Rehabilitation Psychology Copyright 2000 by the Educational Publishing Foundation

2000, Vcl- 45, No. 3, 274-283 0090-5550/00/$5.00 DOI: I0.1037//0090-5550.45.3.274

Effect of Music Therapy on Mood and Social

Interaction Among Individuals With Acute
Traumatic Brain Injury and Stroke
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Sangeetha Nayak
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Kessler Medical Rehabilitation Research and Education Corporation and

University of Medicine and Dentistry of New Jersey

Barbara L. Wheeler
Montclair State University

Samuel C. Shiflett
Kessler Medical Rehabilitation Research and Education Corporation and
University of Medicine and Dentistry of New Jersey

Sandra Agostinelli
Kessler Medical Rehabilitation Research and Education Corporation

ABSTRACT. Objective: To investigate the efficacy of music therapy techniques as

an aid in improving mood and social interaction after traumatic brain injury or
stroke. Design: Eighteen individuals with traumatic brain injury or stroke were
assigned either standard rehabilitation alone or standard rehabilitation along with
music therapy (3 treatments per week for up to 10 treatments). Measures: Pre-
treatment and posttreatment assessments of participant self-rating of mood, family
ratings of mood and social interaction, and therapist rating of mood and participa-
tion in therapy. Results: There was a significant improvement in family members'
assessment of participants' social interaction in the music therapy group relative to

Sangeetha Nayak and Samuel C. Shiflett, Kessler Medical Rehabilitation Research and
Education Corporation, West Orange, New Jersey, and New Jersey Medical School,
University of Medicine and Dentistry of New Jersey; Barbara L. Wheeler, Department of
Music, Montclair State University; Sandra Agostinelli, Kessler Medical Rehabilitation
Research and Education Corporation.
We would like to acknowledge Laurie Hennion and Elisabeth Hayden for their
assistance with music therapy sessions. Support for this research was provided by National
Institutes of Health Grant U24-HD32994.
Correspondence concerning this article should be addressed to Sangeetha Nayak, PhD,
Department of Psychiatry, New Jersey Medical School, University of Medicine and
Dentistry of New Jersey, ADMC 1404, 30 Bergen Street, Newark, New Jersey 07107.
Electronic mail may be sent to nayaksa@umdnj.edu.

274
 Music Therapy 275

the control group. The staff rated participants in the music therapy group as more
actively involved and cooperative in therapy than those in the control group. There
was a trend suggesting that self-ratings and family ratings of mood showed greater
improvement in the music group than in the control group. Conclusions: Results
lend preliminary support to the efficacy of music therapy as a complementary
therapy for social functioning and participation in rehabilitation with a trend toward
improvement in mood during acute rehabilitation.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

A large body of research supports the effectiveness of music therapy within many
areas of physical, cognitive, communicative, social, and emotional rehabilitation
(Standley & Prickett, 1994). The emotional sequelae that frequently accompany
both stroke and traumatic brain injury (TBI) can interfere with an individual's
reentry into the community and ability to obtain maximum benefit from reha-
bilitation. Although there has been little research conducted on the social and
emotional benefits of music and music therapy for individuals with stroke or TBI,
research with a variety of populations has shown that music therapy is associated
with a decrease in depression (Hanser & Thompson, 1994), improved mood
(Thaut, 1989), and a reduction in state anxiety (McKinney, 1990). Music therapy
has also been shown to improve various aspects of social interaction in adults
with and without disabilities, including those with emotional problems (Cassity,
1976) and Alzheimer's disease (Pollack & Namazi, 1992) and their caregivers
(Clair & Ebberts, 1997).
The precise mechanisms underlying these influences are not fully understood
(Behrens, 1988; Hodges, 1980), but there are several theories in the literature
regarding possible reasons for the impact of music on altering emotions and
behavior. Because music is laden with emotional associations and memories, it
provides an effective medium for evoking emotional responses (MacRae, 1992).
Furthermore, the rhythm in music can structure behavior by simultaneously
influencing emotions directly and altering physiological functioning such as heart
rate, muscle tone, blood pressure, and respiration (Sleekier, 1998). Finally, music
therapy sessions provide a comfortable, nonthreatening milieu in which to
encourage successful interaction among patients and between the patient and
therapist. Although reports suggest that music therapy may be useful in changing
mood and improving social skills, there is a paucity of well-controlled research
on TBI or stroke. In a study of individuals with mild TBI, Eslinger, Stauffer,
Rohrbacher, and Grattan (1993) found improved emotional empathy (as reported
by family members and friends but not participants) in those assigned to a
10-week music therapy group but not a social support group. However, they did
not find any significant reductions in depression. In another study of 40 stroke
survivors (Purdie, cited in Marwick, 1996) that compared a music therapy group
(daily music therapy sessions for 12 weeks) with a control group that received
standard care, the individuals in the music group showed signs of being less
depressed and anxious and more emotionally stable.
The primary objective of this study was to evaluate whether music therapy is
 276 Nayak, Wheeler, Shiflett, and Agostinelli

