Human Movement Science 22 (2003) 461478 www.elsevier.

com/locate/humov

Motor and functional skills of children with developmental coordination disorder: A pilot investigation of measurement issues q
Sylvia Rodger a,*, Jenny Ziviani a, Pauline Watter b, Anne Ozanne c, Gail Woodyatt c, Elizabeth Springeld
a

Department of Occupational Therapy, Childrens Research Unit: Communication, Occupation, Movement, School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, Qld 4072, Australia b Department of Physiotherapy, Childrens Research Unit: Communication, Occupation, Movement, School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, Qld 4072, Australia c Department of Speech Pathology and Audiology, Childrens Research Unit: Communication, Occupation, Movement, School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, Qld 4072, Australia

Abstract This paper reports on the motor and functional outcomes of 20 children with developmental coordination disorder (DCD) aged 48 years consecutively referred to a pediatric physiotherapy service. Children with a Movement ABC (M-ABC) score less than the 15th percentile, and with no concurrent medical, sensory, physical, intellectual or neurological impairments, were recruited. The Motor Assessment Outcomes Model (MAOM) [Coster and Haley, Infants and Young Children 4 (1992) 11] provided the theoretical base for measurement selection, and preliminary ndings at the activities and participation levels of the model are reported in this article. Children with DCD performed at the lower end of the normal range on the Peabody Developmental Motor Scales (ne motor total score) (M 85:65, SD 12.23). Performance on the Visual Motor Integration Test (VMI) standard scores was within the average range (M 96:15, SD 10.69). Videotaped observations of the childrens writing and cutting indicated that 29% were left-handed and that a large proportion of all children (31%) utilized unusual pencil grasp patterns and immature prehension of scissors. Measurement at the

q Based on conference presentation at Developmental Coordination Disorder (DCD), V Ban, Alberta, Canada, 1416 May 2002. * Corresponding author. Tel.: +61-7-3365-1664; fax: +61-7-3365-1622. E-mail address: s.rodger@mailbox.uq.edu.au (S. Rodger).

0167-9457/$ - see front matter 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.humov.2003.09.004

462

S. Rodger et al. / Human Movement Science 22 (2003) 461478

participation level involved use of the Pictorial Scale of Perceived Competence and Social Acceptance (PCSA) and Pediatric Evaluation of Disability Inventory (PEDI). Overall, these young children rated themselves towards the more competent and accepted end of the PCSA over the dimensions of physical and cognitive competence and peer and maternal acceptance. The PEDI revealed generally average performance on social (M 49:98, SD 16.62) and mobility function (M 54:71, SD 3.99), however, self-care function was below the average range for age (M 38:01, SD 12.19). The utility of the MAOM as a framework for comprehensive measurement of functional and motor outcomes of DCD in young children is discussed. 2003 Elsevier B.V. All rights reserved.
PsycINFO classication: 2820; 3253 Keywords: DCD; Participation; Function; Self-ecacy; Measurement

1. Introduction The clinical diagnosis of developmental coordination disorder (DCD) involves ascertaining whether a childs motor skills fall below those expected for his or her chronological age and intellectual ability. In addition, if Diagnostic and Statistical Manual IV (APA, 1994) criteria are being followed, these motor skill limitations must have a functional impact on the childs performance of daily activities and academic achievement. Discussion still surrounds the appropriate method of determining motor skill difculties for children with DCD. While some have advocated the Movement Assessment Battery for Children (M-ABC) (Henderson & Sugden, 1992), others have argued that it is not yet able to represent the gold standard for measurement (Crawford, Wilson, & Dewey, 2001). In particular, Crawford et al. found that dierent tests identied dierent children. They found low levels of agreement (less than 80%) between the Bruininks Oseretsky Test of Motor Prociency (BOTMP) and the M-ABC in the identication of children with DCD. Dewey and Wilson (2001) reported that the M-ABC identied more children as having DCD than the BOTMP, especially those with additional learning and attention problems. They found a moderate agreement of 82% or K 0:62 between the two tests. Despite this, the M-ABC has become increasingly used as a standardized measurement of motor skills for the identication of children with DCD, both clinically and in research. Henderson and Sugden (1992) proposed that scores below the fth percentile on the M-ABC indicate denite motor problems, while scores between the 6th and 15th percentile indicate borderline motor problems. Many researchers (e.g., Crawford et al., 2001; Dewey & Wilson, 2001) recognize scores below the 15th percentile on the M-ABC as being indicative of DCD, namely atypical or impaired motor performance. A less well documented measurement issue for children with DCD is how to determine functional manifestations of motor in-coordination (especially in terms of self-care skills) (Dewey & Wilson, 2001). Dewey and Wilson (2001) proposed that participation in school and daily living activities represent two distinct functional issues that should be examined independently. Many studies (e.g., Cantell, Smyth, &

