RESNA 26th International Annual Confence

Technology & Disability: Research, Design, Practice & Policy

June 19 to June 23, 2003
Atlanta, Georgia

Constraint Induced Movement Therapy Device

Kyle Smith and Lynn Wang
Duke University


A novel device has been developed for constraint-induced movement therapy (CIMT) in children with hemiplegia. The device immobilizes the fingers, hand, wrist, and elbow of the child's unaffected arm, forcing use of the affected arm. The device has several advantages over most current devices, which are made of standard casting material. The device is relatively inexpensive, comfortable, attractive, and easily constructed from commercially available products.


Each year, many thousands of people experience cerebrovascular accidents resulting in hemiplegia, the weakening or paralysis of one side of the body (1). Fortunately, the brain has the remarkable ability to reorganize its neurons, passing functions from parts of the brain damaged by stroke to healthy regions, allowing people with hemiplegia to regain functioning of their affected limbs (2). However, since this neural reorganization will only occur in response to external stimuli, the patient must use the affected arm to regain function.

The goal of Constraint-Induced Movement Therapy (CIMT) is to force patients with hemiplegia to use their affected arms by restraining their healthy arms. Several studies have shown significant improvement in patients who undergo CIMT (3). Most studies have involved restraining the unaffected arm using a standard cast from the upper arm to the fingertips for a few weeks in an outpatient or intensive inpatient therapy. Most therapists fabricate their own custom casts for the therapy; no standard device exists for restraint of the unaffected arm.

Problem Statement

Currently, therapists at Lenox Baker Children's Hospital (Durham, NC) use a Delta-Cast that patients undergoing CIMT wear for 3 weeks, 24 hours a day, in an outpatient intensive treatment. The Delta-Cast is a polyurethane resin cast with Velcro straps that is formed around the patient's elbow, forearm, and hand. This device has two major shortcomings. First, the Delta-Casts already in use do not cover the fingertips. Since the fingers remain exposed and the unaffected arm was not supervised, patients succeeded in deriving functionality from their casted arm. Thus, our CIMT device must offer the support of the Delta-Cast with additional restraints to prevent all use of the unaffected arm removal by the patient. The second shortcoming of the Delta-Cast is that it must be custom made for each patient. Therapists must be trained to make casts, and the fabrication process is tedious and time-consuming. The high cost of casting materials and the time involved in manufacturing make the cast expensive. Therefore, our CIMT device must be easily adaptable to multiple patients and must be more cost-effective than the current cast. The device must also be professional in appearance for parents, and both aesthetically pleasing and comfortable for the child to ensure cooperation in therapy participation.


Providing Lenox Baker with a device that is both adaptable to multiple patients and easily reproducible will ensure the ability to accommodate a full range of patient sizes. Such a design will also be both time and cost efficient. Selecting existing commercial products for the main frame of the device will allow for easy reproduction of the device, and a professional appearance that will please parents. Using tried and proven products will ensure a comfortable fit for each patient. 

Design and Development

Figure 1 Elbow Brace
Figure 1 shows the elbow wrap modeled on a possible patient. The elbow wrap, manufactured by Benik, is made entirely of Velcro  sensitive neoprene and consists of two hinge pockets on either side of the arm with an aluminum 135  flexion/extension restraint.  The elbow brace is 9" in length and wraps around the arm at the elbow joint, leaving the elbow exposed.   The fit of the brace is adjusted with Velcro tabs and can be further tightened with removable Velcro  straps that secure onto the brace with Velcro and double-back through a D-ring.
We met with our supervisors, a pediatrician and two occupational therapists, regularly to correctly identify and assess the needs and rationale for the device. Throughout the duration of the project we also met with and interviewed several candidates for CIMT treatment to better understand the needs of the patients wearing the devices and their parents' concerns. Meeting these candidates also allowed us to ascertain the level of restraint and support necessary to eliminate the functionality of the unaffected arm and judge the potential impact of CIMT.

Patient and parent acceptance was a high priority for the device. Our final design therefore comprised three components to maximize ease of use for parents and to offer a sleeker appearance. The three final components of the CIMT device are: (1) elbow brace with elbow flexion/extension restraint, commercially purchased and modified with a small aluminum stay to fix its angle at 135° (Fig. 1); (2) commercially purchased hand splint made of a malleable aluminum frame with a removable lining, and added Velcro supination/pronation straps (Fig. 2); and (3) Stretchable sleeves to enclose the entire device (Fig. 3). Each component is durable and easy to put on; straps used with the hand splint are color coded to ensure correct strap direction.