effective as an aid to enhance a patient's mood, social interaction, and involve-

ment in therapy (physical and occupational) during acute rehabilitation. If effec-
tive, music therapy would provide a cost-effective and time-efficient adjunct to
acute rehabilitation.

METHOD
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Participants

Eighteen participants with TBI or stroke were recruited from the inpatient
facility at Kessler Institute for Rehabilitation. Only participants with moderate to
severe impairments at admission, denned by an admission average Functional
Independence Measure (Hamilton, Granger, Sherwin, et al., 1987) score of 4.5
or lower and a depression score of 4 or higher on the 7-point Faces Scale
(McDowell & Newell, 1996) were included. If the patient's self-reported rating
of mood on the Faces Scale did not meet the inclusion cutoff, a family member
was also asked to rate the patient's mood. If the family member rating was 4 or
higher, the individual was eligible to participate in the study. Patients with
nasogastric tubes, intravenous lines, uncontrolled agitated behavior, or any
serious medical conditions were not included. Also, those with sensory, percep-
tual, or marked cognitive impairments (e.g., hearing difficulties or aphasia) that
might have interfered with their ability to participate in the study were not
included. As part of the standard rehabilitation program, all patients are evaluated
for depression at admission and seen by a rehabilitation psychologist once a week
for counseling. All participants in the music and control groups received coun-
seling at least once a week for the duration of the study. The participants ranged
in age from 31 to 84 years, with a mean age of 59.89 years (SD = 16.3). The
sample included 6 men (4 in the music and 2 in the control group) and 12 women
(6 in the music and 6 in the control group). Informed consent was obtained from
all participants and their family members). Participants were assigned to one of
two conditions: music therapy or control.

Measures

All of the measures described subsequently were administered by a research

assistant to participants, their family, and a staff member before the onset of
treatment (Time 1) and again just before participants were discharged from the
hospital (Time 2).
Self-report. Participants rated their own mood on the Faces Scale (McDo-
well & Newell, 1996), a 7-point scale made up of stylized faces. Each face
consists of a circle with eyes that do not change and a mouth that varies from a
smile (upright semicircle) to gloom (inverted semicircle). Participants were asked
to select the face that came closest to expressing how they felt on that day. This
 Music Therapy 277

scale was selected because it is easy to administer and its nonverbal format made
it especially advantageous to use with this group of participants. Test—retest
reliability of the Faces Scale is about .70, and the validity coefficient is .70
(McDowell & Newell, 1996).
Family rating of participant's mood. A 7-point visual analog rating scale
was used to obtain a family member's assessment of the participant's mood. The
scale consists of a horizontal line with the anchors not depressed at all and very
depressed at either end on the top and the numbers 1-7 below the line. Two
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

