Secondary Logo

Journal Logo


Creative Dance Practice Improves Postural Control in a Child With Cerebral Palsy

Stribling, Kate PT, DPT, PCS; Christy, Jennifer PT, PhD

Author Information
doi: 10.1097/PEP.0000000000000450


Cerebral palsy (CP) is a group of permanent movement and postural disorders attributed to an injury to the developing fetal or infant brain1 and is the most common motor disability in childhood.2 Children with CP often have difficulty with strength, proprioception, and balance.3,4 Physical therapy (PT) interventions for children with CP typically incorporate functional movement through all planes for balancing and strengthening activities.3–5

Dance is a typical childhood activity that provides fun, musically motivated exercise.6,7 The American Dance Therapy Association advocates for the use of dance as movement therapy for adults and children across settings.7,8 Dance can be used as an assessment and as an intervention by focusing on motor control and motor learning.7,8 Similar to PT interventions, dance instruction promotes movement in all planes of motion to improve balance, postural stability, and strength.6 There is growing evidence of the use of dance as an intervention for children with neurological disabilities.9,10 Studies demonstrate improved balance scores in a child with congenital myotonic muscular dystrophy on the Bruininks-Oseretsky Test of Motor Proficiency (BOT-2) following tap dance classes,9 and parent-perceived therapeutic benefits for their children with CP following ballet lessons.10

Visual or tactile cues for body alignment and motor imagery are common aspects of creative dance instruction that combine with active movement to facilitate body-space awareness.3,11 Motor imagery can improve balance and motor skills in adults with hemiparesis following stroke, but no publications were found investigating motor imagery use with children.12–14 The purpose of this case study was to determine whether a creative dance class would improve postural control and balance for a child with CP.


This case study was approved by the Institutional Review Board of the University of Alabama at Birmingham. We refer to the participant as “creative dance student,” (CD). CD's mother provided informed consent and CD assented.

We gathered information from parent interview and during the pretest. At the time of the study, CD had no history of major illnesses. She was 11 years old with a diagnosis of spastic triplegic CP, Gross Motor Function Classification System (GMFCS) level II. “Level II” indicates that CD walked independently but was challenged by long distances or walking on uneven surfaces.15 We recruited CD from the Children's of Alabama pediatric hospital in Birmingham, Alabama, where she received periodic outpatient PT (last session 6 months before the study). She received no outpatient PT during the study. CD had prior creative dance experience 8 years before the study. She was developmentally typical for cognition and excited to participate in dance classes.

CD had not undergone recent surgeries, medical procedures, or changes in medication. Her prior surgical procedures included implantation of a ventriculoperitoneal shunt during infancy, multiple shunt revisions, a selective dorsal rhizotomy, and single Botox injections in the right arm and both ankles (procedure dates not reported). During the study CD was taking Vyvanse (40 mg every morning), Zoloft (25 mg every morning), Intuniv (4 mg every morning), and Trileptal (250 mg every 12 hours). Prior equipment included a posterior rolling walker and ankle-foot orthoses. At the time of our study, she used custom arch supports, but these were not worn during dance interventions.

Using manual muscle testing we found no active extension in the right wrist, mild deficits in right upper extremity strength, and impaired strength in lower extremity muscles, more so on the right than the left. Range-of-motion limitations included a pronation contracture of the right forearm, excessive internal rotation of the right hip, bilateral limitations in ankle dorsiflexion, and hamstring length. Ryder's test was equal bilaterally at 25°. We found a 2-cm leg length discrepancy on the left. Values are reported in Supplemental Digital Content 1 (available at:

CD stood with a crouched posture and bilateral calcaneal eversion (Supplemental Digital Content 1, available at:, forward head/shoulders, and increased thoracic kyphosis. She had difficulty with frontal and transverse plane movements and with walking backward. During gait she used momentum and lower extremity circumduction to assist with forward progression.


A second-year doctor of physical therapy (DPT) student and a PT faculty member at The University of Alabama at Birmingham assessed outcomes 2 days before starting the dance classes and following the final session.

We used the SMART Balance Master/EquiTest system (NeuroCom, Natus Medical, Inc) to assess CD's balance and postural control abilities. This system uses a force platform with a 100-Hz sampling rate and a moveable visual surround and platform to measure the sensory and motor functions involved in balance control. Computerized dynamic posturography (CDP) using NeuroCom's SMART system is commonly recognized as the “gold standard” of balance assessments, with high validity and reliability in adults and children.16,17 CDP testing consisted of the sensory organization test (SOT), the adaptation test (ADT), the motor control test (MCT), and the limits of stability (LOS) protocol.

