Content uploaded by S. Guétin
Author content
All content in this area was uploaded by S. Guétin on Dec 23, 2013
Content may be subject to copyright.
Fax +41 61 306 12 34
E-Mail karger@karger.ch
www.karger.com
Original Research Article
Dement Geriatr Cogn Disord 2009;28:36–46
DOI: 10.1159/000229024
Effect of Music Therapy on Anxiety and
Depression in Patients with Alzheimer’s Type
Dementia: Randomised, Controlled Study
S. Guétin a, c, d F. Portet a M.C. Picot b C. Pommié a, c M. Messaoudi a
L. Djabelkir a A.L. Olsen c M.M. Cano c E. Lecourt d J. Touchon a, c
a Service de Neurologie, Centre Mémoire de Ressources et de Recherches (CMRR), Inserm U888, CHRU
Montpellier, and
b Département d’Information Médicale, CHRU Arnaud de Villeneuve, Montpellier ,
c Association
de Musicothérapie Applications et Recherches Cliniques (AMARC) and
d Laboratoire de Psychologie Clinique et
Psychopathologie (LCPL) EA 4056, Université Paris 5 – Renée Descartes, Paris , France
The effect of music therapy was sustained for up to 8 weeks
after the discontinuation of sessions between weeks 16 and
24 (p ! 0.01). Conclusion: These results confirm the valuable
effect of music therapy on anxiety and depression in pa-
tients with mild to moderate Alzheimer’s disease. This new
music therapy technique is simple to implement and can
easily be integrated in a multidisciplinary programme for the
management of Alzheimer’s disease.
Copyr ight © 2009 S. Karger AG, B asel
Introduction
According to a recent study, 24.3 million people cur-
rently suffer from Alzheimer’s disease or related disor-
ders, and 4.6 million new cases are reported worldwide
each year. The number of patients is expected to double
every 20 years, to reach 43.2 million by 2020 and 81.1
million by 2040
[1] . Alzheimer’s type dementia (AD) is
the most common degenerative disease, with only half of
the cases being diagnosed and one third treated. With the
2-fold increase in the number of cases anticipated over
the next few decades, this progressive disease has become
a major public health problem. Alzheimer’s disease is
characterised by acquired impairment in cognitive func-
tion, with a gradual impact on the patient’s professional
Key Words
Music therapy ⴢ Alzheimer’s disease ⴢ Depression ⴢ Anxiety
Abstract
Background/Aims: Numerous studies have indicated the
value of music therapy in the management of patients with
Alzheimer’s disease. A recent pilot study demonstrated the
feasibility and usefulness of a new music therapy technique.
The aim of this controlled, randomised study was to assess
the effects of this new music therapy technique on anxiety
and depression in patients with mild to moderate Alzhei-
mer-type dementia. Methods: This was a single-centre,
comparative, controlled, randomised study, with blinded as-
sessment of its results. The duration of follow-up was 24
weeks. The treated group (n = 15) participated in weekly ses-
sions of individual, receptive music therapy. The musical
style of the session was chosen by the patient. The validated
‘U’ technique was employed. The control group (n = 15) par-
ticipated under the same conditions in reading sessions. The
principal endpoint, measured at weeks 1, 4, 8, 16 and 24, was
the level of anxiety (Hamilton Scale). Changes in the depres-
sion score (Geriatric Depression Scale) were also analyzed as
a secondary endpoint. Results: Significant improvements in
anxiety (p ! 0.01) and depression (p ! 0. 01) wer e obs erved in
the music therapy group as from week 4 and until week 16.
Accepted : June 2, 2009
Publis hed online: July 23, 200 9
Stéphane Guet in
9 rue L éon Cogniet
FR–75017 Paris (France)
Tel. +33 6 20 47 67 57, E-Mail stephane.guetin@yahoo.f r
© 200 9 S. Karger AG, Basel
1420–8008/09/0281–0036$26.00/0
Accessible online at:
www.karger.com/dem
Effect of Music Therapy in Alzheimer’s
Disease
Dement Geriatr Cogn Disord 2009;28:36–46
37
and social/family activities. Changes in emotions and be-
havioural disorders are generally already present. Vari-
ous types of depressive and anxiety disorder may develop
and are said to be among the earliest noncognitive ex-
pressions of the disease
[2] . Psychological/behavioural
disorders become apparent from the start of progression:
a tendency towards isolation, apathy, lack of interest and
gradual withdrawal from activities. These disorders are
often associated with irritability, aggression and unchar-
acteristic emotional reactions
[3] .