ratings were obtained: mood in the previous 24 hr and the previous week.
Therapist rating of participant's mood. The physical or occupational ther-
apist who regularly worked with the participant was asked to rate the partici-
pant's mood in the previous 24 hr on a 7-point visual analog rating scale. Higher
scores indicated more depression.
Family rating of social interaction. Seventeen questions adapted from the
social interaction subscale of the Sickness Impact Profile (Bergner, Bobbin,
Carter, & Gilson, 1981) were completed by family members using a 7-point
Likert scale.1 Test-retest reliability and Cronbach alpha coefficients of the
psychosocial dimension of the instrument, from which this scale is extracted, are
.79 and .91, respectively (de Bruin, de Witte, Stevens, & Deideriks, 1992).
Correlations with other scales of depression and adjustment range from .45 to .72
(McDowell & Newell, 1996). A total score was computed for the 17 items.
Lower scores indicated better social interaction-behavior.
Staff rating of participation in therapy. This questionnaire includes three
items in which staff rate participants on a 7-point scale on their level of
involvement in therapy. A physical or occupational therapist who worked closely
with the participant completed this scale. The items addressed cooperation,
motivation, and how actively involved the participant was in therapy; higher
scores represented better outcomes.

Design and Procedure

A between-groups, repeated measures design was used. Participants were

assigned to either a treatment condition (music therapy along with standard
rehabilitation) or a control condition (standard rehabilitation alone). The goal of
random assignment was not fully achieved because we attempted to treat par-
ticipants in the music therapy condition in groups of 2 to 3 people to provide
them with an opportunity for social interaction. This process required that 2 or
more patients available at the same time be assigned to the same condition, thus
compromising the randomization process. Despite the attempts to have patients

1
Examples of items are "My family member does not show interest in conversation
while people are visiting." "My family member talks less with those around her/him," and
"My family member shows less affection."
 278 Nayak, Wheeler, Shiflett, and Agostinelli

in groups, on a few occasions participants received treatment individually (with

only the music therapist present) in one or more sessions as a result of the
occasional unexpected lack of availability of other members of the group.
Music therapy plus standard rehabilitation. Ten participants in this group
met two or three times a week for the duration of their stay in the hospital and
received up to a maximum of 10 treatment sessions in addition to their standard
rehabilitation (M = 5.9, SD = 1.73; range: 4-10). A variety of music therapy
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

procedures was used to enhance mood and social interaction. The specific music
This document is copyrighted by the American Psychological Association or one of its allied publishers.

activities used, and their selection, were based on the needs of the group and are
typical of music therapy practice.
Each session began with an opening song or activity designed to set the mood
and to help participants become involved in the session. This was frequently
some type of instrumental improvisation in which participants used simple
percussion and melodic instruments along with the therapist. In some sessions,
participants were asked to play together in a manner that expressed how they
were feeling at that time. This was accomplished by having each group member
select a simple percussive instrument from among a collection. A number of
sessions involved simple pitched instruments (bells or chimes); each participant
was assigned several pitches—instruments and played his or her pitch when cued.
Instruments included drums, tambourines, maracas, xylophones, and tone bars
(each bar is an individual note, beaten with a stick to produce the sound). The
therapist would then structure the improvisation by asking participants to play
their instruments to express how they were feeling or first asking a participant to
describe how he or she was feeling and then asking that member and the rest to
play their instruments in a manner reflecting their mood. Both of these ap-
proaches were followed by a brief discussion of how well the musical improvi-
sation reflected and supported the feelings being experienced. In other sessions,
participants were asked to simply play together and then later cued verbally to
listen to one another as they played or took turns playing musical leadership or
supportive roles.
One or two additional music therapy activities, such as singing, composing,
playing instruments, improvising, performing, and listening, formed the core of
each group. The specific musical activities were chosen on the basis of the
interests and abilities of the participants, and consideration was given as to
whether the activity could be used to address the goals of enhancing mood and
social interaction. The music therapist provided physical cues or participants read
from a color- or letter-coded chart. Another activity that was used frequently was
composition, accomplished by substituting words of participants' choice for the
normal words of a song. These compositions helped participants to express how
they were feeling or what they were thinking. Singing was frequently incorpo-
rated. Each group of people had favorite songs, most of which were on song
sheets or charts. All activities involved verbal processing, including some focus
on mood. The amount and quality of verbal processing varied depending on the
abilities and needs of the participants.
 Music Therapy 279

Standard rehabilitation control condition. Eight participants were assigned

to a standard rehabilitation control group. This group received all of the standard
therapies that are part of the inpatient rehabilitation regimen.