The SOT identifies postural control impairments by systematically examining center-of-gravity (COG) response to different, sometimes conflicting, sensory information (ie, visual, vestibular, and somatosensory) during 6 conditions: (1) eyes open, fixed surface, and visual surround; (2) eyes closed, fixed surface; (3) eyes open, fixed surface, swayed visual surround; (4) eyes open, swayed surface, fixed visual surround; (5) eyes closed, swayed surface; and (6) eyes open, swayed surface, and visual surround. Three trials of each condition are averaged, yielding a sway score between 0 (ie, fall) and 100 (ie, no sway) for each condition and a composite score. In addition, the SOT provides a single data point of the COG alignment at the beginning of each trial.18,19

The ADT and the MCT measure motor functions related to balance control in response to a perturbation of the standing surface. The ADT assesses ability to modify muscle activation to recover postural control and minimize sway in response to sudden, unpredictable toes-up or toes-down surface movements.18,19 The MCT measures ability to recover balance following a sudden, unpredictable anterior or posterior translation of the surface. It assesses the latency, amplitude, and symmetry of the response.18,19

The LOS protocol measures postural control by quantifying volitional center-of-pressure (COP) displacement and movement velocity in 8 directions.18 The participant is asked to make a visual representation of the COP move toward a target on a screen after hearing a short beep, and then hold the position as close to the target as possible for the 8 second trial. Data include reaction time, movement velocity, endpoint excursion, maximum endpoint excursion, and directional control.


A DPT student with experience as a dance instructor carried out the intervention. Classes took place at The Dance Foundation, a nonprofit studio space in Homewood, Alabama, and were one-on-one with CD and the instructor.

Dance classes were 1 hour, twice weekly for 8 weeks. Using the path model for dosing parameters for children with CP as a guide, we chose the frequency, time, and intensity for our intervention considering the availability of similar programs in the community; access to the studio; and time constraints for CD, her family, and the DPT student instructor.20 We used common creative dance methods, including the 9 lines of movement, which is a motor imagery approach for postural alignment in dancers,11,21 and the BrainDance movement sequences.6,22 We incorporated visual demonstrations, verbal and tactile cues for alignment, and verbal instruction in motor imagery and movements. The same music soundtrack was used for every class. Large mirrors in the studio provided visual feedback. CD was either barefoot or in ballet slippers. The study concluded with a choreographed performance for CD's friends and family.

Lesson plans were based on CD's abilities and needs. A detailed lesson plan is included as Supplemental Digital Content 2 (available at: Each dance class started in supine or side-lying positions and consisted of a series of movements on the floor, followed by standing exercises. Exercises incorporated motor imagery and included limb and trunk isolations, balance work, and movement through the studio space. After every session, we gave CD alignment and motor imagery activities for home.


Sensory Organization Test

The COG score is the number of degrees off center when each trial began. A score of 0 indicates that the person's COG was in midline, a negative number indicates that the person's COG was to the left or posterior of center, and a positive number indicates that the person's COG was to the right or anterior of center. Following intervention, CD's COG alignment was more consistent and closer to midline than during pretest for all conditions (Figure and Table). COG at the start of condition 3 had the most improvement and moved closer to midline (preintervention: −0.75° left and 1.35° anterior; postintervention: −0.53° left and −0.37° posterior) with more consistency across posttest trials. Again, this consistency can be seen numerically by looking at the range scores. Pretrial scores fell within 0.9° of each other for left/right and 2.3° for anterior/posterior, whereas postrial scores fell within 0.7° and 0.4°, respectively (Table).

