Aging & Mental Health, May 2007; 11(3): 330–338

ORIGINAL ARTICLE

An investigation of long-term effects of group music therapy on

agitation levels of people with Alzheimer’s Disease

ALISON J. LEDGER & FELICITY A. BAKER

University of Queensland, Australia

(Received 10 February 2006; accepted 16 June 2006)

Abstract
This study aimed to investigate the long-term effects of group music therapy on agitation manifested by nursing home
residents with Alzheimer’s disease. A non-randomised experimental design was employed with one group receiving weekly
music therapy (n ¼ 26) and another group receiving standard nursing home care (n ¼ 19). Agitation levels were measured
five times over one year using the Cohen-Mansfield Agitation Inventory (Cohen-Mansfield, J. (1989). Agitation in the
elderly. In N. Billig & P. V. Rabins (Eds.), Issues in geriatric psychiatry (pp. 101–113). Basel, Switzerland: Karger). Although
music therapy participants showed short-term reductions in agitation, there were no significant differences between the
groups in the range, frequency, and severity of agitated behaviours manifested over time. Multiple measures of treatment
efficacy are necessary to better understand the long-term effects music therapy programs have on this population.

Introduction
Agitation leads to severe disturbances in the quality (Mahoney, Volicer, & Hurley, 2000; Opie,
of life of people with Alzheimer’s disease and Rosewarne, & O’Connor, 1999).
their carers (Cohen-Manfield, 1986, 1999; Foli & Music therapy is one intervention which aims to
Shah, 2000; Marin et al., 1997). Estimates of the create secure, stimulating environments, to meet
prevalence of agitation vary depending on the social and emotional needs, and to reduce agitation
terminology, assessment tool, population, and displayed by people with Alzheimer’s disease and
setting employed (Colenda, 1995; Tractenberg, other types of dementia (Brotons, Koger, &
Weiner, & Thal, 2002), ranging from 24% to 98% Pickett-Cooper, 1997; Nugent, 2002). Short-term
of people diagnosed with dementia (Gruber- reductions in agitated behaviour have been reported
Baldini, Boustani, Sloane, & Zimmerman, in music therapy case studies and research under-
2004; Sourial, McCusker, Cole, & Abrahamowicz, taken with people with dementia (Brotons &
2001). Pickett-Cooper, 1996; Fitzgerald-Cloutier, 1993;
People with Alzheimer’s disease show a range of Groene, 1993; Nugent, 2000; Ridder, 2003;
agitated behaviours, including repetitive acts, behav- Suzuki et al., 2004; Thomas, Heitman, &
iours inappropriate to social norms, and aggressive Alexander, 1997). Explanations for the success of
behaviours towards self or others (Cohen-Mansfield music therapy in reducing agitation include the
& Martin, 1999). The progression of a person’s effects of music on attention (Baker, 2002; Ridder,
agitated behaviours is difficult to predict. Although 2003) and quality of life components (Ruud, 1998).
agitated behaviours are likely to persist and increase Familiar music may serve to regulate a person’s
in severity over time, they do not worsen system- arousal to a moderate level (Baker, 2002; Ridder,
atically (Eustace et al., 2002; Haupt, Kurz, & Jänner, 2003), or redirect a person’s attention from
2000; Levy et al., 1996). Instead, agitated behav- misleading or confusing stimuli (Gerdner, 1999).
iours are more episodic in nature, fluctuating over A person with dementia may then be more able to
time. Possible causes of agitation include interpret his or her environment and any fear
under-stimulating or demanding environments, or or agitation may be lessened. Success in singing,
a person’s reduced ability to communicate and have playing instruments, moving to music, or sharing
his/her needs met (Cohen-Mansfield & Martin, memories or views related to music, may also meet
1999). Interventions that address these possible a person’s unmet needs for self-expression, achieve-
causes of agitation are therefore imperative ment, and meaning in life (Clair & Bernstein, 1990;