S. Rodger et al. / Human Movement Science 22 (2003) 461478

463

Ahonen, 1994; Kaplan, Wilson, Dewey, & Crawford, 1998; Losse et al., 1991) have investigated the presence of academic diculties in reading, spelling, and math in children with DCD. The issue remains, however, whether these diculties are a manifestation of DCD. It could be argued that only motor-based school diculties such as handwriting, cutting with scissors, organization of desk and materials, keeping up in physical education classes should be considered in this diagnostic category. To date the functional impact of DCD has been well documented only in respect of decreased leisure and playground participation (e.g., Hay & Missiuna, 1998; Smyth & Anderson, 2000; Watkinson et al., 2001) and having a negative impact on childrens self-condence in physical abilities, social competence and social-emotional well being (e.g., Henderson & Hall, 1982; Schoemaker & Kalverboer, 1994). These studies have focused on children in the middle childhood years. More specic skill manifestations have not been reported. Measures such as the M-ABC Parent Checklist (Henderson & Sugden, 1992), Developmental Coordination Disorder Questionnaire (DCD Q) (Wilson, Dewey, & Campbell, 1998), and interviews with parents have been used to identify the degree of functional impact of the childs motor coordination diculties on academic achievement and activities of daily living. While the M-ABC Checklist and DCD Q were initially developed as screening tools, many of the items provide specic information about functional impact and the contexts in which these occur. Relevant items relate to walking around the classroom without colliding, using xed playground equipment, turning rope (M-ABC) and in the DCD Q items such as printing and writing in class is fast enough to keep up, cutting out magazine pictures, learning to ride a bike later than friends, and avoiding participation in sports (Wilson et al., 1998). Most of the research on DCD has been conducted with children aged 712 years with relatively few published studies involving younger children. At The University of Queensland School of Health and Rehabilitation Sciences Pediatric Clinics, we became aware of a number of young children under 8 years, being referred for assessment of motor diculties. The functional ramications of motor diculties in this age group have had little attention and provided the mandate for the current study. In order to comprehensively measure the motor and functional diculties experienced by young children with DCD, an assessment model was needed. The Motor Assessment Outcomes Model (MAOM) (Coster & Haley, 1992) was chosen based on a search of the literature and its specic relevance to children with motor skills diculties. This model (see Fig. 1) is consistent with the Conceptual Model of Disablements, initially proposed by Nagi (1991) and subsequently incorporated into the levels of functioning in ICIDH-2 model of human functioning and disablement (World Health Organisation, 1998). Essentially, this model discusses measurement at the level of sensori-motor performance components (impairment), which incorporates assessing foundations for the development of motor skills and underlying physical, developmental, sensory and motor capacities. The next level of perceptual motor skills equates with functional limitations (activities and activities limitations). Measurement at this level acknowledges the discrete motor skills required by the child to enable functioning in the environment, assesses prociency in developmental

464

S. Rodger et al. / Human Movement Science 22 (2003) 461478

Functional Performance (Disability) Participation Ability to engage in appropriate social roles. Influenced by context and environmental variable. Measured by observation, self-report or caregivers (know child in multiple contexts).

Figure 1. Figure Perceptual Motor Skills (Functional Limitation) Coster & 1. Motor Outcomes Assessment Model (adapted from Haley, 1992)Assessment Model (adapted from Coster & Haley, 1992) Activities Discrete motor skills enable child to function in environment. Require proficiency in developmental tasks, associated with standardized testing.

Sensorimotor Performance Components (Impairment) Body Function and Structures Foundations for the development of motor skills, underlying physical capabilities, intrinsic enablers of performance.

Fig. 1. Motor outcomes assessment model (adapted from Coster & Haley, 1992).

tasks, and is usually associated with standardized testing. The nal level of the hierarchy is functional performance or disability, (termed participation in the ICIDH-2). Participation refers to the nature and extent of a persons involvement in life situations in relation to impairment, activities, health conditions and contextual factors (WHO, 1998). This level addresses the childs ability to engage in appropriate social roles, inuenced by contextual and environmental variables. Measurement at this level usually requires observation, self-report, and reports from caregivers and significant others known to the child in multiple contexts. To date, Coster and Haley (1992) have primarily used their model for assessment of children with signicant physical and neurological disabilities such as cerebral palsy. There is no literature to support its use with children with milder motor diculties such as DCD. The major focus of this paper is to present some preliminary ndings of the functional impact of DCD on young children (48 years), and to investigate the utility of the Motor Assessment Outcomes Model, as a mechanism to guide this process.

2. Method 2.1. Participants Children with DCD consecutively referred to the Pediatric Physiotherapy Motor Clinic at The University of Queensland, Brisbane, Australia were recruited. Inclusion criteria were: (1) no known sensory, motor, neurological or intellectual impairment, (2) no known emotional or social problems which may impact on development, and

S. Rodger et al. / Human Movement Science 22 (2003) 461478

465

(3) M-ABC total impairment scores less than or equal to the 15th percentile, and (4) aged between 4 years (48 months) and 7 years 11 months (95 months). Twenty children, 12 males (60%) and 8 females (40%), whose ages ranged from 52 months to 93 months participated (M 69 months, SD 12 months). Nine children (45%) were at kindergarten or preschool, with the remaining 11 children (55%) in grades 1 or 2. Children in Queensland can commence grade 1 in January at the start of the school year as long as they have turned 5 by December 31 of the previous year. The M-ABC total impairment scores for the 20 children in the pilot group ranged from 10 to 21.5 (M 14:3, SD 3.69), placing them between the rst and 15th percentile (M 8:8, SD 4.52). 2.2. Instruments Consistent with the MAOM (Coster & Haley, 1992) measures were selected to represent each of the three levels of the model. At the Impairment level (body function and structures) the Neurodevelopmental Physiotherapy Assessment (NDPA) (Watter, 1996) was utilized to assess the underlying sensori-motor performance components. While it is acknowledged that these sensori-motor components impact on subsequent skill performance, the emphasis of this paper will be on describing performance and function at the activities and participation levels of the MAOM. Findings on the NDPA have been reported elsewhere (Watter, Ozanne, & Rodger, 2002). Instruments used to address performance at the activities level included: Peabody Developmental Motor Scale (PDMS) (ne motor) (Folio & Fewell, 1983), Developmental Test of Visual Motor Integration (VMI) (Beery & Buktenica, 1997), a videotaped sample of handwriting (name, Test of Legible Handwriting (TOLH)) (Larsen & Hammill, 1989), Handwriting Speed Test (HST) (Wallen, Bonney, & Lennox, 1996), and a videotaped sample of cutting using scissors. At the participation level of the model, the Pediatric Evaluation of Disability Inventory (PEDI) (Haley, Coster, Ludlow, Haltiwanger, & Andrellow, 1992) was completed by parent interview and the child completed the Pictorial Scale of Perceived Competence and Social Acceptance (PCSA) (Harter & Pike, 1983, 1984). In this way, a comprehensive mix of standardized measures (VMI, PDMS, HST, TOLH), parent report of the child in multiple contexts using a judgement based measure (PEDI) and therapists observations of pencil prehension, hand preference, scissor prehension, cutting accuracy and strategy, as well as child self-report (PCSA) (Harter & Pike, 1983) were utilized. The PDMS ne motor subscale has been widely used by occupational therapists to assess children with mild to moderate perceptual motor diculties (e.g., CaseSmith, 1995). The PDMS was standardized on 617 American children, evenly split between males and females, who represented by region the total population of the US. The distribution of Anglo American and non-white Americans (84% and 16%) respectively represented the US population statistics in the mid 1970s. Test retest reliability (n 38) for the ne motor subscales was r 0:99 over a period of one week. Inter-rater reliability (n 36) was r 0:99 for total test scores and r 0:94 for item scores in the ne motor subscale. Content validity was presented

