Modified constraint-induced therapy for children with hemiplegic cerebral palsy: a randomized trial


Dr Margaret Wallen, Occupational Therapy Department, The Children’s Hospital at Westmead, Locked Bag 4001, Westmead, NSW 2145, Australia. E-mail:


Aim  Conventional constraint-based therapies are intensive and demanding to implement, particularly for children. Modified forms of constraint-based therapies that are family-centred may be more acceptable and feasible for families of children with cerebral palsy (CP)-but require rigorous evaluation using randomized trials. The aim of this study was to determine the effects of modified constraint-induced therapy compared with intensive occupational therapy on activities of daily living and upper limb outcomes in children with hemiplegic CP.

Method  In this assessor-blinded pragmatic randomized trial, 50 children (27 males, 23 females; age range 19mo–7y 10mo) with hemiplegic CP were randomized using a concealed allocation procedure to one of two 8-week interventions: intensive occupational therapy (n=25), or modified constraint-induced therapy (n=25). Manual Ability Classification System (MACS) levels of the participants were, level I n=2, II n=37, III n=8, and level IV n=1; Gross Motor Function Classification System (GMFCS) levels were, level I n=33, level II n=15, and level III n=1. Participants were recruited through three specialist CP centres in Australia and randomized between January 2008 and April 2010. Children randomized to modified constraint-induced therapy wore a mitt on the unaffected hand for 2 hours each day, during which time the children participated in targeted therapy. The primary outcome was the Canadian Occupational Performance Measure (COPM – measured on a 10-point scale) at completion of therapy. Other outcome measures were Goal Attainment Scaling, Assisting Hand Assessment, Pediatric Motor Activity Log, Modified Ashworth Scale, Modified Tardieu Scale, and a parent questionnaire. Assessments were carried out at 10 weeks and 6 months following randomization.

Results  All participants were included in the analysis. Between-group differences for all outcomes were neither clinically important nor statistically significant. The mean difference in COPM was 0.3 (95% confidence interval [CI] −0.8 to 1.4; p=0.61) and mean difference in COPM satisfaction was 0.1 (95% CI −1.1 to 1.2; p=0.90). Minor adverse events were reported by five of the 25 participants in the modified constraint-induced therapy group and by one of the 25 in the intensive occupational therapy group. All adverse events were related to participants’ lack of acceptance of therapy.

Interpretation  Modified constraint-induced therapy is no more effective than intensive occupational therapy for improving completion of activities of daily living or upper limb function in children with hemiplegic CP.