RESULTS

Analyses of covariance (ANCOVAs) were conducted to examine the effect of

This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

treatment group on each of die outcome measures. Because large mean differ-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

ences were observed at admission on some measures, scores at Time 1 were used
as covariates on all of the tests comparing the music therapy and control groups.
Gender was also entered as a covariate to reduce error variance attributable to
gender. However, before the ANCOVAs were conducted, the homogeneity of
covariate regression on each outcome measure was tested between the two
treatment conditions via F tests. If regression slopes were dissimilar in the two
groups, ANCOVAs were not performed, and beta coefficients are reported and
discussed. Means and standard deviations on all outcome measures are shown in
Table 1. Alpha levels were set at .05; however, because of the small sample size
and the exploratory nature of this study, trends at the .10 level are also discussed.
The test for equality of slopes indicated a significant difference between the two
groups on the social interaction scale, F(l, 14) = 8.34, p < .02. The relations (beta
values) between preintervention and postintervention scores were .93 in the control
group and .76 in the music therapy group (p < .02), indicating that the more
impaired a participant's social behavior at the outset, the more likely he or she was
to benefit from music therapy. The occupational and physical therapists reported, at
the conclusion of the interventions, that the music group was significantly more
involved in therapy, F(l, 14) = 8.64, p < .01, and tended to be more motivated to
participate, F(l, 14) = 4.12, p = .06, than the control group. There was no
statistically significant difference between the groups in terms of staff-perceived
cooperativeness during therapy sessions. The main effect for treatment showed
trends on three of the four mood measures: self-report, F(l, 14) = 3.27, p < .10;
family rating of mood in the previous 24 hr, F(l, 14) = 3.39, p < .10; and family
rating of mood in the previous week, F(l, 14) = 4.52, p < .06. On the fourth mood
measure, staff rating of participant mood, the test for equality of slopes indicated a
trend suggesting a difference between the two groups (F = 3.46, p < .10). The
relations (beta values) between preintervention and postintervention scores were .88
in the control group and -.15 in the music therapy group (p < .09). This trend
suggests that the music group did not benefit from treatment, whereas those in the
control group showing greater levels of impairment at baseline appeared to improve

DISCUSSION

When differences between groups on pretreatment scores were statistically

controlled, several notable treatment effects emerged, indicating a positive effect
 Nayak, Wheeler, Shifiett, and Agostinelli

„ in "n "n m »n >n

8
S -r ^- ^ ^r -: ^r
""

1
* * # di di *
*
*
g 1*, r- o\ CN CN ts TT
0 CD CD ^ CS ^ 00
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

s
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Q CD ^ C ^ S c D 0 § C D
3
04 to —; ^ T^* o oc i— ' o o
04

1 e
H oo o o i n o o o c c D i n o o
£ oo oo oc t^ oo oc v o r ^ o q
1
g II CD C D T j - C D o i •^f'^t'' ^
Tf
c
_0 ^ 31

1
y
1
c Q
^ —H ON ^ ""^ OO in ^O 1
60 -3 - = d d K o d d
1
d-
u _,

fO
i
d
T3
.§ 73
M

c"
H
S OO CD
oo i—"
CD
\D
O
«n
00 O
oo o
CD
-^ J5

o OJ £
•a
5 -| §

S > ' '