466

S. Rodger et al. / Human Movement Science 22 (2003) 461478

as face and logical validity in terms of the description of test development in the manual (Folio & Fewell, 1983). Construct validity as determined by developmental changes with age (increasing performance with age), and correlations between age and total scores of r 0:99 for the ne motor scale. In addition, performance in skill areas was strongly related to normal development in non-clinical children. The PDMS was able to discriminate clinical with motor delays from non-clinical children signicantly and reliably for all ages except for infants under 5 months (Folio & Fewell, 1983). The VMI has been extensively used in research on children with DCD (e.g., Dewey & Wilson, 2001; Missiuna & Pollock, 1995). Normed in 1996 on 2, 614 children from 318 years in 5 major sections of the USA representative of the 1990 US census, the VMI is one of the most extensively used assessments of copying abilities. Split half reliability coecient of 0.88 is reported for the overall test, ranging from 0.76 to 0.91. Coecient alpha for the copying items was r 0:96. Testretest reliability coecients for 122 children who undertook the VMI twice three weeks apart was 0.87. Inter-rater reliability for the VMI was r 0:94 (n 100). Validity is presented in the form of concurrent validity with DTVP-2, and WRAVMA undertaken by 122 children. Moderate correlations of r 0:52 with WRAVMA and r 0:75 with DTVP-2 copying indicated adequate levels of concurrent validity with similar tests. Seven hypothesis in relation to construct validity were supported (Beery & Buktenica, 1997), including relationship between VMI abilities and increasing age r 0:89, relationship between VMI and intelligence, particularly performance IQ r 0:66, academic achievement (comprehensive test of basic skills) r 0:63, and nally, signicant dierences between children with a range of disabling conditions and typically developing children on the VMI have been reported. In addition, paediatricians ratings coupled with VMI results have been signicantly predictive of school grade failures and retentions (Beery & Buktenica, 1997). The PEDI has been extensively used with children with signicant neuromotor dysfunction (e.g., cerebral palsy, traumatic brain injury), however, it has not been widely used with children with milder motor dysfunction. Preliminary investigations by Ziviani and Wright (1995) with children who had been abused suggest that the PEDI may be a sensitive measure of function in children other than those with physical and multiple disabilities. Normed on 412 American children, equally distributed by gender, community size and geographic location, the PEDI is slightly over-represented with parents from higher educational status. Clinical samples of 102 children with paediatric trauma (n 46), severe disabilities (n 32) and neurological and developmental disabilities (n 24) have been studied. The developers report internal consistency reliability coecients ranging from r 0:95 to r 0:99 indicating excellent internal consistency. Inter-interviewer reliability ICCs are reported between 0.96 and 0.99 on all scales for the normative sample, and 0.841.00 for the clinical sample. Construct validity is reported in relation to change in functional performance with age, with strong developmental trends emerging. Evidence to support the division of functional skills and caregiver assistance as two separate constructs is also provided (Haley et al., 1992). Moderate concurrent validity with the Batelle Developmental Inventory Screening Test (r 0:700.73) and high correlations with wee

S. Rodger et al. / Human Movement Science 22 (2003) 461478

467

FIM (r 0:800.97) are also reported. Discriminant validity of PEDI as a determinant of clinical (n 102) versus non-clinical (n 412) status has also been demonstrated. The PCSA (Harter & Pike, 1983, 1984) is a self-report measure of young childrens social acceptance (peer and maternal) and perceived competence (physical and academic). The self-report measure was standardized on 90 preschoolers and 109 early grade school children, with groups equally divided by gender. Most of the children were Caucasian (96%) and attended schools in middle class suburbs. Chronbachs alpha for grade school children ranged from r 0:53 to 0.79. Three discriminant validity studies have been reported (Harter & Pike, 1984) in relation to the students and teachers ratings. Correlations between students and teachers ratings were higher within than across subscales, but were not strong (r 0:30 0.37). Klein and Magill-Evans (1998) examined the testretest reliability of the PCSA with 24 preschool and school age children with language delays over a period of 10 22 days. Moderate to good correlations were reported (r 0:620.81), with competence perceptions more stable than acceptance perceptions. Cognitive competence (r 0:81) was the most stable with maternal acceptance being the least stable (r 0:61). Concurrent validity between the PCSA and All About Me (AAM) (Missiuna, 1998) was demonstrated through highest correlations between PCSA physical subscale and competence factor scores and AAM total score (r 0:77) than with acceptance subscale scores (r 0:510.55). Michael (1990) supported the use for the PCSA for research purposes, but advised caution due to limited and inconclusive reliability, validity and normative data. 2.3. Procedure Ethical clearance was received from the Behavioural and Social Sciences Ethics Review Committee at The University of Queensland. Subsequently, all children within the study age range, referred to the Pediatric Physiotherapy Motor Clinic were invited to participate. All invited parents who provided written consent and their children were involved in the study. Children were rst seen by the physiotherapist who completed the M-ABC and a demographic information questionnaire to ensure study eligibility. Children who met the inclusion criteria, were subsequently assessed using the PNDA. On two further occasions, children were assessed individually using a battery of speech and language assessments by a speech pathologist (Ozanne, Woodyatt, Watter, Rodger, & Ziviani, 2002) and functional perceptual motor (PDMS, VMI, handwriting and cutting samples, HST, TOLH) and participation assessments (PEDI, PCSA) by an occupational therapist. Each assessment session lasted approximately one and a half hours. In this article, only the pilot ndings in relation to motor and functional skills are presented. The pilot results of the language assessment (Ozanne et al., 2002) and physiotherapy assessment (Watter et al., 2002) have been reported elsewhere. Standardized assessments were conducted according to instructions in the test manuals. Each child was videotaped undertaking a series of writing and cutting tasks. All children were videotaped whilst writing their name and undertaking the VMI, and if they were between 7 years and 7 years 11