Figure 2 Hand Splint
Figure 2 shows the elbow wrap and the hand splint modeled on a potential patient.  The hand splint, manufactured by Sammons-Preston, is made of a malleable aluminum frame that is covered with a removable soft blue lining.  The straps and removable finger separators are also made of this blue lining.  The splint holds the hand in a neutral arced position with a thumb strap and two straps that cover the fingers and the knuckles of the hand.  Two additional straps wrap around the forearm and the base of the hand splint.  These straps have been extended with 14 Velcro straps partially lined with 4 ¾  X 2  and 2 7/8  X 1 5/8  neoprene rectangles to serve as supination/pronation restraint straps.  These straps wrap around the arm in opposing directions.
The other main priority of the project was to create a universal design for the device. To balance appearance with universality, we eventually moved away from one cumbersome design with telescoping supports to multiple devices in a full range of sizes and the ability to mix and match these sizes. We also evolved from fabricating our own hand restraint component with a fixed frame, which provided for minimal adaptability, to incorporating a commercial hand splint with a malleable aluminum frame to increase the range of patients fit and to provide the ability to order additional sizes. Due to the orientation of the straps on the commercial hand splint, the supination/pronation restraint and the hand restraint could be merged into one unit. The supination/pronation straps are 2" wide Velcro extensions of the hand splint straps that wrap around the forearm in opposite directions and attach to the Velcro-sensitive elbow brace. Neoplush was placed between the Velcro straps and forearm to prevent abrasion. Maximizing the use of adaptable commercial pieces minimizes the time a therapist must spend fitting the device to the patient, making the device time and cost effective.

Figure 3. Sleeve enclosure
Figure 3 shows the completely assembled device, modeled on a potential patient. The device is completely enclosed with a Lycra  or knit sleeve and is secured at the upper arm with a drawstring cord.
The sleeve of the device was adapted from arm warmers used by bicyclists in cool weather. One of the sleeves developed is made of CoolMax, which is available in a range of sizes and colors; children may also decorate their sleeve because of its low cost. Cords were threaded through the seams on both ends to act as drawstrings. The lower drawstring was permanently cinched and tied to prevent its opening. A cord lock was added to the upper drawstring to hold the drawstring tight and prevent the sleeve from slipping over the elbow brace and the patient from removing the device. A knot can be tied for further restraint as necessary.


Throughout the design process, author Wang wore and evaluated the comfort and appearance of the fabricated devices. Potential problems were identified and solutions were discussed and implemented. One concern that arose from these experiments was the temperature of the device in use. Thus, we performed temperature measurements in the device using three different sleeve materials. Based upon the results, we provided a lighter, more breathable sleeve alternative made out of CoolMax rather than a tight weave of Lycra and cotton.

Cost-Efficiency Comparisons


Current Device (Delta Cast)

CIMT Device

Therapist Time

1.5 hours @ $120/hr = $180




Hand splint= $50

Elbow component=$60





Since minimizing the cost to the patient and therapist time were main goals for the creation of this device, we evaluated the cost and time needed to use the device throughout the design process. The final design meets these objectives and increases return on investment with each new CIMT patient as shown in table to the right.

Discussion and Conclusions

The main advantage of our CIMT device is that it meets its functional objectives while being cost and time efficient, easily reproducible, adaptable to a wide range of potential patients, and easy for parents and patients to use. Therapists do not need to be specially trained to use this device, and customizations made to the device will occur during patient fitting and patient/parent education, and will not involve additional construction. The device conforms to the needs of the therapy. The sleeve can be removed easily by the therapist or parent to monitor the skin, but can be difficult for the child to remove by knotting the drawstring enclosure if the child becomes tempted to remove the device. The only foreseeable disadvantage is if the chosen commercial products become unavailable. However, similar commercial products could be easily substituted into the device. The final device meets all objectives, and will provide a re-useable and cost-effective therapy tool to help patients with hemiplegia to regain motor skills.

We demonstrated the device for our professor, classmates, supervisors, and others familiar with occupational therapy and rehabilitation engineering on several occasions. We discussed the changes they suggested and altered our device when necessary. While the device has not been used in therapy because an appropriate client has not yet been identified, we believe that any problems discovered in clinical use of the device will be easy to correct due to the adaptability of the device.


  1. Taub E, Uswatte G, Pidikiti R. "Constraint-Induced Movement Therapy: A New Family of Techniques with Broad Application to Physical Rehabilitation--A Clinical Overview". Journal of Rehabilitation Research and Development. 36(3): 237-51, 1999 Jul.
  2. Gourley, M. "Regaining upper-extremity function through constraint-induced movement therapy". OT Practice. 7-9, 2002 Feb.
  3. Willis JK, Morello A, Davie A, Rice JC, Bennett JT. "Forced use treatment of childhood hemiparesis". Pediatrics. 110:94-6, 2002 Jul.


We would like to thank Dr. Larry Bohs, our professor, for providing us with this design opportunity and aiding us in the design process. Mark Palmeri, our Teaching Assistant, aided in providing feedback and sourcing information. We would also like to thank our supervisors, Jodi Petry, Erin Eadry, and Dr. Gordon Worley, who advised us on device characteristics. 

Kyle Smith
14510 Lightning Ridge Run
Fort Wayne, IN 46814

Lynn Wang
1341 Cabrillo Ave.
Burlingame, CA 94010

RESNA Conference Logo