S c>i O in f""i o (N oo o
Q
to 1 t
^H d d d ^ t o d d
3 CN

1 § f |

"o" s H o o o o o o o o M |
oo \o in cs ^ cs i~~ ^o
<s c4 m nf oj uS «S uS
1 II ^ u .3
1§
iS 'S ^^
en ^
C •% — oo oo r~ *n o\ oo o> 1 "S
,o Q r~- ^ j - ^ . 0 s i n o o o t - -
to ^-' T-: ,—' d rS o d d
s 1 _^
•o 1
^ u

1
OJ op
1 g g
£j
H
la S § ° j S S 2 g g
^ CD |
1
55

1 H .j M *-
t
1 I S
U

s 1 v
i -t | g "1 |

1 |l|lj||llf||l|.|
11*.
ri s V 3 E S <E 6 .S 11 g^<- S..213'™ 1 |2
1
u ot/i[i4 t o ' o f i u S j
1
 Music Therapy 281

of music therapy on social interaction and participation in therapy and trends

suggesting some improvement in mood. The improvement in mood among
participants in the music group approached statistical significance on three of the
four mood measures: self-report (p < .10), family rating of mood during the
previous 24 hr (p < .10), and family rating of mood during the previous week
(p < .06). Family members of those in the music therapy group reported a
significant improvement in social interaction (relative to the control group)
among participants who were more impaired at the outset (p < .02). These
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

social behaviors could have been facilitated by the very nature of music therapy,
which builds on the social aspects of music and musical participation. In
addition, music therapy sessions in this study focused on encouraging social
interaction through providing activities that promoted musical and verbal inter-
action and through verbal processing of what occurred in the sessions.
The mood outcome is consistent with the report of Purdie (cited in Marwick,
1996), who found a decrease in depression in patients with stroke after 12 weeks
of music therapy. However, these results are unlike those reported by Eslinger et
al. (1993), who found that there was no mood benefit for those with mild TBI.
The difference between the samples in Eslinger et al.'s study and ours is that our
sample included individuals in an acute rehabilitation setting. It is possible that
music therapy has a more pronounced effect on mood early on after injury rather
than later, a notion that will have to be confirmed with further research.
The music therapy group was seen by staff as more actively involved (p <
.01), and there was a trend indicating that the music group was more motivated
to participate in therapy (p < .06), suggesting that alterations in mood had a
positive effect on participation in the rehabilitation process. No differences
emerged on staff ratings of cooperation.

Limitations

The results of this study support the effectiveness of music therapy in

improving mood and social interaction among people who have had a stroke or
TBI. However, this study involves some limitations due to several unexpected
problems. Substantial difficulties in recruiting, irregular attendance, and early
discharge from the hospital contributed to imperfections in the research design.
As indicated earlier, the goal of random assignment was not fully achieved
because we attempted to treat participants in the music therapy condition in
groups of 2 to 3 people to provide them with an opportunity for social interaction.
In addition, the number of individuals participating in a treatment session varied
because some participants were unexpectedly unavailable for one or more of the
scheduled group sessions, usually as a result of feeling too tired or ill to
participate or because of an unexpected visit by family members.
The apparent benefits of music therapy may have resulted from the effects of
increased social interactions provided by, if not required by, a group session
involving several other patients. Because of the recruitment and scheduling
 282 Nayak, Wheeler, Shiflett, and Agostinelli

difficulties outlined earlier, a comparison intervention (i.e., art therapy) that

would have controlled for the effects of attention and participation in a group had
to be eliminated. Future research needs to explore the extent to which the
particular content or type of therapy contributes to or facilitates the effects found
here.

