Dance program for physical rehabilitation and participation in children with cerebral palsy

Objective: This pilot study aimed to examine a classical ballet program created for children with cerebral palsy (CP) as an emerging physical rehabilitation modality. The main program goals were to promote participation and to provide an artistic, physically therapeutic activity. Methods: The study was conducted in collaboration with a tertiary rehabilitation hospital, one outpatient physical therapy clinic, and one community center. As a pilot exploratory study, the research design included questionnaires to assess the participants' (children (n = 16), parents (n = 16), and therapists (n = 13)) perceptions on the therapeutic benefit of the dance program. A binomial statistical model was adopted for the analysis of the results. Results: Main results were that the children reported high enjoyment level (p < .0001) and desire for more classes (.0001); the parents reported perceived therapeutic benefit (p < .0001); and the therapists viewed the class as a positive adjunct to therapy (p < .0001). Conclusions: The main limitation of this work was the utilization of subjective outcome measures. However, this is the first step toward the development of objective measures of an intervention that, to our knowledge, has not been analyzed in the past. We conclude that the program has the potential of developing into an evidence based rehabilitation resource for children with CP.


Background
The prevalence of cerebralpalsy(CP) in children is estimated to be, on average, between 1.5 and 2.5 casesper 1000 live births (Paneth, Hong &Korzeniewski, 2006). Children with CP face challengesi nm ovementa nd posture control. Hypertonia andv arious combinations of motor impairmentss uch as weakness, reduced selective motor control, poor balance and discoordination are common in children with CP. These deficits contribute to impaired movement and posture that compromise adequate function and participation in social activities (van der Heide et al., 2004;Woollacott et al., 2005). Thus, techniques to enhance functional mobility are criticaltoimproving outcomes in pediatric CP. Movementability impacts learning through the child's exploration and manipulation of the environment.Inaddition, movement is the basisfor manyhuman experiences, such as community activities, play and cognitive development.The International Classification of Functions (ICF) of the WorldHealth Organization (WHO) provides auniversal starting point for ac omprehensive understanding of the human experiences of functioninga nd disability where the physical,social and environmental factorsbecome intertwined (Cieza &S tucki, 2008). According to the ICF,p articipation consists of taking part or being involved in everyday life activities and roles. Thus, participation in leisureactivities has emergeda sa ni mportant health outcome for children with disabilities. Leisure activities are typically those in which an individual freely choosest op articipate during free time because it is enjoyable. Benefitso fp articipation in leisure activities include fostering friendships and developing personal interestsa nd identity . In children with CP, impaired mobility leads to decreased participation in the community and reduced contact with peers in activities and play (Fauconnier et al., 2009;Imms, 2008;Michelsen et al., 2009;Parkes, McCullough &M adden, 2010;Pratt, Baker &G aebler-Spira, 2008;Shikako-Thomas,M ajnemer, Law&Lach, 2008). As ac onsequence,t he development of cognitive, motor and social skillsiscompromised (Bottcher, 2010;King et al., 2009). This decrease in participation correlates positivelyw ith increased gross motor impairments ( Parkes et al., 2010). Furthermore, it has been hypothesized that attentional and executived eficits present in children with CP may also contribute to the decrease in societal participation (Bottcher, Flachs &U ldall, 2010). The net effect is a negativefeedback loop on the development of the child as awhole. Indeed, participation intensityofchildren with CP in recreational, active physical and social activities tends to decrease through time (King et al., 2009). It has been noted, however, that participation in self-improvement activities and skill-based activities such as dance does not seem to decline throughtime (King et al., 2009). The need for activity-based therapies in CP has been stressed in the recent literature (Damiano, 2006), since children with CP have restricted access to skill-based movement programsthat combine therapeutic principles of movement rehabilitationw ith community involvement. We created and established a dance program based in classical ballet technique for children with CP to enhancet heir movement abilities, artistic expression, socialization and participation. The main objective of the dance program was to deliver physical rehabilitationt hrough the execution of artistic movement in agroup setting where the children participated by their own choice and perceived enjoyment. We considered the child's right to choose to participate essential,asithas been reported that motivation seems to be key for successful treatment in children with CP (Kwak, 2007;Majnemer et al., 2008;Morris, 2009). Additionally, by allowing the children to participate of their own choice we enhanced the leisure component of the dance class while involving the children in therapeutic movement activity.
