The International Journal of Psychosocial Rehabilitation
The efficacy of music therapy in effecting behaviour change in persons with Cerebral Palsy
Leith Krakouer
Stephen Houghton
Graham Douglas
John West

Centre for Attention & Related Disorders
The Graduate School of Education
The University of Western Australia

Krakouer L., Houghton. S., Douglas, G., & West J. (2001)  The efficacy of music therapy in effecting behaviour
changein persons with Cerebral Palsy.  International Journal of Psychosocial Rehabilitation. 6, 29-37

Address for Correspondence: Professor Stephen Houghton
The Centre for Attention & Related Disorders
The Graduate School of Education
The University of Western Australia
Nedlands 6009, Australia
Tel: ++ 618 9380 2391
Fax: ++ 618 9380 1052


A music therapy program using specialised musical instruments specifically chosen to meet the individual needs of five persons with Cerebral Palsy, was implemented using a multiple baseline research design. An empirical evaluation revealed statistically significant changes in group behaviour following the introduction of the programme. Further statistical analysis using DMITSA 2.0 (Crosbie & Sharpley, 1991), a statistical programme specifically developed to analyse data from interrupted time-series designs, revealed statistically significant changes in the individual target behaviours of all five participants. The findings are discussed in relation to previous research.

The efficacy of music therapy has been investigated with individuals with a wide range of disabilities, including for example: Autism (Wager, 2000), dementia (Korb, 1997), acute brain injury (Nayak et al., 2000), Parkinson’s disease (Pacchetti et al., 2000), Alzheimer’s disease (Aldridge, 1998), attachment disorder (Brotons & Pickett-Cooper, 1996; Burkhardt-Mramor, 1996), chronic schizophrenia (Zhang & Cuie, 1997), depression (Suzuki, 1998), and multiple sclerosis (Davis, 1998). The positive outcomes reported in the majority of these studies suggest that it is an efficacious treatment. Furthermore, a meta-analysis of 21 empirical studies which used music therapy as an intervention with a total of 336 participants with dementia, demonstrated that its effect was highly significant (Koger, Chapin, & Brotons, 1999). More recently, a review of clinical music therapy (see Tervo, 2001) provided additional support by showing how music, and in particular rock music, can give adolescents “the possibility to express, be in contact with and share among themselves feelings of anger, rage, grief, longing and psychological disintegration”.
Case studies appear to be the most frequently reported music therapy based interventions. For example, Burkhardt-Mramor (1996) reported that insight-oriented music psychotherapy addressed the attachment problems of an 11-year old male with a history of physical and emotional abuse. In an innovative music therapy study, Krueger (2000) incorporated the use of information technology with a 10th grade student with behavioural problems, to allow him to “eventually be able to focus on music making for 45 minute periods”. Music therapy has also been shown to be effective with older individuals. For example, Wager (2000) found music therapy was a “valuable on-going leisure activity” for a 36 year old male with autism and mental retardation, while Davis (1998) reported that music therapy created closer family bonds and improved psychosocial and spiritual well-being for a 48 year old male with multiple sclerosis.
Larger scale group evaluations of music therapy have also demonstrated its effectiveness as a treatment. For example, in one recent study (Pacchetti et al., 2000) weekly choral singing, voice exercises, and rhythmic and free body movements were used with a group of 16 individuals diagnosed with Parkinson’s disease. Improvements were subsequently noted in their emotional functions, daily living and quality of life. Similarly, other studies using music therapy intervention have demonstrated significant improvements in: the social interactions, mood and participation rates of groups of individuals with traumatic brain injury and stroke (Nayak et al., 2000); and emotional states of persons with schizophrenia (Zhang & Cuie, 1997). Reductions have also been shown to occur in the agitation behaviours associated with Alzheimer’s patients (Brotons & Pickett-Cooper, 1996), and in depressive symptoms in elderly adults (Suzuki, 1998) as a consequence of music therapy programs.
Although studies confirm the efficacy of music therapy as an intervention with a wide rage of disabling conditions, it appears that limited studies utilise statistical analyses, and that of those which do, none to date appear to have utilised a time series approach. That is, pre and post test group data tend to be analysed, with no focus on the behaviour of individual participants over time. The present research sought to address this by implementing a multiple baseline research design and by using DMITSA 2.0 (Crosbie & Sharpley, 1991) to examine the efficacy of music therapy with individuals with Cerebral Palsy (CP). DMITSA 2.0 is a statistical programme specifically developed to analyse data from interrupted time-series designs. This programme allows tests of statistical significance to be made on data which have been collected over time and which may be divided into at least two sections (i.e., baseline and intervention).
In addition, the specific aims of the present research were to determine: (i) the effectiveness of music therapy with people with CP; and (ii) the suitability of matching specialised musical instruments on an individual basis in a music therapy program to achieve specific outcomes.