Conclusion
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

The results of this study indicate that music therapy had a positive effect on
social and behavioral outcomes and showed some encouraging trends with
respect to mood. In particular, it appears that these effects facilitate participation
in the standard rehabilitation process. These findings have relevant clinical
implications for the rehabilitation psychologist. Because mood and social func-
tioning are frequently goals in acute rehabilitation for patients with brain injuries
and stroke, music therapy could provide an effective means for addressing these
goals.
Rehabilitation psychologists might consider encouraging increased inclusion
of music therapists on the interdisciplinary treatment team. Given the needs of
individuals in acute rehabilitation for improved mood and social interaction and
the preliminary findings of this study, the skills of the music therapist could
complement those of the psychologist. Psychologists might consider conducting
groups with music therapists serving as co-therapists, with the music therapist
using the nonverbal qualities of music and the structure and pleasurable aspects
of this type of therapy to facilitate interaction and the psychologist using this
interaction and setting to address common goals (e.g., developing social skills
and coping mechanisms, minimizing emotional distress, and working on com-
munity reintegration). These groups could serve as a complement to goals being
addressed in individual sessions and could help patients transfer what they learn
in therapy to other contexts. Introducing music, a structured but emotionally
expressive medium, into the acute rehabilitation treatment setting would allow
rehabilitation psychologists access to a cost-effective, pleasant, and entertaining
adjunct to traditional psychotherapy.

REFERENCES

Behrens, G. A. (1988). An objective approach to the expression of feelings.

Music Therapy Perspectives, 5, 16—22.
Bergner, M., Bobbitt, R. A., Carter, W. B., & Gilson, B. S. (1981). The Sickness
Impact Profile: Development and final revision of a health status measure.
Medical Care, 19, 787-805.
Cassity, M. (1976). The influence of a music therapy activity upon peer accep-
tance, group cohesiveness, and interpersonal relationships of adult psychi-
atric patients. Journal of Music Therapy, 13, 66-76.
 Music Therapy 283

Clair, A. A., & Ebberts, A. G. (1997). The effects of music therapy on interac-
tions between family caregivers and their care receivers with late stage
dementia. Journal of Music Therapy, 34, 148-164.
de Bruin, A. F., de Witte, L. P., Stevens, F., & Deideriks, J. P. M. (1992).
Sickness Impact Profile: The state of the art of a generic functional status
measure. Social Science and Medicine, 35, 1003-1014.
Eslinger, P., Stauffer, J. W., Rohrbacher, M., & Grattan, L. M. (1993). Music
therapy and brain injury (Report to the Office of Alternative Medicine).
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Bethesda, MD: National Institutes of Health.

Hamilton, B. B., Granger C. V., Sherwin, F. S., et al. (1987). In M. J. Fuhrer
(Ed.), Rehabilitation outcomes analysis measurement (pp. 137—147). Bal-
timore: Paul H. Brookes.
Hanser, S. B., & Thompson, L. W. (1994). Effects of a music therapy strategy on
depressed older adults. Journal of Gerontology, 49, 265-269.
Hodges, D. A. (Ed.). (1980). Handbook of music psychology. Washington, DC:
National Association for Music Therapy.
MacRae, A. (1992). Should music be used therapeutically in occupational ther-
apy? American Journal of Occupational Therapy, 46, 275-277.
Marwick, C. (1996). Leaving concert hall for clinic, therapists now test music's
"charms." Journal of the American Medical Association, 275, 267-268.
McDowell, I., & Newell, C. (1996). Physical disability and handicap. In Mea-
suring health: A guide to rating scales and questionnaires (2nd ed., pp.
194-198). New York: Oxford University Press.
McKinney, C. H. (1990). The effect of music on imagery. Journal of Music
Therapy, 27, 34-46.
Pollack, N. J., & Namazi, K. H. (1992). The effect of music participation on the
social behavior of Alzheimer's disease patients. Journal of Music Therapy,
29, 54-67.
Standley, J. M., & Prickett, C. A. (1994). Research in music therapy: A tradition
of excellence. Silver Spring, MD: National Association of Music Therapy.
Sleekier, M. A. (1998). The effects of music on healing. Journal of Long-Term
Home Health Care, 17, 42-48.
Thaut, M. H. (1989). The influence of music therapy interventions on self-rated
changes in relaxation, affect, and thought in psychiatric prisoner-patients.
Journal of Music Therapy, 26, 155—166.

Received June 9, 1999

Revision received December 1, 1999
Accepted December 7, 1999