Correspondence: Alison Ledger, Clinical Lecturer in Music Therapy, School of Music, University of Queensland,
Australia 4072. Tel: þ61 7 3365 4146. Fax: þ61 7 3365 4488. E-mail: a.ledger@uq.edu.au
ISSN 1360-7863 print/ISSN 1364-6915 online/07/030330–338 ß 2007 Taylor & Francis
DOI: 10.1080/13607860600963406
 Long-term effects of music therapy on agitation 331
Pollack & Namazi, 1992; Prickett & Moore, 1991; The chosen nursing homes were either community,
Ridder, 2003; Rio, 2002; Short, 1995; church, or privately operated and offered a mixture
Tomaino, 2000). Music therapy groups may espe- of high and low nursing care. Each nursing home
cially promote feelings of belonging among partici- provided residents with access to activities such as
pants with dementia (Ebberts, 1994; Pollack & outings, concerts, arts and crafts, games, cooking,
Namazi, 1992; Rio, 2002). As agitation is viewed or gardening.
as an expression of feelings of despair, frustration,
boredom, or loneliness (Cohen-Mansfield & Martin, Participants
1999), music therapy may reduce agitated
behaviour by lessening such feelings (Ashida, 2000; A sample of 60 participants (30 experimental and
Bright, 1997; Groene, 1993; Pollack & Namazi, 30 control) was required to detect significant
1992; Rio, 2002; Short, 1995). differences between the experimental and control
Although research indicates that music is an groups (p50.05, power of 80%). Sixty-eight were
effective intervention for reducing agitated behav- initially recruited to allow for attrition. Inclusion
iours, there are weaknesses and limitations to criteria comprised:
existing studies, including small sample sizes and (a) Primary diagnosis of Alzheimer’s type dementia.
within-participants designs (Brotons et al., 1997; (b) Stage 4, 5, or 6 on the Global Deterioration
Koger, Chapin, & Brotons, 1999; Lou, 2001; Scale (GDS) for assessment of primary degen-
Nugent, 2002; Sheratt, Thornton, & Hatton, erative dementia (Reisberg, Ferris, De Leon, &
2004b; Ridder, 2005; Vink, Birks, Bruinsma, & Crook, 1982).
Scholten, 2003). Furthermore, no studies have (c) Cognitive impairment evident upon
specifically tested whether music therapy can have Mini-Mental State Examination (MMSE523.
long-term influences on the agitation levels of people Folstein, Folstein, & McHugh, 1975) or Mental
with dementia (Ledger & Baker, 2005). This study Status Questionnaire administration (MSQ > 2
aimed to address the limitations of earlier research, errors; Kahn, Goldfarb, Pollack, & Peck, 1960).
by employing a large sample size and a control
group, and investigating the possibility that music Due to the high prevalence of agitation among
therapy has long-term effects on people with people with moderate to severe dementia
Alzheimer’s disease. The study specifically tested (Sourial et al., 2001), presence of agitation was not
whether music therapy participants with Alzheimer’s specified as an inclusion criterion. Nursing home
disease manifested fewer, less frequent, or less severe residents were to be excluded if they were in poor
types of agitated behaviours over time than people health (indicating that they were unlikely to com-
who were not receiving music therapy. plete the year-long study). Although variability in
demographic characteristics, psychotropic medica-
tion use, levels of pain, depression and fatigue,
quality of staffing, quality of relationships with
Method
others, and past experiences of stress were consid-
Design ered likely to influence participants’ agitation levels,
The study employed a longitudinal repeated mea- it was expected that these variables would be
sures design with an experimental and control group. balanced between the experimental and control
Participants were studied over a one-year period, this participants. Although these variables were not
being appropriate to detect agitation changes in the controlled, data pertaining to these variables was
participants (Levy et al., 1996; Weiner et al., 1998). collected and the influence of these variables
Experimental group participants were offered weekly was explored.
group music therapy treatment, while control group Participants were allocated to either the experi-
participants received their usual nursing and therapy mental or control group based on their current place
care. This design was approved by one of the human of residence (convenience sample). The five nursing
ethics committees of the University of Queensland in homes assigned to the experimental condition were
accordance with the National Health and Medical all within Brisbane, as this condition required weekly
Research Council’s guidelines. visits from a registered music therapist. Consent for
68 eligible participants was obtained, however
eight did not participate in the study because:
Settings
(a) five participants in the experimental group chose
Thirteen nursing homes in Queensland (12) and not to participate in music therapy; (b) one partici-
Victoria (1) participated in the study, two of which pant in the control group moved to a different nursing
already offered music therapy services to its nursing home and subsequently began receiving music
home residents. Nursing homes within organizations therapy; (c) one participant died before receiving
known to show interest in music therapy services the first session of music therapy; (d) one participant
were selected based on location (i.e. within reason- became distressed in music therapy and the service
able travelling distance from research personnel). discontinued after she received five sessions.
332 A. J. Ledger & F. A. Baker
Table I. Characteristics of participants.

Characteristic Experimental group Control groupy

Age 71–75 3 0
76–80 4 3
81–85 6 7
86–90 9 6
91–95 3 2
96–100 1 1
Gender Male 4 1
Female 22 18
GDS stage 4 4 2
5 14 10
6 8 7
Level of cognitive impairment Mild 4 1
Moderate 7 6
Severe 15 10
Not recorded 0 2
Medications Acetylcholinesterase inhibitors 3 5
Antipsychotic 11 4
Antidepressant 7 8
Antianxiety 3 2
Narcotic 1 1
Social Support Once/week or more 21 11
Less than once/week 5 7
Not recorded 0 1
yExperimental n ¼ 26; Control n ¼ 19; Total N ¼ 45