Recent clinical studies, namely in functional neuro-
imaging, have been able to evidence the favourable role
of music therapy in the management of Alzheimer’s dis-
ease
[4, 5] . Music-based therapy corresponds to 2 funda-
mental methods, a ‘receptive’ listening-based method,
and an ‘active’ method, based on playing musical instru-
ments. Music therapy was defined by Munro and Mount
[6] as: ‘the intentional use of the properties and the po-
tential of music and its impact on the human being’. Re-
ceptive music therapy is perceived by Biley
[7] as a ‘con-
trolled method for listening to music, making use of its
physiological, psychological and emotional impact on the
individual during treatment for an illness or trauma’. A
distinction is generally made between 2 types of receptive
method: (1) receptive ‘relaxation’ music therapy
[8, 9] :
this method is similar to other approaches, such as hyp-
nosis, sophrology and relaxation in general, and is often
used in the treatment of anxiety, depression and cogni-
tive disorders; (2) receptive ‘analytical’ music therapy: in
this instance, music is used as a medium for ‘analytical’
psychotherapy
[10] . The aim is to encourage the expres-
sion and development of thought. It may thus allow pa-
tients with cognitive disorders to stimulate, use and dis-
cover their remaining abilities. This psychotherapeutic
approach encourages emotional and self-enhancing sup-
port. It may be perceived as a type of psychotherapy prac-
tised in line with the major current trends in psychother-
apy. The most widely used method in the context of de-
mentia is receptive ‘relaxation’ music therapy.
The use of this method is able to reduce the frequency
and extent of affective and psychological/behavioural
disorders. Music is a major means of triggering emotions
and helping patients express themselves verbally. Music
therapy stimulates intellectual function, acts on anxiety
and depression and thus significantly improves autono-
my in patients suffering from Alzheimer’s disease
[11–
15] . This is because the music is chosen on the basis of
personal experience, which will stimulate memory by
evoking autobiographical events. Listening to music, to-
gether with the resulting relaxation factor, is also effec-
tive in numerous areas. Choosing music connected to the
individual’s personal experience is thus of paramount
importance. These studies confirm that music therapy
has a relaxing effect on patients suffering from Alzhei-
mer’s disease.
A pilot study demonstrated the feasibility and benefit
of individual receptive music therapy sessions. Signifi-
cant improvements in anxiety and depression (p ! 0.001)
were observed from the first session and were maintained
significantly during the subsequent sessions. The physi-
cal and mental burden felt by the main caregiver was re-
duced significantly (p ! 0.01). The sessions helped stimu-
late cognitive function by encouraging memory encod-
ing and recall
[16] . The results obtained made it possible
to estimate the number of subjects required to set up a
randomised controlled study.
The primary objective of this randomised controlled
study is to evaluate the impact of short- and medium-
term music therapy on anxiety disorders in patients suf-
fering from mild to moderate stages of AD. The second-
ary objectives concern depression and the persisting ef-
fect of music therapy up to 2 months after discontinuation
of the sessions.
Materials and Methods
Consent
This study received a favourable opinion from the ethics com-
m it t e e, a s r e qu i re d by F re n ch le g i sl a t io n o n bi o et h i cs , ev e n t h ou g h
the study does not entail any additional risks (music therapy ses-
sion, no impairment of physical or psychological integrity). Dur-
ing the inclusion visit, and before any subjects were included in
the study, potential ly eligible subjects (or their fami ly or legal rep-
resentative) signed the informed consent form (stating that they
did not object) to take part in the project.
T y p e o f S t u d y
The study design corresponded to a randomised, controlled,
comparative, si ngle-centre study, with t he results evaluated u nder
blind conditions. The study was conducted over a total duration
of 18 months, with a follow-up period of 6 months.
S t u d y P o p u l a t i o n
The included patients were residents at the Les Violettes nurs-
ing home in Montpellier over the period from September 2007 to
April 2008. They a ll suffered from mild to moderate stages of AD.