SOT equilibrium center-of-gravity and LOS center-of-pressure alignment. The SOT equilibrium COG alignment for the pretest (top left) is more diffuse compared with the posttest (top right). Symbols represent the alignment of the COG during each of the 6 conditions. LOS testing trace the path of CD's purposely displaced center of pressure testing before (bottom left) and after (bottom right) intervention. CD indicates creative dance student; COG, center of gravity; LOS, limits of stability; SOT, sensory organization test.
TABLE - Comparison of Sensory Organization Test Scores Before and After Interventiona
1 79.3 81.3 97 96 Averagec: 1.03/1.13 Average: −0.86/−0.63
T1: 0.5/1.0 T1: −0.2/−0.1
T2: 1.8/1.4 T2: −1.0/−0.8
T3: 0.8/1.0 T3: −1.4/−1.0
2 80.7 73.3 97 94.3 Average: −0.63/−0.6 Average: −1.4/−1.5
T1: −1.1/−0.6 T1: −1.3/−1.3
T2: −0.1/−0.3 T2: −1.9/−1.3
T3: −0.7/−0.9 T3: −1.1/−1.8
3 64.7 78.7 88.7 96.7 Average: −0.75/1.35 Average: −0.53/−0.37
T1: −0.8/1.2 T1: −0.1/−0.3
T2: −0.7/1.5 T2: −0.7/−0.6
T3: 0.1/−0.8 T3: −0.8/−0.2
4 65 65.7 80.7 86 Average: −0.65/−1.0 Average: −0.67/−1.03
T1: −0.9/0.5 T1: −0.4/−0.6
T2: −0.4/−0.3 T2: 0.0/−0.8
T3: 0.4/−1.7 T3: −1.6/−1.7
5 Fall Fall Fall Fall
6 Fall Fall Fall Fall
Abbreviations: A, after; B, before; COG, center of gravity; E, equilibrium score (amount of sway; 100 is no sway); S, strategy score (100 is total use of ankle strategy, 0 indicates a hip-dominant strategy); SOT, sensory organization test.
aEquilibrium and strategy scores represent the average of 3 trials.
bA positive number means right or anterior; negative number means left or posterior; the first number is right/left and the second is anterior/posterior (eg, for condition 1, COG-B, 1.03/1.13 means she tended to sway to the right and anterior and A 0/0 would mean the person was perfectly in the middle).
cAverage of 3 trials (T1, T2, and T3 are trials 1 through 3, respectively).

The equilibrium score measures the amount of sway during each trial. Less sway is associated with better postural control. A score of 100 indicates no sway (Table). CD's composite score improved from 39 to 42 after intervention, indicating an overall decrease in sway. Sway improved on conditions 1, 3, and 4 from 79.3 to 81.3, 64.7 to 78.7, and 65 to 65.7, respectively, whereas condition 2 showed an increase in sway on the posttest, 80.7 to 73.3. Sway score data were not collected for conditions 5 and 6 because of falls.

Strategy analysis quantifies the amount of movement at the ankles (ie, ankle strategy) and at the hips (ie, hip strategy) used to maintain balance (Table). A score of 100 indicates total use of an ankle strategy. CD's use of an ankle strategy decreased slightly for conditions 1 and 2 (from 97 to 96, and 97 to 94.3, respectively). Her use of an ankle strategy increased for conditions 3 and 4 (from 88.7 to 96.7, and 80.7 to 86, respectively). Falls occurred for conditions 5 and 6.

Adaptation Test

CD completed 5 ADT trials before and after intervention. CD was unable to complete the toes-up test pre- or postintervention because of falls. However, she was able to complete the toes-down test, with more consistent and faster reaction times postintervention. Her pretest mean was 131 milliseconds, with a range of 85 to 147 milliseconds; her posttest mean was 103 milliseconds, with a range of 92 to 112 milliseconds. Although the posttest values are outside of the normal range, CD was able to complete the test without touching the walls, unlike in the preintervention trials.

Motor Control Test

For the MCT a score of 0 indicates complete left-sided use, 100 indicates symmetric use of both sides, and 200 indicates complete right-sided use. After intervention CD showed increased use of her right lower extremity, becoming more symmetric during medium amplitude backward (preintervention: 84, postintervention: 92), large backward (preintervention: 90, postintervention: 94), and medium forward (preintervention: 93, postintervention: 95) platform translations. During large forward translations, weight symmetry moved slightly to the left following intervention (preintervention: 91; postintervention: 89).

CD's average reaction time to platform movement improved slightly from 140 millisecond pretest to 139 milliseconds posttest. However, symmetry of her left and right reaction time improved during all backward translations (large left preintervention = 150 milliseconds, right = 110 milliseconds; large both postintervention 140 milliseconds; medium left preintervention = 110 milliseconds, right = 120 milliseconds; medium both postintervention = 130 milliseconds) and during large forward translations (left preintervention = 130 milliseconds, right = 200 milliseconds; left postintervention = 130 milliseconds, right = 120 milliseconds), indicating more coordinated lower extremity use during reactive balance. For posttest medium amplitude forward translations, reaction times reversed from slower with the left (180 milliseconds) and faster with the right (120 milliseconds), to faster with the left (130 milliseconds) and slower with the right (190 milliseconds).