Of the 60 participants who began the study, only brain injury, cerebral aneurysm, and epilepsy.
45 participants completed all five testing phases. Twenty-two experimental and seventeen control
One participant from the experimental group left the group participants were assessed as in stages five or
study when she moved away and ceased receiving six (moderate or moderately severe Alzheimer’s
music therapy treatment. A further 14 participants disease) on the GDS and the remainder
(three from the experimental group and eleven from (four experimental, two control) were assessed as
the control group) passed away in the course of the in stage four (mild Alzheimer’s disease) of the GDS.
year’s research. Due to the highly individual, Most participants (experimental n ¼ 22, control
fluctuating, and unpredictable nature of agitation n ¼ 16) showed moderate to severe cognitive impair-
(Levy et al., 1996), only the data from those ment on either the MMSE or MSQ. A greater
participants who completed the study were included proportion of the experimental participants (42%
in the analysis. of experimental participants versus 21% of control
Characteristics of the 45 participants (26 experi- participants) had been prescribed antipsychotic
mental, 19 control) who completed the study are medication, which may have contributed to
displayed in Table I. The participants ranged from lower baseline agitation levels (Weiner &
71 to 96 years of age (experimental group Schneider, 2003).
M ¼ 84.81, SD ¼ 6.62; control group M ¼ 85.63,
SD ¼ 5.22) and most were female. Three partici- Intervention
pants in the experimental group and none in the
control group were from non-English speaking Participants in the experimental group participated
backgrounds. About three quarters of the partici- in weekly sessions (30–45 minutes) of group music
pants (21 experimental, 12 control) had contact with therapy for at least 42 weeks within a year. Research
family or friends at least once per week. None of the suggests that the effects of music therapy on
participants had received music therapy treatment agitation can be noted with one-to three-weekly
prior to the commencement of the research. sessions (Brotons & Pickett-Cooper, 1996; Clair &
Experimental group participants who completed Bernstein, 1990; Ebberts, 1994; Jennings & Vance,
the study attended a mean number of 44.65
3.1 2002; Olderog Millard & Smith, 1989;
music therapy group sessions during the year of Sambandham & Schirm, 1995; Suzuki et al.,
research. 2004). Sessions were held at consistent times in
Some participants had neurological or psychiatric accordance with residents’ existing routines–three of
disorders alongside dementia of the Alzheimer’s type the groups took place in the morning and two
(experimental n ¼ 9, control n ¼ 6), such as depres- of the groups took place in the mid afternoon.
sion, schizophrenia, Parkinson’s Disease, stroke, Groups consisted of two to ten research participants
 Long-term effects of music therapy on agitation 333
at any one time. Group music therapy was preferred (Finkel, Lyons, & Anderson, 1992; Miller,
over individual therapy as this intervention is Snowdon, & Vaughan, 1995; Shah, Evans, &
reported in music therapy literature as effective in Parkash, 1998; Sourial et al., 2001; Snowdon,
addressing people’s unmet needs for interaction and Miller, & Vaughn, 1996; Weiner et al., 2000).
belonging (Christie, 1995; Ebberts, 1994; Pollack & As research personnel (the first author and the
Namazi, 1992; Rio, 2002; Silber & Hes, 1995) and research assistant) were only present during music
reducing agitation during and immediately post- therapy sessions, they interviewed nursing staff to
session (Lesta & Petocz, 2006). Participants who obtain more comprehensive CMAI assessments for
resided at nursing homes allocated to the control the research participants. Research personnel
condition received their usual nursing and therapy recorded nurses’ responses on a tick-sheet. Due to
care over the one-year period. the high turnover of nursing staff in the participating
Four of the music therapy groups were conducted nursing homes, it was not always possible to inter-
by the first author (a qualified music therapist) and view the same staff at each of the five time points.
one group was conducted by a research assistant As the goal of the CMAI is ‘‘to achieve the most
(also a qualified music therapist). The musical accurate reflection of the frequency at which these
content of the music therapy sessions was decided behaviours occurred’’ (Cohen-Mansfield, 1991,
by the treating music therapist and varied depending p.4), research personnel interviewed the nurse on
on participants’ assessed needs, abilities, back- duty who had the most frequent contact with
grounds, and musical preferences. Each group research participants during the previous two
had a similar overall structure (greetings-main weeks. An average of three nursing staff were
section-song requests-farewells) and sessions interviewed at each nursing home over the one-year
included techniques that aim to promote period.
self-expression, control, mastery, belonging, and
purpose (Brotons & Pickett-Cooper, 1996; Clair & Analysis
Bernstein, 1990; Ebberts, 1994; Groene, Zapchenk,
Total CMAI scores and subtype scores for each
Marble, & Kantar, 1998; Hanson, Gfeller,
participant at each of the five time points were
Woodworth, Swanson, & Garand, 1996; Prickett &
calculated by assigning values to the frequency
Moore, 1991; Rio, 2002). Participants joined in
ratings for each behaviour (‘‘Never’’ ¼ 0 through to
sessions through listening to music played by the
‘‘Several times an hour’’ ¼ 6). Means and
therapist, choosing or requesting favourite songs,
standard deviations of the experimental and control
guessing song-titles from melodic/lyric clues, sing-
groups’ total and subtype CMAI scores were
ing, playing instruments, moving to music, and
then determined for each time point. A repeated
discussing feelings and memories.
measures multivariate analysis of variance
was employed to test for differences between the
Measures experimental and control groups in the range,
Participants’ agitation levels were assessed using the frequency and severity of agitated behaviours
CMAI-long form (Cohen-Mansfield, Marx, manifested over time.
Rosenthal, 1989) prior to the commencement of the
experimental or control period, and then at three, six, Therapist’s log
nine, and twelve months following commencement of In addition to the CMAI measures, the two qualified
the experimental or control period. The CMAI music therapists conducting the clinical interven-
measures the frequency of 29 agitated behaviours in tions kept a log of observed changes in participants’
the two weeks prior to the CMAI assessment. CMAI agitation from pre to post session using the CMAI
scores can be summarized according to four behav- behaviours as descriptors. Only noticeable changes
iour subtypes – verbal non-aggressive behaviour (e.g., were recorded.
repeated unwarranted requests for attention such as
calling out), verbal aggressive behaviour (e.g., cur-
sing, threatening, or insulting language, verbal sexual Results
advances), physical non-aggressive behaviour
Range and frequency of behaviours
(e.g., repetitive behaviours such as wandering,
tapping, or repeated attempts to undress inappropri- The range and frequency of agitated behaviours are
ately), and physical aggressive behaviour (e.g., reflected in the participants’ mean total CMAI
destroying objects or property, throwing objects, scores, graphed in Figure 1 for the five data
grabbing, pushing people) (Cohen-Mansfield, 1999; collection points. Initial CMAI scores were signifi-
Cohen-Mansfield & Martin, 1999). CMAI scores cantly higher (t ¼ 2.17, p50.05) for the control
correlate significantly with scores on other measures group and both groups showed large standard
of agitation, and demonstrate high internal deviations before the commencement of the
consistency reliability (r  0.82), test-retest reliability music therapy intervention (control group: CMAI
(0.97), and high inter-rater reliability (r ¼ 0.8) M ¼ 39.05, SD ¼ 22.15; experimental group: CMAI
334 A. J. Ledger & F. A. Baker
M ¼ 25.92, SD ¼ 15.42). Total CMAI means for or control; F ¼ 1.61; p ¼ 0.432). This indicated that
both groups fluctuated from one data collection there were no significant differences between the
point to the next. The control group showed a large experimental and control groups in the range and
decrease between the baseline and first reassessment frequency of agitated behaviours manifested
of CMAI, and then a large increase at the next over time.
reassessment (time point 3) before a decline in
agitation at the end of the study (time point 4 and 5).
The experimental group showed an increase
between the baseline and first reassessment of Severity of behaviours
CMAI, followed by a small decrease at the next The means and standard deviations for
CMAI reassessment (time point 3), then an increase CMAI subtype scores (verbal aggressive, verbal
in agitation (time point 4), before returning to an non-aggressive, physical aggressive, and physical
agitation level similar to time points 2 and 3. Control non-aggressive) across the five time points are
group participants showed their highest degree of shown in Table II. In the experimental group, the
agitation at time point 3, but the experimental group course of agitation appeared to remain more stable
showed their highest degree of agitation later, at time for verbal aggressive behaviour (Figure 2) than for
point 4. the other subtypes and are therefore focused on in
The repeated measures multivariate analysis of this paper. However, the repeated measures multi-
variance revealed significant effects within-partici- variate analysis of variance revealed no significant
pants over time (F ¼ 2.61; p50.05), but not
differences between the groups in the manifestation
within-participants over time by group (experimental
of any of the four different types of agitation over
time (verbal non-aggressive F ¼ 0.33, p ¼ 0.57;
45
verbal aggressive F ¼ 0.59; p ¼ 0.45; physical
non-aggressive F ¼ 0.62, p ¼ 0.44; physical aggres-
Experimental
Control sive F ¼ 0.78, p ¼ 0.38). The experimental group did
40
not manifest a significantly lesser degree of
non-aggressive behaviour, nor the more severe
CMAI score