Each patient was required to have a baseline Mini Mental State
Evalu ation (MMSE)
[17, 18] score of between 12 a nd 25 and a base-
line Hamilton Anxiety Scale score of at least 12. The included
patients were men or women aged 70–95 yea rs, with adequate ver-
bal or written expression, visual and hearing abilities (hearing
aids not permitted) in order to carry out the tests. All of the pa-
tients had been receiving stable anticholinergic treatment for 6
months. Psychotropic and anxiolytic treatment was authorised at
Guétin et al.
Dement Geriatr Cogn Disord 2009;28:36–46
38
s ta bl e r ed uc ed do se s. Pa ti en ts c ons id er ed h ig hl y l i ke ly no t t o c om -
ply with the protocol or to drop out of the study as well as those
suffering from a life-threatening illness during the envisaged
study period were not included in the study. Likewise, patients
with ot her neurological dis orders, st roke, Parkinson’s disease, ep -
ilepsy, Lewy body dementia defined by the presence of extrapyra-
midal symptoms, hallucinations, unexplained episodes of confu-
sion, dementia possibly of vascular origin (modified Hachinski
ischaemia score 1 4), frontal dementia (frontal score 1 3) and ps y-
chiatric disorders (schizophrenia, bipolar disorders or depression
as per the major depressive disorder criteria of DSM-IV) were not
included in the study.
S a m p l e S i z e
The number of subject required was estimated at 11 per group
for a type I risk of 5% and a power of 90% with a 2-sided hypoth-
esis. This sample size was based on the results of the preliminary
study
[16] , tak ing an improvement corresponding to 7 u nits (on the
Hamilton Scale) with a standard deviation of 2.6 in the music ther-
apy group versus an improvement corresponding to 3 in the con-
trol group (improvement close to the standard deviation). Consid-
ering the a nticipated number of patients lost to follow-up, the sam-
ple size for the group was increased to 15 subjects per group.
Thirty subjects in total were included in the context of the study.
Authorised Medication/Concomitant Medication
All medicinal products and preparations, including over-the-
counter products, taken by the patient during the study were re-
corded in the case report form stating the name, dosage, indica-
tion and treatment duration.
The intake of me dicina l pro duct s wa s rec orded at e ach f ollo w-
up visit. No modifications in medication or significant changes
in medicinal product intake were observed during the study, ir-
respective of therapeutic class and patient group.
M e t h o d
All of the included patients underwent a clinical evaluation
and neuropsychological assessment at day 0 (D0), week 4 (W4),
W8, W16 and W24. This follow-up was carried out in a visit con-
text. Each subject underwent a clinical examination by a neurolo-
gist experienced in the diagnosis of AD, together with a neuro-
psychologist, and carried out all of the envisaged tests and ex-
aminations.
Thi rty patients in tot al were randomised to one of the 2 g roups,
i.e. 15 patients per group. The subjects were followed up at W4,
W8, W16 and W24 ( fig. 1 ).
In the group of patients undergoing music therapy, the ses-
sions took place once a week between D0 and W16. The patients
in the control group, without music therapy, took part in a differ-
ent type of session (rest and reading), under the same conditions
and at the same intervals.
The results obtained at D0, W4, W8, W16 and W24 were col-
lected by an independent neuropsychologist assessor (D.L.), not
belonging to the care team and unaware of the type of interven-
tion. The assessment at W24 made it possible to observe the po-
tential persisting effect of music therapy.
Intervention Method
The individual receptive music therapy method was used.
This may help reduce a nxiety, depression and ag itation in patients
sufferi ng from Alzheimer’s disease
[19 , 2 0] . The music was chosen
based on the patients’ personal tastes following an interview/
questionnai re. Choosing music connected to the individual’s per-
sonal experience is of paramount importance. The style of music
chosen varies from one patient to another, but also from one ses-
sion to another for a given patient. The Centre Hospitalier Ré-
gional de Montpellier (CHRU) and Association de Musicothéra-
pie Applications et Recherches Cliniques (AMARC) thus de-
signed a computer program for this purpose. This makes it
possible to select a musical sequenc e suited to the patient’s request
from the different musical styles suggested (classical music, jazz,
world music, various). The standard musical sequence, lasting 20
min, is broken down into several phases which gradually bring
the patient into a state of relaxation according to the new ‘U se-
quence’ method
[8, 9, 16] . This works by reducing the musical
rhyth m, orchestral format ion, frequency and volume (descendi ng
‘U’ phase). After a phase of maximum relaxation (bottom ‘U’ seg-
ment), a re-enlivening phase follows (ascending ‘U’ segment)
( fig. 2 ). All of the music sequences, constructed using the ‘U se-
quence’ method, were specially created by the record publishing
company, Music Care ( table 1 ).
