subtype profiles of DCD. The first subtype included children with better gross motor than fine motor skills, although both were still below normal while standing balance and visual-perceptual skills were both within normal ranges. Compared with children of the same age with DCD, children in the second subtype scored high on measures of upper-limb speed and dexterity, visuomotor integration, and visual-perception skills, but they demonstrated poor performance on measures of kinesthetic ability (accuracy in discriminating movement and position of the upper limbs) and balance. Children in subtype 3 demonstrated the greatest overall motor involvement and were the only subtype to have difficulty with both kinesthetic and visual skills. Compared with their peers with DCD, children in subtype 4 performed well on kinesthetic tasks but demonstrated poor performance on tasks requiring visual and dexterity skills. Children in subtype 5 demonstrated poor performance on measurements of running speed and agility compared with their peers with DCD; however, they performed well relative to their peers with DCD in the tasks involving visual-perception skills. The development of different classification systems for DCD may have been influenced by the design of motor tests.21 For example, items testing one motor skill may be influenced by a related motor skill (eg, ball-throwing skills cannot be separated from visuomotor skills). In addition, a test may have an over-representation of one skill that could unduly influence the child's performance on the test. For example, having a greater number of items testing gross motor rather than fine motor skills could either positively or negatively influence a child's score, depending on the child's specific strengths and weaknesses.

Another difficulty in interpreting the literature on DCD is the lack of inclusion criteria. Geuze et al22 reviewed 164 publications on the study of DCD and found that only 60% of the studies had objective inclusion criteria. Because of this lack of inclusion criteria, Geuze et al recommended that a child scoring below the 15th percentile on standardized tests of motor skills and having an IQ score above 69 would qualify for a diagnosis of DCD.

The inconsistency among standardized motor tests used to identify children with DCD is another problem. In one study,5 the Bruininks-Oseretsky Test of Motor Proficiency (BOTMP) and the Movement Assessment Battery for Children (M-ABC) were administered to 157 children with DCD and 155 children with no motor difficulties; the test results were in agreement 82% (kappa=.62) of the time in distinguishing children who had DCD from children who did not have DCD. This is considered a substantial level of agreement.23 In a study of 202 children (101 with DCD and 101 without DCD), the BOTMP and M-ABC agreed only 67% of the time (kappa=.41), which was considered a moderate level of agreement.24 Because the BOTMP and the M-ABC are 2 of the most commonly used tests for identifying children with DCD, the potential lack of agreement by these tests in identifying children who have DCD is a concern.

Two primary factors may explain the difference in outcomes between the BOTMP and M-ABC when used to identify children with DCD.24 First, the BOTMP allows the tester to verbally prompt and correct the child during the testing procedure, allowing the child who is dependent on more external controls to do better on the BOTMP. The BOTMP tends to under-identify children with DCD. Second, the M-ABC requires more careful instruction on the part of the examiner and allows more opportunities for the examinee to practice, but does not allow any verbal or physical prompting by the examiner. Children with attention problems may have more difficulty with the careful instructions for the M-ABC.

Another difficulty in classifying children with DCD is the overlap with other disorders. Approximately 41% of children with attention-deficit/hyperactivity disorder (ADHD) and 56% of children with learning disabilities also have DCD.5,21 Further confusing the classification scheme is that the terms “developmental coordination disorder,” for which no identifiable organic brain damage is present, and “apraxia,” which is caused by identifiable brain damage, have been used interchangeably.3

Characteristics of Children With DCD

Children with DCD may have a wide range of dysfunctions. These dysfunctions can be grouped into 3 areas: gross motor, fine motor, and psychosocial.

Gross Motor

Many children with DCD have neurological soft signs such as hypotonia, persistence of primitive reflexes, and immature balance reactions that interfere with gross motor development.5,25 These children also may demonstrate an awkward running pattern, fall frequently, drop items, and have difficulty imitating body positions and following 2- to 3-step motor commands.8 Because of their gross motor problems, children with DCD also perform poorly in sporting events,2 possibly due, in part, to their slow reaction and movement times.7 Their decreased participation in sports may result in decreased muscle force.8

Fine Motor

Difficulty with handwriting or drawing often is the first identifiable sign of a fine motor problem and is the most frequently mentioned motor problem experienced by children with DCD. Children with DCD frequently have difficulty planning and executing other fine motor skills such as gripping and dressing.2629


Unfortunately, children with DCD may experience problems not limited to fine or gross motor areas. These children also may experience psychosocial problems at school. Children with DCD may have learning disabilities or reading problems and may be at increased risk for lower intelligence.5,8 They may act out in class more than other children,13 may be the class clown, and may exhibit less socially desirable means of gaining recognition and friends.8 Adolescents with DCD have been found to have fewer friends, and they have more feelings of low self-worth and more anxiety than peers without DCD and younger children with DCD.30


Historically, parents have been told not to worry about their child's clumsiness because the child will outgrow the problem.31 However, current researchers in the area of DCD report that the children do not outgrow clumsiness and that, without intervention, they will not improve.1,8,12,27,31 Losse et al31 tested 17 children aged 6 years and retested them at age 16 years. The children with motor difficulties at 6 years of age continued to exhibit problems at 16 years of age.