35
aggressive behaviour, over time.
Univariate test results revealed a significant effect
30 of the variables time and group for verbal aggressive
behaviour (F ¼ 2.70; p50.05). It is therefore possi-
25 ble that the experimental group manifested less
verbal aggressive behaviour over time, only the effect
20 size was too small to be detected through multi-
1 2 3 4 5 variate analysis. This interpretation is supported by
Time-point evidence that the experimental group’s levels of
Figure 1. Mean total CMAI scores for the experimental and verbal aggressive behaviour fluctuated less over time
control groups. (Figure 2).

Table II. CMAI subtype means and standard deviations.

Verbal non-aggressive Verbal aggressive Physical non-aggressive Physical aggressive

Time Pty Group M SD M SD M SD M SD

x
CMAI 1 Exp 7.65 4.31 4.12 4.04 10.35 10.48 3.81 3.79
Control 11.11 4.90 7.21 4.69 14.89 10.39 5.84 8.45
Total 9.11 4.83 5.42 4.55 12.27 10.57 4.67 6.20
CMAI 2 Exp 9.92 5.68 4.88 5.01 13.08 10.68 4.81 6.91
Control 9.68 5.56 4.21 3.90 12.89 8.76 4.37 7.91
Total 9.82 5.57 4.60 4.53 13.00 9.81 4.62 7.26
CMAI 3 Exp 10.54 4.71 4.62 4.12 12.73 11.09 3.27 3.58
Control 11.95 6.44 6.11 4.81 16.47 9.41 7.26 8.13
Total 11.13 5.48 5.24 4.43 14.31 10.47 4.96 6.19
CMAI 4 Exp 12.04 5.65 4.92 4.42 15.46 11.81 5.54 7.14
Control 11.61 6.33 5.29 4.25 13.87 10.07 5.65 6.78
Total 11.86 5.88 5.08 4.30 14.79 11.02 5.59 6.91
CMAI 5 Exp 9.42 5.85 5.23 5.07 3.85 4.79 11.23 8.99
Control 8.58 6.93 5.21 4.89 5.74 9.60 11.53 10.66
Total 9.07 6.26 5.22 4.94 4.64 7.19 11.36 9.62
y
Time Pt ¼ time point. xExp ¼ experimental group. Experimental n ¼ 26; Control n ¼ 19; Total N ¼ 45.
 Long-term effects of music therapy on agitation 335
8 familiar music (Clair & Bernstein, 1994) are no
Experimental longer present.
7 Control
The finding that music therapy participants
6
showed only small, steady increases in verbal
aggressive behaviour over time could be linked to
CMAI score