468

S. Rodger et al. / Human Movement Science 22 (2003) 461478

months and able to write enough, the HST and TOLH were completed. In terms of scissor skills, children were videotaped cutting along a straight line, cutting out a square and a circle. The shapes used were a 50 mm black line, a 100 mm square, and a 100 mm circle. These cutting tasks were developed by Springeld (1998) who investigated the development of cutting skills in typically developing children from 2 to 5 1/2 years of age. Cutting was chosen as it is a complex tool use skill required by children in the preschool and early school years. It requires bimanual coordination, functional prehension of the scissors, appropriate cutting strategies (planning how to approach and complete the task), appropriate strategies for holding the paper for cutting with non-preferred hand, and motor accuracy and control. These shapes were chosen as they represent a developmental continuum from the simple act of cutting accurately (line), change of direction at corners (square) and a continuous cutting task around a shape with no direction change (circle) (Springeld, 1998). Accuracy was calculated using a clear plastic template placed over the shapes to determine deviation in millimeters from the cutting line. Videotapes of cutting and writing were subsequently rated by the last author (ES). Thirty nine percent of the videotapes were rated independently by the rst author (SR) to establish inter-rater reliability. For writing, the raters coded the childs preferred hand used during the VMI, name writing, TOLH and HST (if completed), pencil prehension based on the 11 pencil prehension patterns described by Schneck and Henderson (1990), and non-dominant hand use (rated as consistent, inconsistent, or no stabilization of paper). Inter-rater agreement for writing observations ranged from 79% to 100% (M 89%). In terms of cutting, ratings of cutting accuracy, scissor prehension, scissor loop position, cutting strategy and paper strategy were made based on the rating guide developed by Springeld (1998). Inter-rater agreement for scissor skill observations ranged from 88% to 100% (M 96%).

3. Results The study ndings will subsequently be described according to the activities and participation levels of the MAOM. 3.1. Functional perceptual motor skills (activities level) 3.1.1. Visual motor integration On the VMI, these children achieved standard scores ranging between 74 and 120 (M 96:15, SD 10.69) indicating that generally their performance fell within the average range (standard score 85115). Only two children (10%) had VMI standard scores less than 85. 3.1.2. Fine motor skills (PDMS) The PDMS ne motor scale has four subtests (hand use, grasp, eye hand coordination, manual dexterity) and a total ne motor score. When scoring the hand use and grasp subtests, some scoring modications were made. Ceiling eects are seen

S. Rodger et al. / Human Movement Science 22 (2003) 461478 Table 1 PDMS scores for participants in all ne motor subtests and total ne motor score (n 20) DMQa Range Meanb Standard deviation
a b

469

Hand use 69135 118.5 29.32

Grasp 69135 115.45 30.62

Eye hand 69135 91.5 21.12

Manual dexterity 74122 95.1 13.17

Total ne motor score 65109 85.65 12.23

DMQ Developmental Motor Quotient. Normative Mean DMQ 100, SD 15.

in subtests of grasping at 42 months and hand use at 36 months (Folio & Fewell, 1983). In addition, some children in this sample were older than the age range for the test (83 months), hence scores for these children were calculated based on the 7283 months norms. This may overestimate their ability, as there were no items to test ne motor skills of children above 7 years (84 months). For hand use, children with DCD had scores ranging from DMQ 69135 (M 118:5, SD 29.32). Extrapolated scores were entered for 19 children over 5 years of age, DMQ 135, 99th percentile if they achieved a raw score of 52 consistent with the norms for the oldest age range available (59 months). Sixteen children (80%) had DMQs of 135. Notably, four children (20%) older than 60 months did not show consistent hand preference in block building, resulting in a DMQ of 69. For the grasp subtest, standard scores were unable to be calculated for children over 72 months as all children in the standardization sample obtained complete scores, as the grasping skills tested had fully developed by this age. Percentiles and Developmental Motor Quotients (DMQs) were extrapolated for the ve children over 72 months, such that they obtained a maximum score of 44 for grasp, DMQ 135, 99th percentile. Childrens DMQ scores ranged from 69 to 135 (M 115:45, SD 30.62). The only grasp item above 42 months is holding a marker with tripod nger position. While most children achieved this, ve children were unable to hold the marker in this way, failing this item and resulting in a DMQ of 69, second percentile. In terms of eyehand coordination the group range for DMQs was 69135 (M 91:5, SD 21.21). For manual dexterity the range of DMQs was 74122 (M 95:1, SD 13.17). Both these DMQ means fall within the average range. However, for the ne motor total score, the DMQ range was 65109 (M 85:65, SD 12.23). This group mean score is borderline, placing the group at the lower end of the average range. These ndings are summarized in Table 1. 3.1.3. Writing In terms of hand preference, 71% of children showed right hand preference and 29% demonstrated left hand preference during writing tasks. When observing their nondominant hand use, 87% stabilized the page consistently during writing, while 13% stabilized the page inconsistently. Seventy nine percent of the sample, showed consistent patterns of pencil prehension while copying forms from the VMI and writing their name. For the 21% who demonstrated inconsistent prehension, these children used less