In another study,32 818 children with DCD were tested for reading comprehension at age 7 years and then again at age 10 years. A positive correlation in poor reading comprehension existed for children with DCD at 7 and 10 years of age.

A follow-up study was conducted on 22-year-old individuals (N=55) who at age 7 years had either DCD or attention-deficit/hyperactivity disorder (ADHD), or both.33 The children with DCD and those with both DCD and ADHD had poorer outcomes than their similarly aged peers without DCD and children with ADHD only. The children with DCD and those with both DCD and ADHD were found to have had more criminal offenses, more incidences of substance abuse and other psychiatric disorders, and lower levels of schooling.

Treatment Approaches

Treatment approaches used by occupational therapists and physical therapists can be broadly categorized into either bottom-up or top-down approaches (Tab. 1).10,12,3436 Bottom-up approaches are based on hierarchical theories of motor control. These theories tend to explain the remediation of motor dysfunction through activation of higher levels of neuronal functioning in a child. The bottom-up approaches frequently used in managing children with DCD are sensory integration, the process-oriented treatment approach, and perceptual motor training.34

Table 1.

Summary of Bottom-Up Versus Top-Down Approaches

In sensory integration therapy, the child is provided sensory stimulation designed to promote motor development and higher cortical learning. A child undergoing sensory integration therapy may show some gains in motor development, but these gains often do not generalize to functional skills.34

Kinesthesia (the perception of one's own body parts, weight, and movement) is integral to the acquisition of motor skills in process-oriented treatment approaches. Therapeutic intervention with process-oriented treatment is based on specifically designed kinesthetic training activities. As described by Laszlo and Bairstow,37 this approach has an inherent reward system built into it through its use of positive reinforcement, presentation of desirable activities within the capabilities of the child, and judicious progression of the level of difficulty. The usefulness of the process-oriented treatment approach has been the subject of considerable study.9,10,37,38 Sims and colleagues35 suggested that much of the success of this approach can be attributed to a strong motivation effect, fostered by positive feedback and a sense of self-competence.

Perceptual motor training, an eclectic approach, offers the child with DCD a wide range of motor experiences along with ample opportunities to practice these skills. Often, in outcome studies, children who receive perceptual motor training are compared with children who receive either sensory integration therapy or process-oriented treatment. Children receiving perceptual motor training have been found to demonstrate motor improvements equal to or greater than those of children receiving either sensory integration therapy or process-oriented treatment.34 While perceptual motor training, like process-oriented treatment, may promote learning through positive feedback and reinforcement, these techniques do not facilitate cognitive and problem-solving strategies to the extent that top-down approaches do.

Top-down approaches typically use a problem-solving approach to motor skill development and have been greatly influenced by the dynamic systems approach to motor learning and control. This approach suggests that motor skills develop from an interaction of many systems, both internal and external to the child.39 Top-down approaches also emphasize the context in which motor behavior occurs. Task-specific intervention and cognitive approaches or strategies are the 2 most commonly used.

Task-specific intervention focuses on direct teaching of a skill. The theoretical foundation for task-specific intervention in a child's motor performance is the result of learning focused on a specific task. Motor tasks are broken down into steps, with each step taught independently and then organized to accomplish the entire task.34 Children managed with this approach have demonstrated gains in motor skills.34

Cognitive approaches to motor development emphasize active problem solving.36 The cognitive approach strategy involves the GPDC framework:

Goal: What am I going to do?

Plan: How am I going to accomplish the skill?

Do it: Go ahead and perform the skill.

Check: How well did my plan work?

The child uses verbal self-guidance to apply the GPDC framework to motor learning. In this approach, the therapist acts as a guide by helping the child figure out how to improve his or her motor performance on various motor skills.36

Like task-specific intervention, the results of initial studies of cognitive approaches are encouraging. In one study,25 10 children with DCD who were treated with a cognitive approach to motor development were compared with 10 children who were treated with a bottom-up approach. The children were matched for diagnoses, age, and handedness. Both groups showed improvements on various standardized motor tests after receiving 10 treatment sessions. However, children in the cognitive approach group maintained motor skill longer and generalized to nonclinical situations better than children who were treated using the bottom-up approach.