Limits of Stability

CD's COP control and excursion improved following intervention (Figure): composite endpoint increased from 42% to 44%, maximum excursions increased from 65% to 71%, and composite directional control improved from 34% to 57%. CD's movement velocity decreased slightly from 4.6° to 3.6° per second, and reaction time reduced from 0.71 to 0.62 seconds, possibly indicating more controlled and purposeful movement. Falls occurred during the pretest but not during the posttest.


For 8 weeks, CD received 2 hours per week of creative dance instruction incorporating postural alignment imagery, movement isolations, and balance exercises. Additional motor imagery practice occurred at home.

Following intervention, CD's COG alignment at equilibrium (Figure) appeared more consistent for all SOT conditions and more closely resembled than that of a child with typical development.4 Following intervention, CD's scores for SOT conditions 1, 2, and 3 (81.3, 73.3, and 78.7, respectively) were better than or equal to scores from the Yi et al23 study for children with GMFCS level I (on conditions 1, 2, and 3: 78.0, 73.8 and 71.7, respectively), indicating a significant improvement for CD, who was classified as GMFCS level II. Yi et al did not report scores for conditions 4 through 6.

CD predominately used an ankle strategy to balance during all SOT conditions. The exact amounts of hip versus ankle strategy varied from the pretest to the posttest, decreasing slightly for conditions 1 and 2, and increasing slightly for conditions 3 and 4. These findings agree with those from Shumway-Cook and Woollacott,24 who found a diverse set of functional strategies are used by children with CP for maintaining balance following perturbations.

CD showed more symmetric use of lower extremities during the MCT and improved reaction times on both the MCT and the ADT. These results appear to indicate improved use of the right lower extremity and improved postural control in response to perturbations. However, because there is no established clinically important difference in posturography values for children with CP, we are unable to state whether this change was clinically relevant.

A study25 from El-Shamy et al25 examined LOS for children with CP GMFCS level II using the Biodex Balance System. They found significant improvements in directional control and efficiency following balance training. When comparing CD's results with those of the 2 groups in the El-Shamy study, we found that her directional control at baseline and following intervention (34%, 57%, respectively) exceeded that of the group who received balance training on the Biodex (%21 ± 0.56%, 34.36% ± 0.51%, respectively).25 The Biodex testing requirements differed slightly and CD's baseline scores were already greater than the children in this study, so although our findings mirrored those of El-Shamy, we are unable to state whether there was clinically relevant improvement.

During the posttest, CD continued to have challenges with maintaining balance when her vision was obscured (ie, SOT conditions 2 and 5). These findings may be due to CD's age in addition to poor postural control. Cherng et al4 found no significant difference between children 11 years and younger with ambulatory CP and their typical peers when visual input was unreliable or occluded, as long as the somatosensory input was steady, as in SOT condition 2. When unreliable somatosensory input was coupled with occluded or unreliable vision, as in condition 5 and 6, children with CP had more challenges than typical peers.4 Our findings were consistent with these.

Similarly, CD fell both pre- and posttest during the ADT sudden toes up perturbations. Large forward translation of the platform during the MCT also resulted in falls pre- and posttest. It is possible that dose or duration of the interventions or environmental factors were not adequate to impact these outcomes.

Motor imagery and repeated physical practice led to improvements in balance recovery efficiency on aspects of the SOT, MCT, and ADT; more consistent organization overall on the SOT; improved symmetry of response on the MCT; and the improved directional control on the LOS test. Studies correlate clinical findings like ours with reduced fall risk; more functional standing balance; and improved walking, running, and jumping.3,25 Following our program, CD stated she felt more stable and was excited to begin a community dance class. These perceived benefits and participant motivation to continue are consistent with other studies examining dance as an intervention.9,10

The results of our study should be interpreted with caution because this is a single-subject case study with limited generalizability. Undoubtedly the dose frequency, duration, and intensity of our intervention impacted the outcomes; however, these were based on availability of resources. More research needs to be done to determine the effects these have.

Other topics for further research could include larger sample sizes of children with CP using single-subject design and analysis. We recommend separating testers from the intervention, assessing gait and posture using 3-dimensional analysis, and incorporating participation-oriented and movement-quality assessment tools.


Our study was a first assessment of creative dance intervention for CP. Our results support that dance incorporating motor imagery is a viable intervention for improving balance (stability recovery and directional control) in children with CP. Our results agree with similar research on balance training in children with CP4,23–25 and are consistent with studies investigating dance-based therapies for children with neurological disabilities.9,10 The interventions used in this study could be easily implemented by clinicians across settings.


The authors thank the participant and her family for their time and The Dance Foundation in Homewood, Alabama, for their generous donation of studio space.