Assisting the motor and cognitive development of the child while preventing secondary injury is the main goal of treatmenti nC P ( Sanger, 2008). Physical and occupational therapy are standardtreatment and the mainstay management for the motor impairments. Although further research is needed to establish the effectiveness of various forms of therapy, rehabilitationc ommonly focuses on ar ange of motion exercises to prevent and delay contractures (continued contraction of am uscle in the absence of stimuli), flexibility exercises to increase range of motion, progressive resistanceexercises to improves trengtha nd variousi nterventions to improvep osture, balance and the acquisition of functional skills (Anttila, Suoranta, Malmivaara, Makela &A utti-Ramo, 2008a;Anttila, Autti-Ramo, Suoranta, Makela &M almivaara, 2008b;Damiano, 2009). The intensity and frequencyo ft herapy vary across children, but manyh ave weekly therapy that extends into adulthood. Most therapy is delivered on an individual basis. Therapy may take the place of time that would have been devoted to group activities typical of normally developing children. Current literature stresses the need for newer,
task-related and intense programso ft herapy with life-style modifications (Bower, Michell, Burnett, Campbell &M cLellan, 2001;Damiano, 2006Damiano, , 2009. Given the marginalization that many children with CP experience, dance can be valuableb oth physically and socially, as well as beinganintroduction to an art form that is both aesthetic and athletic. Thus, therapeutic movement that enhances motor control and mimics community activityanswersseveral concerns for rehabilitationofchildren with CP. Our pilot class was based on these concerns and focused on two domains of the ICF namely: Body Structures and Functions, and Activities and Participation. We report on the rationale, implementation and design of the dance program for children with CP and on a pilot studyi ntendedt os urveyt he therapeutic benefits and modifications in participation related to the dance class as perceived by the children, parents and therapistsi nvolved (Lopez-Ortiz, Gladden, et al., 2009). Thep aradigm of dance,a nd in particular of classical ballet technique, offers aw elltested language-liker epresentation of whole-body movementsa ss equences of constitutive elements endowed with static and dynamic stability that is suitable for quantitative analysis. Thei nherent static and dynamic stability of classical ballet vocabulary is largely due to the use of positionsa nd movements that are guided by the limits of the mechanical range of motion of the joints and/or by actively stabilizing the joints in such positions. Since canonicalpositions and movements are defined within the anatomical Cartesian planes, quantitative analysisi ss implified. Moreover, classical ballet technique organizes body movement in space and time with al anguage structure where static positions become the building blocks or primitives for the creation and organization of complex and rich motor performance. Elite dance training has modular and hierarchical organization similar to some physical therapy techniques. Conservatory ballet training involves the training of static postures connected by, initially, slow movements (Kostrovitskaya, 2004). As training advances, the movements are executed at greater speedsa nd the static postures may blendi nto the movement. Therefore, in early training balletic movement tasks are in fact as equence of point-to-point reaching movements in free space. Point-to-point reaching movements have been successfully studied and used for stroke recovery (Rohrere ta l., 2004). Movementc ontrol in stroke survivorsp resents similar characteristics to that of children with CP, such as spasticity, discoordination and reduced selective muscle activation patterns. Thus, the inclusion of the ballet arm postures and movementfor training is consistent with current perspectives on motor control and motor learning by "compositionality" (Hammer, 2003;Krebs, Aisen, Volpe &Hogan, 1999;Miyamoto, Morimoto,Doya &Kawato, 2004; Morasso &Mussa Ivaldi,1 982). Here, compositionality refers to the construction of movement trajectories by combination of building blocks or primitives. The movement primitives in this consist of movement trajectory segments that that concatenate in the points of minimum velocity.
As an art form, classical ballet technique and training principless ystematically enhancealignment,flexibility, core strength, postural control and selective motor control. All these training goals are shared in the rehabilitationofchildren with CP. It is knownthat children with CP have less joint position sense as compared to neurologicallyi ntact children, with bias in the direction of internal rotation of the hip and pronation of the forearm (Wingert, Burton, Sincalir, Brunstrom &D amiano, 2009). One example of a ballet movement that addresses thisbias is the practice of external rotation and abduction of the hip joints that leads to astance with the heelsseparated in the coronal plane known as secondp osition in classical ballet terminology (see Figure 1). In this pilot dance program,a ugmentation of sensory information by tactile cues and closev olunteerw ork aid in accommodating for the lacko fs ensory input.P ositions that counter possible orthopedicd eformity due to CP werep racticed while avoiding physical pain. For the children with intellectual impairments, the ability to follow three-step directions was a prerequisite for enrollment. We assumed that children with borderline IQs couldlearn the basic routine of dance exercises, as well as execute the sequence of movements that were part of each class. Medical co-morbidities influence the child'se xperience of the dance class. Sensorydeficits such as hearing occur in up to 30% of children with CP and vision disturbances may be seen in up to 70% of children with CP (Ghasia, Brunstrom, Gordon & Tychsen, 2008;Venkateswaran &S hevell, 2008). These deficits limit processing of information and reduce visual and motor learning (Ostensjo, Carlberg &Vollestad, 2003). Seizures,i ntellectual disability and medications could reducet he attention to task and learning (Himmelmann,B eckung, Hagberg &U vebrant, 2006). Since dance hasa lready Care is taken to encourage proper musculoskeletal alignment using the barre for trunk support through the upper limbs while executing demi-plié .P hoto reproduced with permission.