Task-specific interventions and the cognitive approach both provide repetition and practice of specific motor skills, and the cognitive approach has the added advantage of promoting independent problem solving. The greater success of top-down approaches, when compared with bottom-up approaches,36,40 might be a result of the top-down approaches' inclusion of both spatial and motor learning sequences, combined with requirements for attention to task and working memory as the child actively engages in problem-solving activities. The results of these studies, in addition to outcome measures of the different approaches in children with DCD, would suggest that approaches that integrate systems theory (with emphasis on sensory information being only part of the picture) and motor learning theory might be most effective for these individuals. Table 2 provides a brief summary of several recent clinical trails evaluating the effectiveness of various treatment approaches used with children with DCD.10,35,36,4043

View this table:
Table 2.

Overview of Selected Clinical Studies Evaluating Treatment Approaches for Children With Developmental Coordination Disorder (DCD)a

The neuronal group selection theory can provide a framework for interpreting the differences in reported outcomes among the various approaches. While children with moderate to severe cerebral palsy (CP) have a limited repertoire of primary neuronal networks, which guide crude, nonspecific movements, children with DCD are believed to experience difficulty at the level of secondary variability, that is, in selecting and reinforcing the most efficient and effective pathways for a given situation.15,44 During normal postnatal development, experience plays a primary role in the neuronal group selection process that establishes the circuitry necessary for efficient, goal-directed, and coordinated movements.17 According to the neuronal group selection theory, this refinement of motor skill occurs during the stage of secondary variability as a combined result of trial-and-error exploration of neuronal groups, selection of specific neurons within each group, repetition of synaptic firing within and among neuronal groups, and sensory experience.

Functional synaptic connections, which act in parallel and involve cortical and subcortical (striatal and cerebellar) structures, form following exposure to a variety of motor experiences. Formation of these connections is highly dependent on sensory information. Bottom-up approaches such as sensory integration, process-oriented treatment, and perceptual motor training emphasize sensory experience, with less emphasis on cognitive processing and cortically driven motor programming. Although an intervention based entirely on information processing may provide the experience necessary to select the most effective neuronal networks, bottom-up approaches may not provide sufficient opportunity for motor practice of cognitively initiated and goal-directed tasks in order to reinforce and establish these connections. Top-down approaches focus less on the specific impairments contributing to decreased coordination and more on the gestalt of coordinated movement, that is, the dynamic interrelationships among a number of CNS structures and systems and the environment within which the task is performed.

Treatment based on a top-down approach uses task-specific interventions that provide the child the opportunity to engage in conscious problem solving, while coincident afferent input provides subcortical structures the feedback and error signals needed to identify and select the most efficient movement strategies for the task.36 Because sensorimotor integration, internal representation of motor programs, and appropriate motor commands occur at the level of secondary variability,15 the emphasis on sensory rather than cognitive factors by the bottom-up approaches may, in part, be responsible for the observed differences in outcomes following treatment.

A number of motor learning theories have been proposed in an attempt to explain the process through which previously learned actions are incorporated into more complex movements.39 According to the motor control and procedural learning theory proposed by Hikosaka and colleagues,45 motor sequence circuits that involve the basal ganglia and cerebellum become encoded following long-term practice. Once these bidirectional and parallel functioning neuronal circuits become established through practice, the child is able to incorporate previously learned sequential motor actions into yet more complex movements. Hikosaka and colleagues proposed 2 stages to such learning, the first of which relies primarily on sensory input to encode neuronal sequencing. It is during the second stage that the sequential processes necessary to accomplish the specific motor tasks become firmly established through the parallel and sequential pathways involving the basal ganglia and cerebellum. Learning new sequences in the first stage of learning requires attention and working memory. The motor sequences then become established in the late stage of learning through repetition and practice. Interventions using top-down approaches meet both of these requirements, whereas bottom-up approaches stress serial information processing only. Determining the location and nature of the neural deficiency in children with DCD is a difficult, if not impossible, task. Motor control processes are complex and depend on integrated functioning of sensory, perceptual, cognitive, and motor systems. Not only are children with DCD a heterogeneous group in terms of functional disabilities but, the specific locus of the problems observed can vary greatly from one child to the next. Because of these factors, an integrated approach to the management of children with DCD is advocated.


Developmental coordination disorder is a complex disorder affecting approximately 5% to 6% of school-aged children. Without intervention, these children will continue to exhibit poor motor skills and show deficits in other areas as well. Directions for future research may include determining: (1) the most appropriate level of intervention intensity, (2) which interventions produce results that generalize to the environment and provide long-term improvement in motor function, (3) what effect environmental adaptations have on the child's motor performance, and (4) whether improved motor skills lead to improved academics and, if so, the process involved that leads to the improvement.


  • Mr Barnhart provided concept/idea, and Mr Barnhart and Ms Davenport provided writing. Dr Epps and Dr Nordquist provided consultation (including review of manuscript before submission).


    View Abstract