4
music therapy’s success in maintaining participants’
expressive capabilities. During group sessions, music
3 therapy participants shared memories, joked, greeted
2 and complimented others, and commented on the
music, their achievements, and their experiences
1
within sessions. Participants also voiced concerns
0 regarding aging and losses in health, beauty,
1 2 3 4 5
independence and possessions. Some spoke of
Time-point
anger towards family members and nursing staff.
Figure 2. Mean CMAI scores for verbal aggressive behaviour. The treating music therapists noted that few music
therapy participants exhibited declines in verbal
communication skills during the course of the year
Therapists’ observations and improved language functioning has been
reported as an outcome in previous studies of
Prior to sessions, the therapists noted that most
participants wandered, fidgeted, grabbed or insulted music interventions for people with dementia
others, yelled, complained, made anxious statements (Brotons & Koger, 2000; Quinn, 2003;
(e.g., about others stealing), or asked repetitive Sambandham & Schirm, 1995; Smith, 1986;
questions (e.g., asking to ‘‘go home’’). These Suzuki et al., 2004). The verbal non-aggressive
behaviours were seen considerably less frequently behaviour of the music therapy participants steadily
within music therapy sessions and in many cases, increased over the first nine months of the study, also
participants remained less agitated immediately after supporting an improvement in language functioning.
sessions. The control group’s degree of verbal aggressive
and verbal non-aggressive behaviour decreased in
the final six months of the study. This could indicate
Discussion that control participants were losing the ability to
express agitation in a verbal form. Koss et al. (1997)
The results of this study showed interesting trends attributed limited verbal agitation to increasing
with regard to the longitudinal effects of weekly language difficulties in participants in the lower
music therapy sessions on the agitated behaviours of MMSE stratum. Through exercising verbal skills
people with Alzheimer’s disease. First, there were no within music therapy groups, experimental group
significant differences between the two groups in the participants may have sustained abilities that dete-
range and frequency of agitated behaviours mani- riorated in the control group during the course of
fested over time. Furthermore, there were no the study.
significant differences between the groups in the
manifestation of any of the four different agitation
Limitations of the study
subtypes over time.
The finding that there were no significant differ- High inter- and intra-participant variability in agita-
ences between the groups over time may indicate tion levels may account for the lack of significance
that music therapy has only immediate effects on found between the two conditions. The pooled
agitated behaviours displayed by people with sample’s standard deviation values for the
Alzheimer’s disease (rather than cumulative and total CMAI scores were large (Table III) and total
long-term effects). The treating therapists’ observa- CMAI scores fluctuated substantially from one time
tion that music therapy treatment led to short-term point to the next (Figure 1).
reductions in agitation is supported by the findings Furthermore, the control group showed a greater
of other music therapy studies which examined degree of agitation at baseline and time points 3 and
improvements in the behaviour of people with 5, but the experimental group showed a greater
dementia (Brotons & Pickett-Cooper, 1996; degree of agitation at time points 2 and 4.
Fitzgerald-Cloutier, 1993; Groene, 1993; Jennings The observation that participants showed high
& Vance, 2002; Nugent, 2000; Thomas et al., inter- and intra-participant variability over time is
1997). It is plausible that music therapy only reduces consistent with other music and agitation studies
agitation while it is regulating people’s orientation (Casby & Holm, 1994; Clair & Bernstein, 1994;
and arousal levels (Baker, 2002; Ridder, 2003) or Clark, Lipe, & Bilbrey, 1998; Groene, 1993;
promoting a safe and secure environment (Gerdner, Nugent, 2000; Sherratt, Thornton, & Hatton,
1999). The effects of music therapy may disappear 2004a) and reports in general gerontology literature
once the predictable and structural qualities of (Cohen-Mansfield, 1999, 2000; Levy et al., 1996).
336 A. J. Ledger & F. A. Baker
Table III. Total CMAI means and standard deviations for the measures may be particularly effective in detecting
pooled sample.
changes in agitation levels over time (Sherratt et al.,
Time point M SD 2004b).