The music was streamed via headphones in the patients’
rooms. The patients were either in a supine position or seated in
a comfortable armchair. They were also offered a mask so as to
avoid visual stimul i, thus encouraging them to concentrate on the
music.
Clinical evaluations
Patient recruitment
12 weeks
With music therapy (n = 15)
Without music therapy (n = 15)
W24W16
W24W16
W8W4D0
Fig. 1. Study flow chart.
Effect of Music Therapy in Alzheimer’s
Disease
Dement Geriatr Cogn Disord 2009;28:36–46
39
R a n d o m i s a t i o n
The patients were allocated to the different groups by ran-
domisation at the end of the inclusion visit (V0), after patient in-
formation, verification of inclusion and exclusion criteria, and
signing the consent form.
Randomisation was generated in blocks of 4 by the method-
ological team (Clinical Research Unit, Montpellier CHRU).
S t u d y E n d p o i n t s
• The prima ry study endpoint corresponded to a nxiety bet ween
D0 and W16, measured using the Hamilton Scale, with the
total score ranging from 0 to 56
[21, 22] . This scale consists of
14 items covering all of the sectors of psychosomatic anxiety.
• The secondary endpoints corresponded to depression mea-
sured by means of a score obtained from the Geriatric Depres-
sion Scale (GDS) questionnai re. This i s a self-asse ssment ques-
tionnaire consisting of 30 dichotomous questions, perceived
as the reference diagnostic tool for evaluating depression in
the elderly. The maximum score is 30
[23] .
Statistical Analysis
A ll of the ra ndomised patients were included in the intent-to-
treat population. An overall description of each variable consid-
ered was drawn up for each group. The quantitative data were
described in terms of sample size, mean, standard deviation and
range (minimum and maximum). The qualitative data were de-
scribed by their distribution in terms of sample size and percent-
age by class. The normality of data was verified using the Kol-
mogorov-Smirnov test. The comparability of the 2 groups was
verified on the baseline data (D0). The means were compared us-
ing Student’s t test or the Mann-Whitney nonparametric test.
Qualitative variables were compared with the 2 test or Fisher’s
exact test. A multivariate analysis was performed by means of
ANOVA with repeated measures, in order to study the overall
changes in the endpoints measured during follow-up. The differ-
ences between 2 consecutive time points and between each time
point and D0 were tested. The tests were 2-sided, with a signifi-
cance limit of 5%. The statistical analysis was performed using
SAS 쏐 software V9.1.
R e s u l t s
Figure 3 illustrates the patient distribution within the
groups. Two patients were prematurely withdrawn from
the study in the intervention group: 1 between W8 and
W16 owing to an intercurrent event not related to the
study (life-threatening situation, hospitalisation), and
the second died between W16 and W24. Four patients
were withdrawn from the study in the control group: 1
between W4 and W8 due to dropping out, 1 between W4
and W8 owing to an intercurrent event not related to the
study (hospitalisation), 1 patient died between W4 and
W8, and the last patient dropped out between W16 and
W24.
40 >T > 30
OF: 1–3
Stimulating rhythm
Slow rhythm (relaxation)
20 min
Moderate rhythm
80 >T > 60
OF: 8–10
60 >T > 40
OF: 3–8
80 >T > 60
OF: 5–10
95 >T > 80
OF: 10–20
60 >T > 40
OF: 2–5
Fig. 2. New music therapy technique: the
‘U’ sequence. Arrows indicate volume lev-
el. T = Tempo (beats per minute); OF = or-
chestral formation (number of instru-
ments).
Tab le 1. Choice of suggested music styles
Classical Jazz World Various
Piano Piano Cuba Popular accordion music
Violin Guitar Andes World accordion music
Flute Saxophone India Classic vocals
Harp Trumpet Ireland Popular vocals
Oboe Trombone Spain New age music
Guétin et al.
