To assess the effectiveness of motor skill interventions in children with developmental coordination disorder (DCD) or equivalent conditions, and to answer the following questions.

Which, if any, of the three theoretical approaches to motor skill interventions is supported by the evidence?

How do the age of the participants, research design, intervention setting, and intervention duration affect motor outcomes?

ERIC, CINAHL, MEDLINE, PsycINFO and SPORTDiscus were searched from 1970 to 1996.

Study designs of evaluations included in the review

Experimental studies with at least one control group or single-subject design studies were eligible. The included studies were: randomised controlled trials (RCTs), including RCTs in which the participants were matched on key variables; controlled studies in which the participants were matched on key variables but not randomly assigned; and studies of single-subject design.

Specific interventions included in the review

Motor skill interventions were eligible. The following theoretical approaches to interventions were included in the review:

the general ability (GA) approach, which consisted mainly of facilitation of balance and other physical abilities and training specific perceptual and motor tasks;

the sensory integration (SI) approach, which involved activities that consisted of full body movements and training in specific perceptual and motor skills;

the specific skills (SS) approach, which included task-specific instruction; and

combinations of the theoretical approaches, i.e. GA plus SI, GA plus SS and SI plus SS.

The intervention settings were small group performed by a teacher, home programme performed by parents, and one-to-one performed by a physical or occupational therapist. The duration of the interventions, where stated, ranged from 2 weeks to 8 months.

Participants included in the review

Children identified as having DCD or motor problems consistent with DCD were eligible. The terminology for the participants included DCD, clumsy or motor impaired, and sensory integrative dysfunction. The individual studies used combinations of the following criteria to select the participants: average intelligence quotient; normal neurology; normal vision and hearing; and no academic delay. Children aged 3 to 13 years were included.

Outcomes assessed in the review

The inclusion criteria were not defined in terms of the outcomes. The included studies used the following measures to assess outcomes: STOTT, Cratty GMT, balance task, Arnheim-Sinclair Battery, Bender Purdue Reflex Test, Hockey slap shot, Neurodevelopmental test, TOMI, Perceptual-Motor Abilities Test, BOTMP, SCSIT, SCPRNT, VMI, observation, distance throw, distance hop, target kick, volley ball bounce, Peabody Developmental Motor Test, Vineland Adaptive Behaviour Scale, letter writing, KST, PEST, and fine motor task. The references to 19 studies, which described these measures, were provided.

How were decisions on the relevance of primary studies made?

The authors do not state how the papers were selected for the review, or how many of the reviewers performed the selection.

No formal assessment of validity was undertaken.

The authors do not state how the data were extracted for the review, or how many of the reviewers performed the data extraction.

The following data were extracted: authors; terminology used for participants; participant criteria; theoretical approach; intervention characteristics; dependent measures used; age and gender of the children; research design; statistically-significant differences in outcome; and whether the means and standard deviations were reported. The effect size (ES) was calculated for studies presenting adequate data, using the methods described by Rosenthal, Becker, and Hedges and Olkin (see Other Publications of Related Interest nos.1-3), such that a positive ES score indicated a positive intervention effect. Each ES was multiplied by a correction factor (see Other Publications of Related Interest no.4). The ESs were calculated when a trained group was compared with an untrained group or, if no control existed, for a trained group compared with a control group receiving another type of intervention. For each study, a mean ES was calculated by taking the means of all ES values of dependent variables.

How were the studies combined?

Data from those studies that reported means and standard deviations for each treatment group were combined in a meta-analysis. The overall, mean, main ESs were calculated using the means of the ESs from all of the studies.

How were differences between studies investigated?

The influence of study design, age and intervention (approach, setting, duration and frequency) on the results was explored by calculating the ES for the following subgroups.

For study design: participants randomly assigned to treatment group, participants matched on key variables and then randomly assigned, participants matched on key variables but not randomly assigned, and single-subject design.

For age of the participants: 3 to 5 years and 6 to 13 years;

For theoretical approach of the intervention: GA, SI, SS, or a combination of theoretical approaches.

For intervention settings: small group, home programme and one-to-one.