470

S. Rodger et al. / Human Movement Science 22 (2003) 461478

Table 2 Results (in percentages) of criteria observed during writing tasks (name writing, VMI) (n 20) Criteria observed Left hand preferred for writing Right hand preferred for writing Non-dominant hand consistent Non-dominant hand inconsistent Consistent prehension VMI and handwriting Inconsistent prehension VMI and writing Consistent hand pencil and scissors Inconsistent hand pencil and scissors Table 3 Prehension patterns utilized by children with DCD in this study (n 20) Prehension patterns (Schneck & Henderson, 1990) Dynamic tripod Static tripod Crossed thumb Four ngered Other Percentage (%) 19 19 19 12 31 Percentages (%) 29 71 87 13 79 21 88 12

mature patterns whilst copying shapes during the VMI than they did while writing their name, for example, a static tripod during name writing may have reverted to a four-ngered grasp during the VMI. There were insucient children over 7 years in this pilot study to summarize their results on tests of handwriting legibility or speed. Eighty eight percent of children consistently preferred the same hand for handwriting and scissor use, while 12% showed inconsistent preference (in all cases right hand for scissors and left hand for writing). An overview of these ndings can be found in Table 2. A range of pencil prehension patterns was observed. Based on the operational definitions described by Schneck and Henderson (1990), four main grasps were identied as described in Table 3. While 19% of children used a dynamic tripod grasp, 19% used a static tripod, 19% used a crossed thumb grasp, and 12% used a four-ngered grasp, another 31% used other grasps. The most common type of other grasp involved either three or four ngers on the pencil with ngers relatively extended. When utilizing these other grasps children tended to move their hand as a unit with limited nger movement (due to extended ngers). The thumb approximated opposition to the index nger. This is distinctly dierent to either of Schneck and Hendersons four ngered or tripod grasps, as the ngers were extended rather than exed. On occasion, the thumb was crossed over the pencil to the extended index. 3.1.4. Cutting with scissors Prehension of scissors, cutting strategy and paper strategy were coded from the videotapes of children cutting along a line, square and circle. Cutting accuracy was scored from 0 deviates from line > 10 mm to 4 cuts on the line. The group mean for children with DCD for cutting the line was 3.15 (SD 0.4) indicating

S. Rodger et al. / Human Movement Science 22 (2003) 461478

471

deviation of less than 2 mm from the line, 2.4 (SD 0.77) for square (deviates from line between 2 and 5 mm) and 1.81 (SD 0.86) for the circle (deviates between 5 and 10 mm from the line). Based on Springelds research the circle is the most dicult shape of the three to cut. Scissor prehension while cutting the line, square and circle was coded in terms of scissor loop position and prehension of the scissors based on classications developed by Springeld (1998). A mature scissor prehension pattern referred to thumb in one loop, index or index and middle nger in the other loop, remaining ngers exed or thumb in one loop, middle in the other loop with index stabilizing scissors, extended along blade of scissors. All other patterns were coded as immature. The mature scissor loop position was scissor loop resting between distal interphalangeal (DIP) joints and proximal interphalangeal (PIP) joints. Immature position was coded if scissor loops were proximal to PIP joint or distal to DIP joints or use of multiple positions. Across all three cutting tasks the majority of children with DCD used immature prehension patterns and scissor loop positions. For cutting the line, only 18% used mature scissor prehension, 31 % used mature prehension for cutting the square, and 23% used mature prehension for the circle. In terms of scissor loop position, only 12% of these children used mature loop position for cutting the line, 19% for the square and 35% for the circle. In contrast, between 73% and 83% of typically developing 251 year olds in Springelds study (1998) of 168 children used mature 2 patterns of scissor prehension during cutting of the line, square and circle, and 9195% used mature scissor loop positions during cutting these same shapes. Cutting strategy refers to how the child approaches the task (whether they can cut out the shape in one continuous cutting action) and where they choose to start cutting from (starting position). Between 41% and 53% of the children in this study used immature strategies when approaching the task, that is, they were unable to cut out the shape in one continuous action. Between 65% and 69% of children used immature or incorrect starting positions, resulting in their inability to complete the task in one continuous cutting action, requiring snipping, cutting to the edge of the paper, restarting, completing the task in several smaller steps. Hence they had poor strategies in terms of where to start and how to complete the task. Paper strategy refers to paper prehension or how the child holds the paper he/she is cutting and prehension relative to the blade refers to where the child holds the paper in relation to the blade of the scissors. Because cutting is a bimanual task involving tool use the actions of the non-preferred hand in manipulating the paper are important for ensuring skilled performance. Children with DCD used more immature patterns for more complex cutting tasks (41% immature strategies for line, 20% for square and 53% for circle). Mature paper prehension refers to thumb of non-dominant hand on top of the paper with other ngers exed under the paper. In terms of prehension relative to the blade they showed more immature than mature patterns (59% immature for line, 75% for square and 75% immature patterns for circle). Immature patterns involved paper held in front of the blade, behind the blade, crossed over scissors, or multiple positions. These immature patterns of holding the paper in relation to the scissors make it dicult to cut accurately and smoothly in a coordinated fashion. Cutting ndings are summarized in Table 4.

472

S. Rodger et al. / Human Movement Science 22 (2003) 461478

Table 4 Observations during cutting tasks (cutting accuracy, scissor prehension, loop position, cutting and paper prehension and prehension relative to blade) Criteria observed Cutting line Cutting accuracy M 3:15a Scissor prehension pattern Mature 18% Immature 92% b Mature 31% Immature 69% Mature 23% Immature 77% Scissor loop position Mature 12% Immature 88% Mature 19% Immature 81% Mature 35% Immature 65% Cutting strategy Prehension relative to blade Mature 59% Paper prehension Mature 41%

SD 0.4 Cutting square M 2:4

Immature approach to task 53% Incorrect start position 65% Immature approach to task 50% Incorrect start position 69% Immature approach to task 41% Incorrect start position 65%

Immature 41% Mature 25%

Immature 59% Mature 50%

SD 0.77 Cutting circle M 1:81

Immature 75% Mature 35%

Immature 50% Mature 47%

SD 0.86

Immature 65%

Immature 53%

a Accuracy refers to descriptions by Springeld (1998) based on scoring 0 deviates from line > 10 mm, 1 deviates from line between 5 and 10 mm, 2 deviates from line between 2 mm and 5 mm, 3 deviates < 2 mm from line, 4 cuts on line. b Immature refers to an immature pattern based on the descriptions reported by Springeld (1998) in her study of typically developing children between 21 and 5 years of age. Immature patterns are briey 2 described in the text.