been proveneffective in improving motor and mental scores of patients with Parkinson's disease (Erhardt, 2009), we expected positive outcomesi nt he children enrolled in the ballet-based dance class. Dance classes include music for the execution of movement. It has recently been suggested that early experiences during development encourage the simultaneous and interconnected wiring of movement and auditory representations in the brain ( Trainor, 2008). There is recent evidence of the presence of auditorymirror neurons that seem to be involved in tracking rhythmic auditory events in anticipation of their use in conjunction with the motor system (Chen, Penhune &Zatorre, 2008). Thus, it was plausible to expectthat the execution of movement wouldbefacilitated by the presence of abeat or musical rhythm in at least someforms of CP. Apreliminary studyinchildren with CP of Gross MotorFunction Classification Score (GMFCS) III and IV showed that music induces movement trajectories of the arms in ballet-like point-to-point reachingmovements that exhibit less abnormality of curvaturea nd increased blending of movement segments as comparedt ot he no-music condition (Lopez-Oritz, Gladden, et al., 2009;Lopez-Oritz, O'Shea, Mussa-Ivaldi & Gaebler-Spira, 2009). In light of this evidence,the dance program included live music.

Methods
Two dance classes were created, one for children with GMFCS Ia nd II and one for children with GMFCS III and IV. The children enrolled by their own choice with the supportoftheir parents.The program was supported by atertiary rehabilitation hospital, one outpatient physical therapy clinicand one community center for children. Volunteers from the medicalcommunity and some parents assisted during the class. The classes were held once weekly in session of five to eight weeks depending on location availability. The children were surveyed only once at the end of the first full session. Eight children, their assistants, teacher and pianist participated in ademonstration of the dance class in aformal dance recital. The children and their parents were surveyed againafter the performance.
Classical ballet training is progressive and repetitive, thereby allowing the student to perform movements consistently for mastery and to achieve the smooth execution of complexm ovement phrases. Preparatory conditioning exercises were developed with progressive levels of difficulty, with the end goal of executing the following classical ballet exercises.
. At theb allet barre:b alleticp osture control( voluntarye ffortt oa chieve anatomically correct alignment of the skeleton), demi-plié (bending of the knees while keeping the heels in full contact with the floor), relevé (raising of the heels with straight knees until full anklefl exion is achieved), positions of the feet (first position:t he heels are together with the lateral rotation of the femur leading to moderately turned out feet, secondposition: the feet are separated at approximately one foot distance, third position: the heel of one foot is pressed against the lateral aspect of the arch of the otherfoot, fourth position:one foot is placed in front on the other while still maintaining the outwardrotation at the hips, fifth position, the heel of one foot is placed in front of the toes of the otherfoot), battement tendú devant and à la seconde (one leg reaches to the frontorthe side while maintaining the hip turnout on both legs and extended knees, the foot brushed the floor to reach full plantar flexioninall joints), retiré (abduction of the hip joint while the knee bends to allow the fully plantar flexed foot to slide on the supporting leg), relevé lent devant (slow hip flexiona bduction and extension while maintaining turn out, extended knees and full planter flexionofthe foot), grand battement devant (fast hip flexion motionw ith extended knees and fully plantar flexed feet), demi-ramassé (trunk flexionf rom the hip joint while maintaining an eutral position of the vertebral column), cambré (back extension while attempting to maintain the lower limbs in a vertical position) (Kostrovitskaya &P isarev, 1995). . In the center: balleticp osture control, plié , relevé ,p ositions of the feet, battement tendú devant and à la seconde, retiré , relevé lent devant,positions of the arms and port de bras, sauté s (jumps starting and ending with demi-plié ), gallops,b alletic walking,b alleticr unning (quick small steps in relevé ). Thep reparatorye xercises were typicallyperformed prone or supine in agroup circle formation. As strength and selective motor control improved, the exercises were executed with the ballet bar as support, in the center in agroup circle formation, or along diagonals across the floor.