Dement Geriatr Cogn Disord 2009;28:36–46
40
Randomised Comparative Study
The comparability of the 2 groups was verified at in-
clusion ( table 2 ) for the main demographic, sociocultural
and medical characteristics.
The 2 groups were comparable at inclusion in terms of
demographic and sociocultural data and history of the
disease, apart from there being a higher number of wom-
en in the music therapy group.
The data relating to patient clinical examination are
described and compared between the 2 groups in table 2 .
The score for the Hamilton Anxiety Scale, the MMSE
score and the GDS score, obtained during the baseline
visit, are shown. No statistically significant differences
are observed between the 2 groups as regards the scores
obtained for the Hamilton Scale, GDS and MMSE at in-
clusion.
Patients meeting the criteria
n = 30
Randomisation
D0
Intervention group
n = 15
Control group
n = 15
Patient withdrawn
from study
(hospitalisation)
Intent-to-treat analysis
group (D0)
n = 15
Intent-to-treat analysis
group (D0)
n = 15
Institutionalised Alzheimer’s patients
n = 38
Excluded patients
n = 8
W4 (n = 15)
W8 (n = 15)
W16 (n = 14)
Patient withdrawn
from study
(death) W24 (n = 13)
W4 (n = 15)
W8 (n = 12)
W16 (n = 12)
W24 (n = 11)
Patients withdrawn
from study
(drop-out,
hospitalisation, death)
Patient withdrawn
from study
(drop-out)
Fig. 3. Distribution of the included patients into 2 groups.
Effect of Music Therapy in Alzheimer’s
Disease
Dement Geriatr Cogn Disord 2009;28:36–46
41
Primary Endpoint: Effect of Music Therapy on
Anxiety
The Hamilton Scale score, which makes it possible to
evaluate patient anxiety, was determined at each visit.
Figure 4 illustrates the changes in this score in each
group over time. All of the visits are shown (follow-up
over 24 weeks).
Changes between D0 and W16. ANOVA with repeated
measures (D0, W4, W8 and W16) evidenced a significant
difference (p ! 0.0001); the 2 groups progressed in a dif-
ferent manner during follow-up. At D0, it appeared that
the level of anxiety was comparable between the 2 groups:
22 ( 8 5.3) for the music therapy group and 21.1 ( 8 5.6) for
the control group. This level decreased further in the mu-
sic therapy group at W16, 8.4 ( 8 3.7) versus 20.8 ( 8 6.2)
for the control group. The changes between D0 and W16
were significantly different between the 2 groups as re-
gards this endpoint (p ! 0.001).
Table 3 indicates the values recorded for the Hamilton
scale during the 4 examinations (D0, W4, W8 and W16)
and the variations observed from one examination to the
other. After 16 weeks, the improvement corresponded to
approximately 13.2 ( 8 5.2) points, i.e. 60% (relative varia-
tion), in the music therapy group. In the control group,
this improvement was in the region of 0.9 ( 8 7.4) poi n t s,
i.e. 4.3%.
Persistence of the Effect of Music Therapy at W24. In
order to determine whether music therapy has a persis-
tent effect at 6 months, i.e. 2 months after stopping the
sessions, the scores obtained were compared between the
2 groups. ANOVA evidenced a significant difference (p !
0.0001); the 2 groups progressed in a different manner
during follow-up, up to 6 months. Table 4 describes and
compares the Hamilton score at W24, the difference be-
tween D0 and W24, and also between W16 and W24. A
score of 10.6 ( 8 6.3) was obtained in the music therapy
group versus 20.5 ( 8 5.4) in the control group at W24. The
difference between D0 and W24 appeared to be signifi-
cant regarding this endpoint (p = 0.002), together with
the difference between W16 and W24 ( table 4 ).
Effect of Music Therapy on Depression
The effect of music therapy on depression was also
evaluated. Figure 5 and table 5 show the values for the
GDS obtained during the different visits, together with
the variations observed from one examination to the
other.
Changes between D0 and W16. At D0, the mean score
was 16.7 ( 8 6.2) for the music therapy group versus 11.8
( 8 7.4) for the control group. ANOVA with repeated mea-
sures, with adjustment to the GDS score at D0, showed a
significant difference between the 2 groups (p = 0.001).
Although the overall changes were not significant over
time, each group nonetheless progressed in a different
manner during follow-up (significant time/group inter-
action p = 0.0095).