Participants and setting

Five individuals (three males and two females) whose ages ranged from 20 to 45 years participated. All had received a clinical diagnosis of Cerebral Palsy (CP) by medical personal; the consequent degree of disability varied across the individuals. Of the five participants all were confined to wheelchairs with the exception of one female.
The study was undertaken at one residential hostel, which forms part of The Cerebral Palsy Association of Western Australia. One large room in the hostel was set aside without interruption for the duration of the study. All Musical Therapy sessions were conducted in this room by the first author.
Research Design
A multiple baseline research design across participants was used in the research. According to Barlow, Haynes, and Nelson (1986) the multiple baseline is probably one of the best designs available for practitioners. It does not require withdrawal, it is fairly simple, and applied opportunities for its use abound once systematic measures are being taken (Barlow, et al. 1986). All observations of participants undertaken during baseline and intervention phases were recorded on videotape. A frequency count was subsequently obtained for each participant's target behaviours per session. The convention used was that where continuous or repetitive behaviour occurred a tally of one was recorded. If there was a five-second break between the target behaviour(s) and then it was re-initiated it was counted as a new behaviour.
As can be seen in Figure 1 the number of observations for each participant varied due to the nature of the research design. For Participant 1 there were five baseline observations, for Participant 2 eight, for Participant 3 ten, for Participant 4 fifteen, and for Participant 5 twenty. Each observation session was 45 minutes in length. Hence, the total amount of observation time varied for each participant, ranging from 3.75 hours (Participant 1) to 15 hours (Participant 5). During baseline participants were observed in their regular group activity of listening to the first author play his guitar and sing a selection of songs. Participants were then randomly assigned to the order in which they received the intervention.
Following baseline observations the intervention phase comprising Music Therapy was introduced. Participants were provided with specific musical instruments commensurate with the activities required to remediate their inappropriate behaviour(s) (see Table 1). Observations, each of 45 minutes duration, continued as follows: Six 45 minute observations for Participant 1, eight for Participant 2, and 10 for Participants 3, 4 and 5. Thus the total amount of video recordings obtained during intervention ranged from 4.5 hours (Participant 1) to 7.5 hours (Participants 3, 4 and 5).
Table 1. Participant’s target behaviour and specialist musical instrument used in the programme.
Target behaviour and instrument