CMAI 1 31.47 19.46

CMAI 2 32.04 19.15
CMAI 3 35.64 17.67 Conclusion
CMAI 4 37.31 18.94
As a unique investigation of long-term effects of
CMAI 5 30.29 17.72
music therapy on people with Alzheimer’s disease,
this study highlights possibilities for further research.
The large number of uncontrolled variables could The music therapists observed that sessions had
have contributed to this high degree of variability in lasting effects on areas of functioning aside
agitation levels. There were many outside factors from agitation. The suggestion that music therapy
that could have affected participants’ agitation levels participants maintained verbal abilities over time
over the year–illnesses, hospitalizations, changes in particularly requires further longitudinal study.
medications, bedroom changes, and deaths among It may also be worth investigating long-term effects
family and friends. Differences in the nursing home of music therapy on quality of life. Observations
environments, timing of music therapy sessions, throughout the year suggested that music therapy
music therapy group sizes, music therapy techniques was a safe, useful intervention for reducing existing
employed, and therapists conducting the sessions agitated behaviours when they occurred. Group
could also have influenced the results of the study. music therapy participants who displayed agitation
Future researchers may wish to collect more detailed prior to sessions typically showed less agitated
information on potential outside influences on behaviour during and immediately after sessions.
agitation, to enhance the reliability of their findings. As agitated behaviours have major implications for
Low endorsement of CMAI behaviours may also the well-being of people with Alzheimer’s disease
account for the lack of significance found between and those who care for them, interventions that may
the two conditions. The CMAI instruction manual reduce agitation are vital, whether outcomes are
(Cohen-Mansfield, 1991) puts forward that a person short or long-term.
can be considered ‘agitated’ if they display aggressive
behaviour at least several times per week, physical
non-aggressive behaviour at least once a day, or Acknowledgements
verbal agitation at least once a day. Although
medical charts and the treating therapists’ observa- Deborah Simpson and Loretta Quinn helped with
tions indicated that participants reached these levels data collection, along with staff of Anglicare,
of agitation, CMAI scores did not reflect this. Blue Care, Regis, and Dorothy Impey Nursing
The pooled sample’s mean total CMAI scores Homes, Australia. Dr Mark Bahr provided initial
ranged from only 30.29 (time point 5) to 37.31 statistical advice and Associate Professor Gerard
(time point 4) out of a possible 174 points, Byrne and Aaron Frost assisted in the analysis and
suggesting that the participants displayed an agitated interpretation of the data. Study participants and
behaviour less than once per week on average. their families are also acknowledged for their
Nugent (2000) similarly identified low scoring of contributions to the research.
behaviours as an obstacle to achieving statistical
significance in her research. It is therefore advised
that future researchers using the CMAI as an References
outcome measure determine a minimum CMAI
score for selecting research participants. Ashida, S. (2000). The effect of reminiscence music therapy
sessions on changes in depressive symptoms in elderly persons
As high inter-and intra-participant variability and
with dementia. Journal of Music Therapy, 37, 170–182.
low endorsement of CMAI behaviours could have Baker, F. (2002). The effects of live, taped and no music on
been obstacles to achieving statistical significance, people experiencing posttraumatic amnesia. Journal of Music
it is recommended that future studies evaluate Therapy, 38, 82–104.
changes in the particular, unique behaviours dis- Bright, R. (1997). Music therapy and the dementias: Improving the
quality of life, (2nd ed.). St. Louis: MMB Music.
played by individual music therapy participants.
Brotons, M., & Koger, S. M. (2000). The impact of music
Rating scales such as the CMAI are quick and therapy on language functioning in dementia. Journal of Music
cost-effective ways of measuring agitated behaviours Therapy, 37, 183–195.
(Neville & Byrne, 2001), but it can be difficult to Brotons, M., & Pickett-Cooper, P. K. (1996). The effects of
achieve statistical significance based on changes in music therapy intervention on agitation behaviours of
Alzheimer’s disease patients. Journal of Music Therapy, 33,
the total scores or subtype scores. Music therapy
2–18.
researchers may be more successful in establishing Brotons, M., Koger, S. M., & Pickett-Cooper, P. (1997).
effects on agitation by employing more than one Music and dementias: A review of literature. Journal of Music
method of agitation measurement. Observational Therapy, 34, 204–245.
 Long-term effects of music therapy on agitation 337
Casby, J. A., & Holm, M. B. (1994). The effect of music Gerdner, L. A. (1999). Individualized music intervention
on repetitive disruptive vocalizations of persons with dementia. protocol. Journal of Gerontological Nursing, 25, 10–16.
The American Journal of Occupational Therapy, 48, 883–889. Groene II, R. W. (1993). Effectiveness of music therapy
Christie, M. E. (1995). The influence of a highly participatory 1:1 intervention with individuals having senile
peer on motivating group behaviors of lower dementia of the Alzheimer’s type. Journal of Music Therapy,
functioning persons who have probable Alzheimer’s type 30, 138–157.
dementia: A feasibility study. Music Therapy Perspectives, 13, Groene II, R., Zapchenk, S., Marble, G., & Kantar, S. (1998).
87–90. The effect of therapist and activity characteristics on the
Clair, A. A., & Bernstein, B. (1990). A preliminary study of music purposeful responses of probable Alzheimer’s disease partici-