3.1.5. Participation level 3.1.5.1. Perceived competence and social acceptance (Harter & Pike, 1983). PCSA addresses the childs self report of their cognitive competence (academic competence with numbers, reading, writing words, spelling), physical competence with gross motor skills (swinging, climbing, ball skills, skipping, running) as well as peer and maternal acceptance (friends to play with, mother talking with them). Scores range from 1(least competent) to 4 (most competent or accepted). The higher the score the more competent the childs self-perception. While there was a range in scores, mean scores for the DCD group indicate the children in this pilot study rated themselves towards the more competent end of the spectrum. Mean scores ranged from 3.04 for peer acceptance to 3.52 for cognitive competence, which are representative of positive self-perceptions. Table 5 provides a breakdown of the ranges, means, and standard deviations of scores in each subscale. 3.1.5.2. Pediatric evaluation of disability inventory. One of the applications of the PEDI is to detect the presence and extent of functional decits or delays and the content area of the delay. The functional skills scales measure the childs capabilities in terms of functional activities in self-care, mobility and social activities. The children in this study performed within the average range on mobility (M 54:71,

S. Rodger et al. / Human Movement Science 22 (2003) 461478

473

Table 5 Means and ranges for perceived competence and social acceptance (Harter & Pike, 1983) for study participants (n 20) Cognitive competence Range Mean Standard deviation 1.84.0 3.52 0.56 Physical competence 2.04.0 3.25 0.59 Peer acceptance 1.74.0 3.04 0.73 Maternal acceptance 2.174.0 3.09 0.46

Table 6 Functional performance on the PEDI (Haley et al., 1992) for study participants (n 20) Standard scorea Functional self care 13.655.3 38.01 12.19 Functional mobility 44.760.7 54.71 3.99 Functional social 33.291 49.98 16.62 Caregiver assistance self care 36.168.9 52.74 9.88 Caregiver assistance mobility 41.959.2 55.45 4.99 Caregiver assistance social 39.474.1 56.10 12.2

Range Mean Standard deviation

a

M 50, SD 10.

SD 3.99) and social (M 49:98, SD 16.62) subscales, however performance on the self-care subscale was greater than one standard deviation below the mean (M 38:01, SD 12.19), where the mean score is 50 and standard deviation is 10. This provides useful information regarding the extent of the self-care diculties experienced by young children with DCD. Specic items such as putting toothpaste on brush, using hairbrush for tangles, nose care and shoelace tying were dicult for these children. Level of caregiver assistance needed to accomplish major functional activities was also measured by parent report. The caregiver assistance scores for this group fell within the average range, indicating that overall they were not provided with any more physical assistance than typically developing children in order to accomplish their self-care, mobility or social occupations. While parents reported that their children were less capable with self-care tasks, they did not necessarily have to provide them with any more physical assistance than typically developing children on a daily basis. The PEDI was developed for assessing the functional performance of children with more signicant physical disabilities than children with DCD. As a result, items related to caregiver assistance described in the PEDI (e.g. assistance with transfers to enable self-care) are not likely to be as relevant for children with milder motor diculties. Hence the results for caregiver assistance fell within the average range. Table 6 provides ranges, means and standard deviations for these childrens functional performance on the PEDI.

4. Discussion The rst issue raised by this pilot study of younger children with DCD dand measurement relates to the importance of clinicians observations made during the

474

S. Rodger et al. / Human Movement Science 22 (2003) 461478

conduct of standardized assessments such as the VMI and during other functional tasks such as writing and cutting. For example, in relation to the VMI, while all but two of the children in this study performed within the average range on the VMI, clinical observations made during the VMI and while name writing, can be useful indicators of movement and performance issues such as consistency of hand use, prehension patterns during various writing tasks, non-preferred hand use, strategy use, time taken to complete the test etc. In this study, we attempted to categorize and quantify observations that clinicians typically make during standardized assessments, as well as during completion of daily functional tasks. For example, we classied childrens prehension patterns using the classication system used by Schneck and Henderson (1990) and coded strategy use during cutting as developed by Springeld (1998). This provides a standard and consistent basis from which to document these observations. These systems enabled us to also collect frequency data (converted to percentages) as a method of describing patterns and strategies used. Missiuna and Pollock (1995) investigated the VMI with 24 DCD children and 24 age-matched and 24 motor-matched controls. Consistent with the current study most of their children with DCD performed within the average range on the VMI. However, in Missiunas and Pollocks (1995) study children in the DCD group often had age equivalent scores up to three years below their age matched peers. While they were within the normal range for age, they were delayed relative to their age-matched classmates. They also found the time taken to complete the VMI was longer for children with DCD than for classmates. Given that the VMI is not a timed, children with DCD generally tend to perform as well as their peers. Children with DCD are likely to have more problems when working under pressure of time. The speedaccuracy trade o is well documented in children with motor impairments. Second, this study found that children with DCD had signicant diculties with strategy use evident from videotaped observations of their scissor skills. The importance of videotaped observation of performance of functional daily activities such as cutting and writing has been highlighted, as analysis of performance diculties can reveal where the performance of children with DCD breaks down during daily occupations. The comparison of these points of breakdown in relation to the performance of typically developing children, can assist practitioners in targeting appropriate interventions. These points of breakdown are often not as obvious during standardized assessment as they are during functional occupational or task performance. Third, this study also represents an investigation of younger children (preschool into early school years) than has typically been studied by DCD researchers. Hence the developmental trends in this age group must be considered. For example, with respect to prehension patterns, Schneck and Henderson (1990) considered the cross thumb, static tripod and four ngers grasp to be transitional grasps that may continue into the childs sixth year. These grasp patterns were seen in relatively even proportions in this group of 48 year olds, hence many of these transitional patterns were developmentally appropriate. It remains to be seen whether these prehension patterns are retained or mature. Missiuna and Pollock (1995) found the type of grasp in older children with DCD (M 7 years) was transitional or immature compared to age-matched and motor-matched controls. While awkward grip may not impede