1. Male 45 years Restricted movement in lower torso.
Reduced frequency of hand movements
Using a drum, beater and go go bells to increase hand movements.
2. Female 23 years 
Little movement in lower torso. Impaired communication. 
Poor hand-eye co-ordination when using compic board to (tap) communicate with others
Using a Bungbun to facilitate specific downward hand movements in extended regular motion similar to skill required for compic board. 
3. Male 28 years Doubling over at the chest which creates pressure on lungs and serious potential respiratory problems
Difficulty with hand-eye co-ordination. Failure to attain an upright posture (leading to possible respiratory problems).
Using a castanet to facilitate a definite movement backwards to an upright position, thus correcting posture.
4. Female 29 years Limited control over body movements particularly as excitement level increases.
Waving of arms and feet in an erratic uncontrolled manner. 
Using bells with hands to perform controlled shaking motion and movement of feet to beat of music. 
The Music Therapy Intervention Programme
The theme of the music therapy intervention program was emotions, specifically "happiness". The number of sessions varied across participants with the minimum number of sessions being six and the maximum 10. Each session, which attempted to encourage participants to use their instrument(s) to express happiness, began with the music therapist (first author) playing a bright melody on a guitar. Each 45-minute music therapy intervention session was divided into three movements or activities, each lasting 15 minutes. When an individual began the intervention phase the first 15 minutes involved the therapist playing passive melodies to assist him/her to relax. The second 15 minutes introduced activities where he/she was encouraged to respond and participate either vocally or through hand movements (clapping etc.). The therapist stopped playing the guitar at specific points and moved into a phase which consisting of vocal type chanting and singing. The basis of this was percussive sounds to encourage variations in responses from the participant. In the final 15 minutes the individual used the chosen instrument fundamental to achieving their specific goal (refer to Table 1 for a description of the instruments assigned).
Consent to participate was obtained from the relevant University Departmental committee, the Cerebral Palsy Hostel personnel, and from the participants themselves. For each session the therapist met with participants on a group basis. Once each participant’s series of baseline observations had concluded the music therapy intervention was introduced. Individuals with lengthened baselines continued in the sessions but did not participate in the vocal and body movements (second 15 minutes) and use of musical instruments (final 15 minutes). Rather these individuals continued listening to the activities without active participation. All sessions were video recorded and these were viewed and scored by the therapist at a later date.
Inter-rater reliability
To establish the reliability of the therapist’s observations a second person not involved in the research, but trained in the use of behavioral observation, viewed all of the video recorded sessions. When an instance of agreement occurred a tally was made and an identical procedure was followed with disagreements. The level of inter-rater reliability was calculated as the number of agreements divided by the number of agreements plus the number of disagreements. This value was then converted to a percentage. For this study the inter-rater reliability was calculated to be 97%
Statistical Analysis
Data were analysed in two ways. First, the Wilcoxon signed rank test was used to examine whether there was a significant increase in target behaviours from baseline to intervention. Second, data for individual participants were analysed using DMITSA 2.0 (Crosbie & Sharpley, 1991) which is a statistical programme specifically developed to analyse data from interrupted time-series designs. This programme allows tests of significance to be made on within-subject data which have been collected over time and which may be dividedinto at least two sections (i.e., baseline and intervention).
As can be seen in Figure 1 the baseline trends for all five participants are relatively stable with the frequency of desired behaviour(s) being low (occurring between 0 and 2 instances per 45 minute session). This stability continued irrespective of the length of baseline. When the music therapy intervention was introduced the frequency of target behaviours appeared to increase for all participants since the trends depicting the frequency of behaviours are all ascending. It should be noted, however, that in the case of participants 2 and 3 there is some tentative evidence of a plateauing effect if not a descending trend in behaviour as the intervention continued.
Figure 1: Frequency of behaviours across baseline and intervention
In only three of the five 45 minute baseline sessions conducted for participant 1 was the target behavior (hand movement) observed and then only once in each session. Following the introduction of music therapy the frequency of hand movement increased to seven per session, culminating in nine in the final 45-minute period. Similarly, low frequencies of the target behaviours i.e., zero or 1 to 2, were recorded for participants 2 and 3 during baseline. An apparent increase in the frequency of the target behaviours appeared to occur following the introduction of music therapy. For participants 4 and 5 (participant 4 was observed for two desired behaviours i.e., hand and feet movement) there was an increase in target behaviours when music therapy was introduced (from 6 to 14 for participant 4, and from 8 to 12 for participant 5).
Statistical Analysis
The results of the Wilcoxon signed ranks test provided evidence to suggest that the mean ranks of the target behaviours from pre-intervention and post-intervention were not equivalent (p = .028). More specifically, there was a significant increase in observed target behaviours from pre- to post-intervention, with means of 0.75 and 9.62 respectively. This suggests a significant increase in the frequency of target behaviours for the group following the introduction of music as a therapy.
Individual Data
Data for each individual participant’s target behaviour are presented in Table 2. As can be seen the frequency of target behaviour per minute was low for all participants during baseline. Participant 4's second behaviour (feet movement) was the least frequently occurring of all behaviours at .008 instances per minute. However, following the introduction of music therapy this behaviour exhibited the greatest increase in occurrence rising to .296 instances per minute. The DMITSA analysis, which is shown in Table 2, indicated that there was a significant increase in the frequency of all participants targetbehaviours
Table 2. Frequency of individual’s target behaviours across baseline
and intervention mean rates per minute
Participant Baseline
mean rate/minute
mean rate/minute
1. .022 .167 13.887 p<.009
2. .019 .189 32.511 p<.001
3. .015 .193 109.702 p<.001
4i .013 .175 41.845 p<.001
4ii .008 .296 51.149 p<.001
5 .017 .213 49.280 p<.001