1. Rosenbaum P, Paneth N, Leviton A, Goldstein M, Bax M. A report: the definition and classification of cerebral palsy April 2006. Dev Med Child Neurol Suppl. 2007;109:8–14.
2. Accardo PJ. Neurodevelopmental Disabilities in Infancy and Childhood. Atlanta, GA: Centers for Disease Control and Prevention; 2008. Accessed February 15, 2015.
3. Dewar R, Love S, Johnston LM. Exercise interventions to improve postural control in children with cerebral palsy: a systematic review. Dev Med Child Neurol. 2015;57(6):504–520.
4. Cherng R, Su F, Chen JJ, Kuan T. Performance of static standing balance in children with spastic diplegic cerebral palsy under altered sensory environments. Am J Phys Med Rehabil. 1999;78(4):336–343.
5. Law M, Darrah J. Emerging therapy approaches: an emphasis on function. J Child Neurol. 2014;29(8):1101–1107.
6. Gilbert AG. Creative Dance for All Ages. Reston, VA: NDA/AAHPERD; 1992.
7. Strassel JK, Cherkin DC, Steuten L, Sherman KJ, Vrijhoef HJ. A systematic review of the evidence for the effectiveness of dance therapy. Altern Ther Health Med. 2011;17(3):50–59.
8. American Dance Therapy Association. ADTA informational brochure. Accessed July 1, 2016.
9. Biricocchi C, Drake J, Svien L. Balance outcomes following a tap dance program for a child with congenital myotonic muscular dystrophy. Pediatr Phys Ther. 2014;26(3):360–365.
10. López-Ortiz C, Gladden K, Deon L, Schmidt J, Girolami G, Gaebler-Spira D. Dance program for physical rehabilitation and participation in children with cerebral palsy. Arts Health. 2012;4(1):39–54.
11. Krasnow D, Wilmerding MV. Motor Learning and Control for Dance: Principles and Practices for Performers and Teachers. Champaign, IL: Human Kinetics; 2015.
12. Hosseini SA, Fallahpour M, Sayadi M, Gharib M, Haghgoo H. Impact of mental practice on stroke patients postural balance. J Neurol Sci. 2012;322(1/2):263–267.
13. Nagano K, Nagano Y. The improvement effect of limited mental practice in individuals with poststroke hemiparesis: the influence of mental imagery and mental concentration. J Phys Ther Sci. 2015;27(8):2641–2644.
14. Malouin F, Richards CL. Mental practice for relearning locomotor skills. Phys Ther. 2010;90(2):240–251.
15. Palisano R, Rosenbaum P, Bartlett D, Livingston M. Gross Motor Function Classification System. Dev Med Child Neurol. 1997;39:214–223.
16. NeuroCom International, Inc. Clinical Justification Computerized Dynamic Posturography. Published 2004. Accessed February 29, 2015.
17. Liao H, Mao P, Hwang A. Test-retest reliability of balance tests in children with cerebral palsy. Dev Med Child Neurol. 2001;43:180–186.
18. NeuroCom International, Inc. NeuroCom Balance Manager Systems Sole Source. Accessed February 29, 2015.
19. Natus Balance & Mobility. CDP Protocols. Accessed February 15, 2016.
20. Gannotti ME, Christy JB, Heathcock JC, Kolobe TH. A path model for evaluating dosing parameters for children with cerebral palsy. Phys Ther. 2014;94(3):411–421.
21. Sweigard LE. Human Movement Potential: Its Ideokinetic Facilitation. Forest Grove, OR: Allegro Editions; 1974.
22. Gilbert AG. BrainDance [video]. Reston, VA: NDA/AAHPERD; 2003.
23. Yi S, Hwang JH, Kim SJ, Kwon J. Validity of pediatric balance scales in children with spastic cerebral palsy. Neuropediatrics. 2012;43(6):307–313.
24. Woollacott MH, Shumway-Cook A. Postural dysfunction during standing and walking in children with cerebral palsy: what are the underlying problems and what new therapies might improve balance? Neural Plasticity. 2005;12(2/3):211–219.
25. El-Shamy SM, Abd El Kafy EM. Effect of balance training on postural balance control and risk of fall in children with diplegic cerebral palsy. Disabil Rehabil. 2014;36(14):1176–1183.

case report; cerebral palsy; child; dance; physical therapy

Supplemental Digital Content

© 2017 Wolters Kluwer Health, Inc. and Academy of Pediatric Physical Therapy of the American Physical Therapy Association