therapy programming for severely regressed persons with pants. Journal of Music Therapy, 35, 119–136.
Alzheimer’s-type dementia. The Journal of Applied Gerontology, Gruber-Baldini, A. L., Boustani, M., Sloane, P. D., &
9, 299–311. Zimmerman, S. (2004). Behavioral symptoms in residential
Clair, A. A., & Bernstein, B. (1994). The effect of no music, care/assisted living facilities: Prevalence, risk factors, and
stimulative background music and sedative background music medication management. Journal of the American Geriatrics
on agitated behaviours in persons with severe dementia. Society, 52, 1610–1617.
Activities, Adaptation & Aging, 19, 61–70. Hanson, N., Gfeller, K., Woodworth, G., Swanson, E. A., &
Clark, M., Lipe, A., & Bilbrey, M. (1998). Use of music to Garand, L. (1996). A comparison of the effectiveness of
decrease aggressive behaviours in people with dementia. differing types and difficulty of music activities in programming
Journal of Gerontological Nursing, 24, 10–17. for older adults with Alzheimer’s disease and related disorders.
Cohen-Mansfield, J. (1986). Agitated behaviours in the elderly II: Journal of Music Therapy, 33, 93–123.
Preliminary results in the cognitively deteriorated. Journal of the Haupt, M., Kurz, A., & Jänner, M. (2000). A 2-year follow-up of
American Geriatrics Society, 34, 722–727. behavioural and psychological symptoms in Alzheimer’s
Cohen-Mansfield, J. (1989). Agitation in the elderly. In N. Billig disease. Dementia and Geriatric Cognitive Disorders, 11,
& P. V. Rabins (Eds.), Issues in geriatric psychiatry 147–152.
(pp. 101–113). Basel, Switzerland: Karger. Jennings, B., & Vance, D. (2002). The short-term effects
Cohen-Mansfield, J. (1991). Instruction manual for the of music therapy on different types of agitation in
Cohen-Mansfield Agitation Inventory (CMAI). Rockville, adults with Alzheimer’s. Activities, Adaptation and Aging, 26,
Maryland: Research Institute of the Hebrew Home of 27–33.
Greater Washington. Kahn, R. L., Goldfarb, A. I., Pollack, M., & Peck, A. (1960).
Cohen-Mansfield, J. (1999). Measurement of inappropriate Brief objective measures for the determination of mental
behaviour associated with dementia. Journal of Gerontological status in the aged. American Journal of Psychiatry, 117,
Nursing, 25, 42–51. 326–328.
Cohen-Mansfield, J. (2000). Heterogeneity in dementia: Koger, S. M., Chapin, K., & Brotons, M. (1999). Is
Challenges and opportunities. Alzheimer Disease & Associated music therapy an effective intervention for dementia? A
Disorders, 14, 60–63. meta-analytic review of literature. Journal of Music Therapy,
Cohen-Mansfield, J., & Martin, L. S. (1999). Assessment of 36, 2–15.
agitation in older adults. In P. A. Lichtenberg (Ed.), Handbook Koss, E., Weiner, M., Ernesto, C., Cohen-Mansfield, J.,
of assessment in clinical gerontology (pp. 297–331). New York: Ferris, S. H., Grundman, M., et al. (1997). Assessing patterns
John Wiley & Sons. of agitation in Alzheimer’s disease patients with the
Cohen-Mandsfield, J., Marx, M. S., & Rosenthal, A. S. (1989). Cohen-Mansfield Agitation Inventory. Alzheimer’s Disease and
A description of agitation in a nursing home. The Journals of Associated Disorders, 11, S45–50.
Gerontology, 44, M77–84. Ledger, A., & Baker, F. (2005). Longitudinal research designs in
Colenda III, C. C. (1995). Agitation: A conceptual overview. music therapy with people who have dementia. The Australian
In B. A. Lawlor (Ed.), Behavioral complications Journal of Music Therapy, 16, 88–103.
in Alzheimer’s disease (pp. 3–17). WA: American Psychiatric Lesta, B., & Petocz, P. (2006). Familiar group singing:
Press. Addressing mood and social behaviour of residents with
Ebberts, A. G. (1994). The effectiveness of three types of music dementia displaying sundowning. The Australian Journal of
therapy interventions with persons diagnosed with probable Music Therapy, 17, 2–17.
dementia of the Alzheimer’s type who display agitated Levy, M. L., Cummings, J. L., Fairbanks, L. A., Bravi, D.,
behaviours. Unpublished master’s thesis, University of Calvani, M., & Carta, A. (1996). Longitudinal assessment of
Kansas, USA. symptoms of depression, agitation, and psychosis in 181
Eustace, A., Coen, R., Walsh, C., Cunningham, C. J., patients with Alzheimer’s disease. American Journal of
Walsh, J. B., Coakley, D., et al. (2002). A longitudinal Psychiatry, 153, 1438–1443.
evaluation of behavioural and psychological symptoms of Lou, M. F. (2001). The use of music to decrease agitated
probable Alzheimer’s disease. International Journal of Geriatric behaviour of the demented elderly: The state of the science.
Psychiatry, 17, 968–973. Scandinavian Journal of Caring Sciences, 15, 165–173.
Finkel, S. I., Lyons, J. S., & Anderson, R. L. (1992). Reliability Mahoney, E. K., Volicer, L., & Hurley, A. C. (2000).
and validity of the Cohen-Mansfield Agitation Inventory in Management of challenging behaviors in dementia. Baltimore:
institutionalized elderly. International Journal of Geriatric Health Professions Press.
Psychiatry, 7, 487–490. Marin, D. B., Green, C. R., Schmeidler, J., Harvey, P. D.,
Fitzgerald-Cloutier, M. (1993). The use of music therapy to Lawlor, B. A., Ryan, T. M., et al. (1997). Noncognitive
decrease wandering: An alternative to restraints. Music Therapy disturbances in Alzheimer’s disease: Frequency, longitudinal
Perspectives, 11, 32–36. course, and relationship to cognitive symptoms. Journal of the
Foli, S., & Shah, A. (2000). Measurement of behavioural American Geriatrics Society, 45, 1331–1338.
disturbance, non-cognitive symptoms and quality of life. Miller, R. J., Snowdon, J., & Vaughan, R. (1995). The use of the
In J. O’Brien, D. Ames & A. Burns (Eds.), Dementia, Cohen-Mansfield agitation inventory in the assessment of
(2nd ed., pp. 87–100). London: Arnold. behavioral disorders in nursing homes. Journal of the
Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). American Geriatrics Society, 43, 546–549.