The results of this research provide support for the efficacy of music therapy in bringing about significant changes in specific behavious of persons with Cerebral Palsy, a disability which has yet to be reported in the music therapy evaluation literature. This provides additional support to the current evidence pertaining to a range of disabilities (e.g., Autism, dementia, acute brain injury, Parkinson’s disease, Alzheimer’s disease, attachment disorder, chronic schizophrenia, Depression, Down's Syndrome, and multiple sclerosis).
Unlike previous research the present study analysed data on each individual's behaviour change over time, in addition to that of group change. Overall, there were significant changes in group behavior and in the behavior of all participating individuals following the introduction of Music Therapy. This provides support to previous research which has adopted a case study approach (e.g., Burkhardt-Mramor, 1996; Wager, 2000; and Davis, 1998), or has involved group evaluations (e.g., Pacchetti et al., 2000; Nayak et al., 2000; Brotons & Pickett-Cooper, 1996; and Suzuki, 1998).
The therapist in the present study carefully matched the target behaviour(s) of each individual with a central program theme (emotions, with an emphasis on happiness) and to specialised musical instruments to increase the likelihood of specific outcomes (i.e., behaviour change) being achieved. That this approach was effective in changing a range of target behaviours (e.g., hand eye coordination, head movements, and torso position) provides strong evidence of the importance of program planning.It should be noted, however, that the therapist in the study was very experienced and conversant with the variety of instruments utilised. Furthermore, the combination of musical skills, therapy skills and personal relating skills he used enhanced the success of the program. Inexperienced music therapists or therapists with excellent therapy based skills, but limited knowledge of such specialist instruments, may not be so successful.
At the conclusion of the program the participants communicated to the therapist that the approach adopted was very important because it allowed them to be more fully involved and to achieve the targets set. The participants also communicated that the length of the music therapy sessions and the intervention itself were slightly too long, and that at times participation became arduous. Future interventions should therefore address this issue.
In conclusion, the present research adds further support for the effectiveness of music therapy in bringing about significant changes in the behaviour of persons with a disability, in this case Cerebral Palsy.

Aldridge, D. (1998). Music therapy and the treatment of Alzheimer’s disease. Journal of Clinical Geropsychology, 4(1), 17-30.
Barlow, D., & Haynes, S., & Nelson, R. (1986). The Scientist Practitioner, Research and Accountability in Clinical and Educational Settings. Sydney, Australia: Pergamon Press.
Brotons, M. & Pickett-Cooper, P. (1996). The effects of music therapy intervention on agitation behaviors of Alzheimer’s disease patients. Journal of Music Therapy, 33(1), 2-18.
Burkhardt-Mramor, K. (1996). Music therapy and attachment disorder: A case study. Music Therapy Perspectives, 14(2), 77-82.

Crosbie & Sharpley. (1991). DMITSA 2.0 A Statistical Programme for Analysing Data from Interrupted Time-Series. Victoria, Australia: Monash University.

Davis, K. (1998). “To never surrender”: Music therapy in the fight against multiple sclerosis. Canadian Journal of Music Therapy, 6(1), 20-34.

Koger, S., Chapin, K., & Brotons, M. (1999). Is music therapy an effective intervention for dementia? A meta-analytic review of literature. Journal of Music Therapy, 36, 2-15.

Korb, C. (1997). The influence of music therapy on patients with diagnosed dementia. Canadian Journal of Music Therapy, 5(1) 26-54.

Krueger, V. (2000). The use of information technology in music therapy: Behavioural problems in contemporary schools. Nordic Journal of Music Therapy, 9(2), 77-83.

Nayak, S., Wheeler, B., Shiflett, S., & Agostinelli, S. (2000). Effect of music therapy on mood and social interaction among individuals with acute traumatic brain injury and stroke. Rehabilitataion Psychology, 45(3), 274-283.

Pacchetti, C., Mancini, F., Aglieri, R., Fundaro, C., Martignoni, E., & Nappi, G. (2000). Active music therapy in Parkinson’s disease: An integrative method for motor and emotional rehabilitation. Psychosomatic Medicine. 62(3), 386-393.

Suzuki, A. (1998). The effects of music therapy on mood and congruent memory of elderly adults with depressive symptoms. Music Therapy Perspectives, 16(2), 75-80.

Tervo, J. (2001). Music therapy for adolescents. Clinical Child Psychology & Psychiatry, 6(1) 2001, 79-91.

Wager, K. (2000). The effects of music therapy upon an adult male with autism and mental retardation: A four year case study. Music Therapy, 18(2), 131-140.

Zhang, X. & Cuie, K. (1997). Effect of music therapy on chronic schizophrenia. Chinese Journal of Clinical Psychology, 5(1), 48-49.

Copyright © 2001, Southern Development Group, S.A.  All Rights Reserved.
A Private Non-Profit Agency for the good of all, published in the UK & Honduras