The children with GMFCS Iand II were typicallyassisted by one or two volunteers while the children with GMFCS III and IV were typicallya ssisted by twoo rt hree volunteers depending on their individual needs. Thep resence of ap ianist, or pianist and two violin players in the class was instrumental to establish an adequate tempo, time signature and desired movement quality for the execution of the movements and postures. Table 1shows the topics covered in the dance class that impact to body structuresa nd functions according to the ICF framework. Similarly, Table 2l ists the aspects of activities and participation included in the design of the dance class within the framework of the ICF. Surveys were developed based initially on the LIFE-H questionnaire (Noreau et al., 2007) and questions werecreated to investigatespecific aspects germane to the dance class not reflected in the pertinent Life-H questions. The surveys focused on participation and perceived therapeutic benefit from the dance class and dance performance. We developed questionnaires specific for the parents,children and the therapists involved in the program. The questions are presented alongwith the results in Tables 3-5. We obtained informed consent/assent from all the participants beforecompletion of the surveys as approved by the Institutional Review Board of NorthwesternUniversity.
Ther esults of the questionnaires that corresponded to ay es or no answer were convertedtoanumerical scalewith YES ¼ 1and NO ¼ -1.Abinomial statisticaltest was performed on the numeric results to establish the evidence against the null hypothesis that anegativeorpositiveresponse had equal randomprobability of occurrence (Snedecor &Cochran, 1996). In otherwords, asmall p -value indicated astrong positive trend in the data. Additional comments reported in the surveythat could not be scored in a numericscale are includedi nT ables 3-5.

Results
The results are presented in Table 3f or the children's survey, Table 4f or the parents' survey and Table 5for the therapists' survey. The children expressed the desire for more classes ( p , .0001),ahigh enjoyment level ( p , .0001),new interest in participation in a school group ( p , .04), new interest in watching ad ance show ( p , .0001) and new interest in attending an art show ( p , .004). The parents rated the class highly with overall enjoyment ( p , .0001),t herapeutic benefit ( p , .0001),p ositive influence in other ongoing therapy ( p , .04) and all woulde nrollt heir children again. Parents' comments included" In otice improved behavior,h appier, enjoying the activities", "They loved it, and they don't feel like they are in atherapy room". The therapists identified advantages workingi nt he dance class setting ( p , .002), would make changes in their personal
therapy sessions because of this program ( p , .002), had new treatment ideas as aresult of their participation in the class ( p , .04), felt that the children gained benefitsthat might not be achieved in atypical therapy program ( p , .01) and viewed the class as apositive adjunctt ot raditionalt herapy(p , .0001).T he parentsd id not perceive any improvements in head, trunk, arm and leg control.T herapists' comments included "This setting is very conductive for socialization and hands-ona pproach", "I am always Arts &H ealth looking for participation-based treatment ideasa nd dance incorporates so many rehabilitation principles and [this is] more motivating and engaging [than otherforms of therapy]", "A chancetobewith agroup of friends with common disabilities is always an encouragement", "allows children to explore and integrate other aspects of self". One therapist commented on the possibility of improving the dance class by incorporating quantitative progression measures everyc lass, while another desired incorporation of specific functional goals. Overall, the main areasofimprovement suggested were location, accessible parking, time of class, musicalvariety and qualitative outcome measures. Eight children participatedinaformal dance performance at ahigher-level educational institution. The children were excited and cooperative. Although one child showed signs of stress before the performance, she expressed pride in her achievement once the performanceh ad ended. As ag roup, the children that participated in the performance expressede njoyment in: dancing, dancing with their friends, being watchedb yo ther people, being on stage and receivingastandinga pplause from the audience.A ll the children surveyed in thisg roup ( n ¼ 6) reportedh aving fun duringt he performancea nd expressedinterest in participating in moreperformances. The children who participated in the dance performanceshowed improved attendance to the classes after the performance and look forward to having more classes.