‘‘Mini-Mental State’’: A practical method for grading the Neville, C. C., & Byrne, G. J. A. (2001). Behaviour rating scales
cognitive state of patients for the clinician. Journal of Psychiatric for older people with dementia: Which is the best for use by
Research, 12, 189–198. nurses? Australasian Journal on Ageing, 20, 166–172.
338 A. J. Ledger & F. A. Baker
Nugent, N. (2000). The effects of live versus taped preferred Sherratt, K., Thornton, A., & Hatton, C. (2004b). Music
songs on individuals with dementia of the Alzheimer’s type interventions for people with dementia: A review of the
displaying agitated behaviours. Unpublished master’s thesis, literature. Aging & Mental Health, 8, 3–12.
University of Melbourne, Australia. Short, A. E. (1995). Insight-oriented music therapy with elderly
Nugent, N. L. (2002). Agitated behaviours in Alzheimer’s disease residents. The Australian Journal of Music Therapy, 6, 4–17.
and related disorders: Music and music therapy research. Silber, F., & Hes, J. P. (1995). The use of songwriting with
The Australian Journal of Music Therapy, 13, 38–50. patients diagnosed with Alzheimer’s disease. Music Therapy
Olderog Millard, K. A., & Smith, J. M. (1989). The influence of Perspectives, 13, 31–34.
group singing therapy on the behaviour of Alzheimer’s disease Smith, G. H. (1986). A comparison of the effects of three
patients. Journal of Music Therapy, 26, 58–70. treatment interventions on cognitive functioning of Alzheimer
Opie, J., Rosewarne, R., & O’Connor, D. W. (1999). The efficacy patients. Music Therapy, 6A(1), 41–56.
of psychosocial approaches to behaviour disorders in dementia: Snowdon, J., Miller, R., & Vaughan, R. (1996). Behavioural
A systematic literature review. Australian and New Zealand problems in Sydney nursing homes. International Journal of
Journal of Psychiatry, 33, 789–799. Geriatric Psychiatry, 11, 535–541.
Pollack, N. J., & Namazi, K. H. (1992). The effect of music Sourial, R., McCusker, J., Cole, M., & Abrahamowicz, M.
participation on the social behaviour of Alzheimer’s disease (2001). Agitation in demented patients in an acute care
patients. Journal of Music Therapy, 29, 54–67. hospital: Prevalence, disruptiveness, and staff burden.
Prickett, C. A., & Moore, R. S. (1991). The use of music to aid International Psychogeriatrics, 13, 183–197.
memory of Alzheimer’s patients. Journal of Music Therapy, 28, Suzuki, M., Kanamori, M., Watanabe, M., Nagasawa, S.,
101–110. Kojima, E., Ooshiro, H., et al. (2004). Behavioral
Quinn, L. J. (2003). The impact of music therapy on the quality and endocrinological evaluation of music therapy for
of life of older persons with dementia of the Alzheimer’s type in elderly patients with dementia. Nursing and Health Sciences, 6,
11–18.
the home setting: Perceptions of clients and their carers.
Thomas, D. W., Heitman, R. J., & Alexander, T. (1997).
Unpublished master’s thesis, University of Melbourne,
The effects of music on bathing cooperation for residents
Australia.
with dementia. Journal of Music Therapy, 34, 246–259.
Reisberg, B., Ferris, S. H., De Leon, M. J., & Crook, T. (1982).
Tomaino, C. M. (2000). Working with images and recollection
The global deterioration scale for assessment of primary
with elderly patients. In D. Aldridge (Ed.), Music therapy in
degenerative dementia. American Journal of Psychiatry, 139,
dementia care: More new voices (pp. 195–211). London: Jessica
1136–1139.
Kingsley.
Ridder, H. M. (2003). Singing dialogue: Music therapy with
Tractenberg, R. E., Weiner, M. F., & Thal, L. J. (2002).
persons in advanced stages of dementia. A case study research
Estimating the prevalence of agitation in community-dwelling
design. Unpublished doctoral thesis, Aalborg University,
persons with Alzheimer’s disease. The Journal of
Denmark. Neuropsychiatry and Clinical Neurosciences, 14, 11–18.
Ridder, H. M. (2005). An overview of therapeutic initiatives when Vink, A. C., Birks, J. S., Bruinsma, M. S., & Scholten, R. J. S.
working with people suffering from dementia. In D. Aldridge (2003). Music therapy for people with dementia. The Cochrane
(Ed.), Music therapy and neurorehabilitation (pp. 61–82). Database of Systematic Reviews, 4, Article CD003477. Retrieved
London: Jessica Kingsley. February 15, 2005, from http://www.mrw.interscience.wiley.
Rio, R. (2002). Improvisation with the elderly: Moving from com.ezproxy.library.uq.edu.au/cochrane/clsysrev/articles/
creative activities to process-oriented therapy. The Arts in CD003477/frame.html
Psychotherapy, 29, 191–201. Weiner, M. F., & Schneider, L. S. (2003). Drugs for behavioural,
Ruud, E. (1998). Music, health, and quality of life. In Music psychological, and cognitive symptoms. In M. F. Weiner &
therapy: Improvisation, communication and culture (pp. 49–67). A. M. Lipton (Eds.), The dementias: Diagnosis, management, and
Gilsum, NH: Barcelona. research, (3rd ed., pp. 219–283). Washington: American
Sambandham, M., & Schirm, V. (1995). Music as a nursing Psychiatric Publishing.
intervention for residents with Alzheimer’s disease in long-term Weiner, M. F., Koss, E., Patterson, M., Jin, S., Teri, L.,
care. Geriatric Nursing, 16, 79–83. Thomas, R., et al. (1998). A comparison of the Cohen-
Shah, A., Evans, H., & Parkash, N. (1998). Evaluation of three Mansfield Agitation Inventory with the CERAD behavioral
aggression/agitation behaviour rating scales for use on an acute rating scale for dementia in community-dwelling persons with
admission and assessment psychogeriatric ward. International Alzheimer’s disease. Journal of Psychiatric Research, 32,
Journal of Geriatric Psychiatry, 13, 415–420. 347–351.
Sherratt, K., Thornton, A., & Hatton, C. (2004a). Emotional Weiner, M. F., Tractenberg, R., Teri, L., Logsdon, R.,
and behavioural responses to music in people with Thomas, R. G., Gamst, A., et al. (2000). Quantifying
dementia: An observational study. Aging & Mental Health, behavioral disturbance in Alzheimer’s disease patients.
8, 233–241. Journal of Psychiatric Research, 34, 163–167.