For intervention duration: less than 3 months and 3 months or more.

For frequency of the intervention: less than 3 times per week and 3 to 5 times per week.

Twenty-one studies (961 children) were included in the review. These included 15 RCTs of which 4 matched the participants on key variables before randomisation, 4 controlled studies in which the participants were matched on key variables but not randomly assigned, and 2 studies of single-subject design. Thirteen studies were included in the meta-analysis.

Eighteen mean main ESs were calculated from 13 studies.

Theoretical approach of the intervention.

Motor skill interventions using the SS approach were the most effective. The ES was 0.71 for GA (9 means of experiments), 0.21 for SI (4 means of experiments), 1.46 for SS (3 means of experiments), 0.20 for GA plus SS (1 experiment), and 0.07 for GA plus SI (1 experiment).

Age.

Interventions involving children aged 6 to 13 years gave higher ESs than those involving children aged 3 to 5 years. The ES was 0.14 for children aged 3 to 5 (2 means of experiments), and 0.77 for those aged 6 to 16 (16 means of experiments).

Research design.

All types of assignment of participants seemed to support the intervention. The ES was 0.56 for RCTs (13 means of experiments), 1.15 for matched RCTs (2 means of experiments), and 0.99 for matched but not RCTs (3 means of experiments).

Intervention setting.

One-to-one programmes gave lower ESs than either of the other two settings. The SS approach was not used in any of the one-to-one programmes. The ES was 0.96 for small group (5 means of experiments), 1.41 for home programme (2 means of experiments), and 0.45 for one-to-one (11 means of experiments, including 9 RCTs).

The length and frequency of the intervention were not related to ES. The ESs were similar for interventions lasting less than 3 months (ES=0.72) and 3 months or more (ES=0.69), and were also similar for interventions carried out less than 3 times per week (ES=0.60) and 3 to 5 times per week (ES=0.86).

The findings indicated that the motor skill intervention is most effective when applied to children with DCD over 5 years of age, using the SS approach with an intervention frequency of at least 3 to 5 times per week. No clear findings were found with respect to other variables.

The aims of the review were stated and the inclusion criteria were defined in terms of the intervention, participants and study design. Five databases were searched, but the key search terms were not reported and it was unclear whether any language restrictions were applied. In addition, the methods used to select the studies were not described. The lack of an attempt to locate unpublished material raises the possibility of publication bias. The validity of the primary studies was not formally assessed.

Relevant information on the included studies was presented in tabular format, but the methods used to extract and code the data were not described. The data were combined in a meta-analysis where adequate information was reported, and the influence of various factors on the results was explored. However, statistical heterogeneity was not assessed and it might have been inappropriate to average the ESs of the many and varied measures used to assess the outcomes in the individual studies. The number of participants providing data for each overall mean ES was not reported and neither were any confidence intervals of the ESs; these omissions hinder an assessment of the precision of the ESs. Interactions between independent variables were not considered. In addition, treatment approaches were not directly compared in the primary studies. Hence, any conclusions about the relative effectiveness of different approaches are tentative rather than definitive.

The authors' conclusions should be interpreted with caution.

Practice: The authors state that the motor skill intervention is most effective when applied to children with DCD over 5 years of age, using the SS approach with an intervention frequency of at least 3 to 5 times per week.

Research: The authors state that there is a need for future research on well-defined subgroups of children with DCD.

Gillberg Foundation; Samaritan Foundation; University of Uppsala; the Vardal Foundation.

1. Rosenthal R. Parametric measures of effect sizes. In: Cooper HM, Hedges LV, editors. The handbook of research synthesis. New York: Russell Sage Foundation; 1994. p. 231-44. 2. Becker BJ. Synthesizing standardized mean-change measures. Br J Math Stat Psychol 1988;41:257-78. 3. Hedges LV, Olkin I. Statistical methods for meta-analysis. New York: Academic Press Inc.; 1985. 4. Thomas JR, Nelson JK. Research methods in physical activity. 3rd ed. Champaign (IL): Human Kinetics; 1996. 5. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th ed. Washington (DC): The Association; 1994.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.