Discussion
Ac lassical ballet based dance class provides an organized systematic approach to movement dexterity development that has been created and perfected through centuries of practical experience. Several of its training principlesa re aligned with physical therapy practices,b ut in addition it provides al anguage structure to movement that is absenti n other forms of physical training. This language structure allowsf or the coordination of complexm ovements far beyond thoser equiredf or daily life. Withini ts intrinsic organization, classical ballet positions and steps act to counter the main movementdeficits present in CP from flexibility, to postural control, to selective motor control. This places ballet training as ak ey instrument for the rehabilitation of children with CP. Moreover,
since ballet training is an expressive art, it incorporates cognitive, emotional and behavioral functions that are not necessarily part of traditional therapy. When presented as an optional movement activity for the children's enjoyment, they do not perceivei ta s "work" or "therapy". The inclusion in each class of children of similar ability provides a less stressful environment in which they can relate with their peersand happily encourage each other during the class. Individualized attention from the assistant therapists to each child and live music created ag roup class environment in which somatic empathy (i.e. attention to non-verbal, somatic cues, such as posture and gesture thereby gaining insight into whatthe child may be experiencing)from the instructor and therapists allowed each child to learn new movements and enjoythe process. This somatic empathy may be key for improved physical rehabilitation outcomesincerebralpalsy rehabilitation (Chaitow et al., 2010). Empathy was alsopresent in the musicians that played the music for the class and accommodated the mood changesand movements qualities required in each exercise. Table 3. Results of the Children Survey. The "Comments" column includes the general trends in the responses. For the question, Did you like the class? the children answered to: Please color in the face below that best shows how much you liked or didn't like the class: For the statistical purposes, all the answers of levels 0and 1were pooled as "yes" levels 2to5were polled as "no".
Children (  Was anything stressful for you or your child? 6 , .8 "long commute, hard at first but got used to it" "yes, she is too young for the class" "driving in the snow" "earlier time slot would be better" What differences do you see between dance and therapy?

12
, .01 * "Important that she did this with her peers" "the group dynamic was excellent" "any kind of creative expression -m usic, art, dance, can be extremely therapeutic" "she doesn't see dance class as work" "dance is more for enjoyment" "dance is more enjoyable for her" "dance is more fun and less stress" More classes ( n ¼ 3) Satisfied ( n ¼ 4) "Classes closer to home" "Teacher and assistants are wonderful" , .04 * "Incorporating breathing with movement" "Arm movement" "Rhythm coupled with movement" 'More relaxation techniques" "Music to home exercise program" "A whole class on just port de bras" "Body-based practice" "Tactile guidance for alignment" What benefits do you feel the children gained from this 10 , .01 * "Movement to music that was appropriately paced" program that might not be achieved in at ypical therapy program?
"Transitions are smoother in dance class" "Socialization" "Alternative structure from traditional therapy" "Body as ac reative soul" "Self confidence" "Exposure to music and rhythm" "Motivation and excitement"
Future plansi nclude addressing the research and location shortcomings as well as increasing the enrollment and frequency of the dance classes. We envision creating dance pieces for performances in addition to formal class demonstrations. In these performances, as in the dance class demonstrations, emphasis will be placed in the movements and dancing of the children, while the assistants strive to remain in the background provided supportasneeded.
Challenges to be addressed include the quantitative evaluation of the physical mobility outcomes, avalidated questionnaire specifictothe dance class and location accessibility.
Quantitative evaluation of movement characteristics remains statistically elusive as class sizes are small and variability of conditions and age is large. Additionally, the qualitative nature of the outcome measures is aweaknessofthe present study. However, as apilot exploratory study, we observed potential benefits in this intervention and the next step will be to incorporate validated quantitative assessmentsofprogress.Despite the need for improvement in theseareas, given that the children enrolled in the ballet class by their own choice and with parentalsupportand they all reportedenjoying the class, at the very least, they will have apositive memoryofacreative activity in which their own body and movements are the tools of expression.

Conclusions
The surveys revealedt hat this program incorporates key parentala nd ICF goals in occupational and physical therapy, such as the importance of facilitating improved motor control,e nhancing posture and trunk stability and coordinating motor movements in response to verbal commands and visual cues. From at herapy perspective,w hens uch goals are addressed in an environment that is engaging for the child, motivation and participation increase. The group setting creates natural peer modeling and promotes social interaction that is incrediblyvaluable to children who otherwise spend muchoftheir therapeutic time in one-on-one settings with adults. This class improves the child's What is your overall impression of the program as an adjunct to traditional therapy?

13
, .0001* "All the families enjoyed having an activity on Saturdays" "Dance is anice way to capture more holistic view of childphysical movements with emotions" "Offers the children something fun that doesn't feel like work" "Integration of other aspects of self" "It's aw onderful concept especially for young children" What were the dissimilarities to a therapy session?
Class structure Peer contact Less hands on Less "breaks" between activities * denotes statistical significance.
repertoire of activities and appreciation of as killed art. The dance program allowed children to develop interest and appreciation of an art form that has the potential to enhancetheir lives, physically,creativelyand emotionally. The enjoyment of the class by all the persons involved provided apositive therapeutic environment for the children and their families. There is aneed for better outcome measurement tools in the qualitative and quantitative aspects on the class. Combining movement and expressive art in the dance class, augments the potential for improving children's posture, movement abilities and societal participation as well as generating rewarding experiences throughout alifetime.