Tab le 2. Randomised comparative study
Variable Music therapy Control
Gender1
male 2 13.3 6 40
female 13 86.7 9 60
total 15 – 15 –
Marital status1
single 3 20 4 26.7
lives with partner 2 13.3 0 0
widowed or divorced 10 66.7 11 73.3
total 15 – 15 –
Place of residence1
large town 9 60 8 53.3
medium-sized town 2 13.3 4 26.7
rural setting 4 26.7 3 20
total 15 – 15 –
Education level1
<GSCE level 12 80 9 60
A level 2 13.3 2 13.3
higher education 1 6.7 4 26.7
total 15 – 15 –
Most recent occupation1
unemployed 3 20 2 13.3
farmer 0 0 1 6.7
middle management 5 33.3 3 20
labourer 7 46.7 6 40
independent profession 0 0 1 6.7
executive 0 0 2 13.3
total 15 – 15 –
Physical medicine1
cognitive stimulation 2 40 1 33.3
physiotherapy 2 40 2 66.7
speech therapy 1 20 0 0
total 5 – 3 –
Age, years285.28675/93 86.985.2 74/95
Diagnosis history, years242.4822.6 0/84 40819.1 12/84
Age at diagnosis, years281.586.4 71/93 83.685.9 70/93
Hamilton Anxiety Scale22285.3 14/29 21.185.6 12/29
GDS score216.786.2 6/26 11.887.4 1/27
MMSE219.884.4 12/25 20.783.4 12/25
1 Figures are numbers and percentages.
2 Figures are means8SD and ranges (min./max.).
GSCE = General Certificate of Secondary Education.
Guétin et al.
Dement Geriatr Cogn Disord 2009;28:36–46
42
Music therapy Control p
nmean8SD min./max. n mean8SD min./max.
Value
Anx. D0 15 22.085.3 14/29 15 21.185.6 12/29 NS
Anx. W4 15 15.583.7 6/21 15 20.784.7 12/28 0.002
Anx. W8 15 12.685.2 6/24 12 22.284.5 14/28 <0.001
Anx. W16 14 8.483.7 2/15 12 20.886.2 7/28 <0.001
Variation
D0–W4 15 –6.585.2 –15/0 15 –0.482.7 –4/8 <0.001
W4–W8 15 –2.985.5 –12/5 12 0.883.9 –8/9 NS
W8–W16 14 –4.684.8 –13/2 12 –1.485.9 –19/4 NS
D0–W16 14 –13.285.2 –21/–4 12 –0.987.4 –20/13 <0.001
Anx. = Anxiety; NS = nonsignificant.
Tab le 3. Anxiety measured using the
Hamilton Scale: values at D0, W4, W8
and W16, and variations between the 4
measurements
Music therapy Control p
nmean8SD min./max. n mean8SD min./max.
Value
Anx. W24 13 10.686.3 2/20 11 20.585.4 10/27 <0.001
Variation
D0–W24 13 –11.587.2 –22/–1 11 –1.586.8 –17/9 0.002
W16–W24 13 2.183.7 –4/8 11 –0.882.8 –7/3 0.046
Tab le 4. Study of the persistence of the
effect of music therapy on anxiety
W8
Treatment Evaluation after
treatment
0
5
10
15
20
25
30
D0
Follow-up visits
Hamilton Scale score
***
*
Control group
Music therapy group
W4 W16 W24
Fig. 4. Changes in the mean Hamilton
Scale score over time. * p ! 0.01: signifi-
cant test.
Color versi on available onlin e
Effect of Music Therapy in Alzheimer’s
Disease
Dement Geriatr Cogn Disord 2009;28:36–46
43
The level of depression decreased further in the music
therapy group at W16, 8.9 ( 8 3.3) for the intervention
group versus 11.2 ( 8 6.1) for the control group. The chang-
es between D0 and W16 appeared to be significantly dif-
ferent between the 2 treatment groups as regards this
endpoint (p = 0.002; table 5 ).
After 16 weeks, the improvement corresponded to ap-
proximately 7.7 ( 8 4.6) points, i.e. 47.1% in the music ther-
apy group with a mean depression score of 16.7 ( 8 6.2)
vers us an improve ment in the re gion of 0. 2 ( 8 4.4) points,
i.e. 1.7%, in the control group with a mean depression
score of 11.8 ( 8 7.4).