S. Rodger et al. / Human Movement Science 22 (2003) 461478

475

handwriting it may lead to fatigue when larger amounts of writing are required and may slow down letter formation, contributing to performance problems. In relation to participation, the children in this study perceived themselves towards the more competent end of the scale on both physical and cognitive competence and peer and maternal acceptance. These ndings are in contrast to other studies of older children with DCD (69 years, M 7 years) that have found children with DCD to perceive themselves as being less competent and socially accepted than control children (Schoemaker & Kalverboer, 1994). There may be several reasons for this nding. First, younger children may be providing socially desirable responses in their self-report. This criticism plagues all self-report research and is therefore not unique to this younger age group (Sturgess, Rodger, & Ozanne, 2002). Second, younger children may not be as able to discriminate between potential responses as older children and therefore tend to answer towards the extreme ends of the scale. In this case, responses appear to have been primarily towards the positive end of the scale (identifying with the more competent child). Third, younger children may not yet have started comparing themselves with their peers (Schunk, 1989). It is thought that by 9 years of age, evaluation against others is a standard means of self-rating. It is well recognized that self-ecacy is a developmental process and that young children see that eort leads to outcomes (Schunk, 1989), whereas with age the concept of ability emerges. It seems possible that these children with DCD see themselves as trying hard at tasks, even when dicult for them. If they attribute eort to competence, they may perceive themselves as being competent due to their eorts. To date, research on perceived ecacy in children with DCD has focused on older children (above 7 years) and has utilized the Self Perception Prole for Children (Harter, 1985) for children in middle childhood. If in fact, children under 8 years with DCD still perceive themselves as being competent with various physical and cognitive tasks, this may be a prime time to intervene to enhance their motor skills and further bolster their self-ecacy, as it is well known that self-ecacy aects eort expenditure and persistence (Schunk, 1989). While the PEDI has not to date been used with children with DCD, it appears to have potential as a measure of functional impact of DCD. The results of this pilot study demonstrated decreased functional self-care skills in children with DCD compared to the standardization sample, whilst social and mobility function were within normal limits. Case-Smith (1995) highlighted the self-care scale of the PEDI was heavily based on items requiring ne motor skills such as dressing, grooming, and bathing. However, despite the intuitive link between ne motor skills and self-care, some studies have failed to nd strong relationships between ne motor skills as measured by the PDMS and the PEDI self-care scores (Case-Smith, 1995). Some reasons for this may include: Dierences between judgment based (parents perceptions) and observational evaluation (PDMS); dierences in the context of performance (e.g., in a time pressured home routine there may be limited opportunities for parents to enable practice and monitoring of skills), and cultural inuences such as parents perception of their caregiving role versus developing independence in their children. Further investigation of this relationship between ne motor scores (PDMS) and self-care scores (PEDI) would be a worthwhile direction for future research for

476

S. Rodger et al. / Human Movement Science 22 (2003) 461478

children with DCD, especially when larger numbers are recruited into the present study. In addition, triangulation of data obtained from standardized assessments such as the PEDI and PDMS with observations of academic and self-care tasks in ecologically valid contexts such as school and home would further inform our understanding of childrens participation (Burgman, 1998; Case-Smith, 1995). Shumway-Cook and Woollacott (1995) recommend assessment should be directed at all three levels of Nagis Model of Disablement and the ICIDH for clinical decision making, that is functional skills, strategies used to accomplish these skills, and underlying systems. If measurement for research purposes addresses all three levels, the possibility of investigating relationships between various levels such as ne motor skills (activities level) and functional performance (participation level) becomes possible. As no one assessment addresses all areas of skill, the clinician must determine the missing tasks and environmental contexts in order to identify the battery of assessments and observations needed in a comprehensive evaluation of motor and functional skills. The importance of systematic contextually relevant assessment of tasks and performance quality cannot be underestimated in both research and clinical protocols (Burgman, 1998). Dierent methodologies and tools as well as use of multiple sources of information (caregiver report, self-report, and observation) to those traditionally used in measurement in DCD (standardized assessment of the child) are required if we are to adequately address function and participation. Assessments based on tasks such as dynamic performance analysis (Polatajko, Mandich, & Martini, 2000) and Performance Quality Rating Scale (Miller, Polatajko, Missiuna, Mandich, & Macnab, 2001) are examples of contextually relevant assessments of tasks and performance quality, that have potential to increase our understanding of performance quality and breakdown in children with DCD. Time use logs, daily diaries of participation in physical activity, leisure and social and occupational roles, and personal projects (Little, 1998) also have promise as measures of childrens participation in various contexts. This pilot study has provided preliminary support for the MAOM as a comprehensive framework to guide measurement selection for young children with DCD. While beyond the scope of the current pilot data set, analysis of an expanded data set has the potential for providing additional evidence as to the utility of this model. We are undertaking this task at present. To date, the measures used at the activities and participation level have been described and preliminary results presented, indicating the pervasive impact of DCD on childrens performance of functional perceptual motor tasks and their participation in daily occupations. Some issues in relation to measuring self-ecacy in children under 8 years of age have also been addressed. The utility of the MAOM to guide measurement and the specic utility of videotaped observations of functional tasks and the PEDI, have also gained preliminary support in this current study. References
American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: American Psychiatric Association. Beery, K. E., & Buktenica, N. A. (1997). Developmental test of visual motor integration. Austin, TX: Proed.