Evaluation after
treatment
Treatment
Follow-up visits
Control group
Music therapy group
D0 W4 W8 W16 W24
0
5
10
15
20
25
GDS score
p = 0.06
*** ** *
Fig. 5. Changes in the mean GDS score
over time. * p ! 0.05; * * p ! 0.01: signifi-
cant test.
Music therapy Control p
nmean8SD min./max. n mean8SD min./max.
Value
Dep. D0 15 16.786.2 6/26 15 11.887.4 1/27 NS
Dep. W4 15 13.186.1 5/26 15 12.187.2 4/25 0.046
Dep. W8 15 11.485.0 4/22 12 12.485.6 6/23 0.009
Dep. W16 14 8.983.3 4/14 12 11.286.1 4/25 0.002
Variation
D0–W4 15 –3.584.6 –13/3 15 0.382.8 –3/6 0.04
W4–W8 15 –1.782.8 –7/2 12 0.684.2 –5/8 NS
W8–W16 14 –2.282.7 –9/1 12 –1.385.0 –12/5 NS
D0–W16 14 –7.784.6 –15/–1 12 –0.284.4 –8/6 0.002
Dep. = Depression.
Tab le 5. Depression measured by the
GDS: values at D0, W4, W8 and W16, and
variations between the 4 measurements
Color versi on available onlin e
Guétin et al.
Dement Geriatr Cogn Disord 2009;28:36–46
44
Persistence of the Effect of Music Therapy at W24. The
scores obtained at W24 were compared between the 2
groups. ANOVA with repeated measures evidenced a sig-
nificant difference (p = 0.006); the 2 groups progressed
in a different manner during follow-up, up to 6 months.
Table 6 describes and compares the GDS score ob-
tained at W24, the difference between D0 and W24, and
also between W16 and W24, with adjustment on D0. The
depression score at W24 was 12.5 ( 8 6.4) in the music
therapy group and 12.1 ( 8 7.6) in the control group. The
difference between D0 and W24 appeared to be signifi-
cant regarding this endpoint (p = 0.03; table 6 ).
Additional Analyses: Changes in Cognition
As regards the MMSE, the score changed from 19.8
( 8 4.4) at D0 to 19.6 ( 8 4.4) at W16 in the music therapy
group and from 20.7 ( 8 3.4) at D0 to 19.8 ( 8 3.3) at W16
in the control group. No significant differences were evi-
denced between the 2 groups. This result was confirmed
by ANOVA with repe ated meas ures, conduc ted on 26 pa-
tients.
Discussion
This randomised controlled study, the endpoints of
which were evaluated under blind conditions, enabled a
stringent assessment of the impact of music therapy in
patients suffering from mild to moderate stages of AD.
The results obtained over the entire follow-up period
show a significant difference between the 2 groups re-
garding anxiety, the primary study endpoint. Signifi-
cantly different changes were observed between the 2
groups between D0 and W4. A reduction in the score was
thus found for the music therapy group, whereas the
mean score remained constant in the control group. Sim-
ilarly, significant changes between D0 and W8 and be-
tween D0 and W16 were evidenced between the 2 groups.
These results confirm the beneficial effect of music ther-
apy on symptoms of anxiety, from the fourth week of
treatment. The significant intergroup difference ob-
served between D0 and W24 demonstrates the persistent
effect of music therapy on symptoms of anxiety for up to
2 months after stopping the sessions ( fig. 4 ).
As regards the depression score (GDS), the 2 groups
progressed in a different manner between each follow-up
time point. Hence, between D0 and W4, a significant re-
duction was observed in the score for the music therapy
group, whereas in the group not receiving music therapy,
the mean score showed a tendency towards a slight in-
crease ( fig. 5 ). Likewise, significant changes between D0
and W16 were evidenced, together with significant varia-
tion between D0 and W24. The significant intergroup
difference observed between D0 and W24 tends to show
that the effect of music therapy on depression is main-
tained for up to 2 months after stopping the sessions
( fig. 5 ).