S. Rodger et al. / Human Movement Science 22 (2003) 461478

477

Burgman, I. (1998). Functional measures and pediatric assessment. Physical and Occupational Therapy in Pediatrics, 18(3/4), 109121. Cantell, M. H., Smyth, M. M., & Ahonen, T. P. (1994). Clumsiness in adolescence: Educational, motor and social outcomes of motor delay detected at 5 years. Adapted Physical Activity Quarterly, 11, 115129. Case-Smith, J. (1995). The relationships among sensorimotor components, ne motor skill, and functional performance in preschool children. American Journal of Occupational Therapy, 49, 645655. Coster, W., & Haley, S. (1992). Conceptualization and measurement of disablement in infants and young children. Infants and Young Children, 4, 1122. Crawford, S. G., Wilson, B. N., & Dewey, D. (2001). Identifying developmental coordination disorder: Consistency between tests. Physical and Occupational Therapy in Pediatrics, 20, 2950. Dewey, D., & Wilson, B. N. (2001). Developmental coordination disorder: What is it? Physical and Occupational Therapy in Pediatrics, 20, 527. Folio, M. R., & Fewell, R. R. (1983). Peabody Developmental Motor Scales and Activity Cards. Hingham, MA: DLM Teaching Resources. Haley, S. M., Coster, W. J., Ludlow, L. H., Haltiwanger, J. T., & Andrellow, P. J. (1992). Pediatric evaluation of disability inventory. Boston, MA: New England Medical Centre. Harter, S. (1985). The self-perception prole for children. Denver, CO: University of Denver. Harter, S., & Pike, R. (1983). The pictorial scale of perceived competence and social acceptance for young children. Denver, CO: University of Denver. Harter, S., & Pike, R. (1984). The pictorial scale of perceived competence and social acceptance for young children. Child Development, 55, 19691982. Hay, J., & Missiuna, C. (1998). Motor prociency in children reporting low levels of participation in physical activity. Canadian Journal of Occupational Therapy, 65, 6471. Henderson, S. E., & Hall, D. B. (1982). Concomitants of clumsiness in young school children. Developmental & Medicine Child Neurology, 24, 448460. Henderson, S. E., & Sugden, D. A. (1992). Movement Assessment Battery for Children (Manual). Kent, UK: Psychological Corporation. Kaplan, B. J., Wilson, B. N., Dewey, D. M., & Crawford, S. J. (1998). DCD may not be a discrete disorder. Human Movement Science, 17, 471490. Klein, S., & Magill-Evans, J. (1998). Reliability of perceived competence measures for young school aged children. Canadian Journal of Occupational Therapy, 65, 293298. Larsen, S. C., & Hammill, D. D. (1989). Test of Legible Handwriting. Austin, TX: Proed. Little, B. R. (1998). Personal project pursuit. Dimensions and dynamics of personal meaning. In P. T. P. Wang & P. S. Fry (Eds.), The humans quest for meaning: A handbook of psychological research and clinical application (pp. 193212). Mahwah, NJ: Lawrence Erlbaum Assoc. Losse, A., Henderson, S. A., Elliman, D., Hall, D., Knight, E., & Jongmans, M. (1991). Clumsiness in children: Do they outgrow it? A 10-year follow up study. Developmental Medicine & Child Neurology, 33, 5568. Michael, W. B. (1990). Review of the pictorial scale of perceived competence and social acceptance for young children. In J. J. Krame, & J. C. Conoley (Eds.), The supplement to the tenth mental measurement yearbook (p. 59). Lincoln, NE: University of Nebraska Press. Miller, L. T., Polatajko, H. J., Missiuna, C., Mandich, A. D., & Macnab, J. J. (2001). A pilot trial of a cognitive treatment for children with Developmental Coordination Disorder. Human Movement Science, 20, 183210. Missiuna, C. (1998). Development of All About Me, a scale that measures childrens perceived motor competence. Occupational Therapy Journal of Research, 18, 85108. Missiuna, C., & Pollock, N. (1995). Beyond the norms: Need for multiple sources of data in the assessment of children. Physical and Occupational Therapy in Pediatrics, 15, 5771. Nagi, S. Z. (1991). Disability concepts revisited: Implications for prevention. In A. M. Pope & A. R. Tarlov (Eds.), Disability in America. Washington, DC: Academy Press. Ozanne, A., Woodyatt, G., Watter, P., Rodger, S., & Ziviani, J. (2002). Communication disorders in children with DCD and implications for management. In Proceedings of DCD V Conference, 5th Biennial Workshop on Children with DCD, Ban, Alberta, Canada, May 1416.

478

S. Rodger et al. / Human Movement Science 22 (2003) 461478

Polatajko, H., Mandich, A., & Martini, R. (2000). Dynamic performance analysis: A framework for understanding occupational performance. American Journal of Occupational Therapy, 54, 6572. Schneck, C. M., & Henderson, A. (1990). Descriptive analysis of the developmental progression of grip position for pencil and crayon control in nondysfunctional children. American Journal of Occupational Therapy, 44, 893900. Schoemaker, M. M., & Kalverboer, A. F. (1994). Social and aective problems of children who are clumsy: How early do they begin? Adapted Physical Activity Quarterly, 11, 130140. Schunk, D. H. (1989). Self-ecacy and cognitive achievement: Implications for students with learning problems. Journal of Learning Disabilities, 22, 1422. Shumway-Cook, A., & Woollacott, M. H. (1995). Motor control: Theory and practical application. Baltimore, MD: Williams & Wilkins. Smyth, M. M., & Anderson, H. I. (2000). Coping with clumsiness in the school playground: Social and physical play in children with coordination impairments. British Journal of Developmental Psychology, 18, 389413. Springeld, E. (1998). The performance of scissor skills in two to ve year old children. Master of Occupational Therapy Thesis. Brisbane: The University of Queensland. Sturgess, J., Rodger, S., & Ozanne, A. (2002). A review of the use of self-report assessment with young children. British Journal of Occupational Therapy, 65, 108116. Wallen, M., Bonney, M., & Lennox, L. (1996). The Handwriting Speed Test. Adelaide: Helios Art Book & Co. Watkinson, E. J., Causgrove Dunn, J., Cavaliere, N., Calzonetti, K., Wilhelm, L., & Dwyer, S. (2001). Engagement in playground activities as a criteria for diagnosing developmental coordination disorder. Adapted Physical Activity Quarterly, 18, 1834. Watter, P. (1996). Physiotherapy management Minor coordination dysfunction. In Y. Burns & J. Macdonald (Eds.), Physiotherapy and the growing child (pp. 415432). London: W.B. Saunders. Watter, P., Ozanne, A., & Rodger, S. (2002). Physiotherapy ndings in children rated normal on Movement-ABC. In V11th International Conference of the Australian Physiotherapy Association, Sydney, May 2528. Wilson, B. N., Dewey, D., & Campbell, A. (1998). Developmental Coordination Disorder Questionnaire (DCD-Q). University of Calgary. World Health Organisation (1998). ICIDH Environmental Factors International Network. Vol. 9 (2/3) (pp. 166). Ziviani, J., & Wright, S. (1995). Pediatric Evaluation of Disability Inventory Review and application. New Zealand Journal of Occupational Therapy, 46, 1518.