The main results are similar to those observed in the
international scientific literature
[4, 24] . Koger et al. [4]
thus carried out a review of the literature combining 69
articles published between 1985 and 1996. This analysis
reflects a favourable response to music therapy but high-
lights the lack of specific information on the action mech-
anism of this method. The variables used are extremely
heterogeneous: music therapy methods, type of music
therapist professional involved, type of dementia, degree
of cognitive impairment, sample size, etc. Koger et al.
[4] ,
Clark et al.
[25] and Sherratt et al. [24] also confirmed
these results through reviews of the literature. It is inter-
esting to note that the majority of the concerned studies
institutionalised individuals and were mainly conducted
(in two thirds of the cases) in North America
[22] . In
1999, Koger et al.
[4] emphasised the lack of published
randomised controlled studies. Only 1 review of the lit-
erature focused on the effect of music therapy on agita-
tion
[5] . Based on the analysis of 7 studies, the author
Music therapy Control p
nmean8SD min./max. n mean8SD min./max.
Value
Dep. W24 13 12.586.4 2/27 11 12.187.6 1/29 0.003
Variation
D0–W24 13 –4.084.6 –12/3 11 1.383.9 –7/8 0.003
W16–W24 13 3.484.4 –3/14 11 0.982.4 –3/5 NS
Tab le 6. Study of the persistence of the
effect of music therapy on depression
Effect of Music Therapy in Alzheimer’s
Disease
Dement Geriatr Cogn Disord 2009;28:36–46
45
noted that music therapy had a beneficial effect on this
symptom. Other studies focused on psychological and
behavioural disorders, and evaluated the effect of music
therapy on behaviour and psychoaffective symptoms.
Gerdner and Swanson
[20] examined the effects of recep-
tive music therapy on agitation and behaviour among
Alzheimer’s patients. In an initial study, they demon-
strated that individual receptive music therapy had a sig-
nificant effect on behavioural disorders and agitation
(Modified Cohen-Mansfield Agitation Inventory) in pa-
tients. This symptomatic effect was maintained for up to
1 h after stopping the sessions. In a second study, Gerdner
[19] compared the effect of individually adapted music to
that of more ‘standard’ relaxation music on patients suf-
fering from Alzheimer’s disease. Personalised music
therapy gave rise to a more marked effect on behavioural
disorders, particularly agitation.
Other studies have focused on the impact of music
therapy on cognition. For instance, in a recent study, Irish
et al.
[26] evaluated patients on 2 occasions, under differ-
ent experimental conditions: the first interview was ac-
companied by music (The Four Seasons – Vivaldi), while
the second was not. Under the conditions ‘with back-
ground music’, the authors observed considerable im-
provements in autobiographical recall (Autobiographical
Memory Interview) among the patients in comparison
with the conditions ‘without music’ (p ! 0.005). These
results were correlated with the scores obtained for the
anxiety scale (p ! 0.001; State Trait Anxiety Inventory).
Relaxing background music is therefore able to reduce
anxiety levels and thus encourage autobiographical
memory recall. These results confirm the findings ob-
served by Thompson et al.
[27] on verbal fluency in the
same type of population.
In the context of our study, the sessions were, more-
over, extended by a period of time spent listening to the
patient. This period of time thus served to create a ‘psy-
chotherapist’-type of therapeutic relationship and cer-
tainly reinforced the effect triggered by listening to mu-
sic. The actual choice of a personalised method is con-
firmed by other studies. Personalised music, which
represents music forming part of the patient’s life, sig-
nificantly reduces agitation among patients suffering
from Alzheimer’s disease, compared with neutral ‘relax-
ation’ music (p ! 0.01)
[19] .
The patients’ impressions recorded at the end of the
session, such as ‘This music reminds me of my childhood
and my family,’ or ‘I pictured myself at the ball, dancing
how we used to,’ or ‘ Th is rem in ds m e of my j ourne ys w ith
my husband,’ indicate that certain patients recall their
long-term memories. This aspect does not suggest an ef-
fect on memory processes but enables recall of older
memories
[28, 29] . The period of time spent choosing the
music according to the patients’ cultura l references there-
fore appears to represent an important moment in proto-
col implementation. The music thus has a connection
with the patient’s personal experience. Emphasis must
therefore be placed on adapting musical works to the pa-
tients’ acceptance criteria from varied styles (classical,
modern, jazz, variety, rock, world music, etc.).
The impact of music therapy may be due to neuro-
physiological effects, specific to the music, acting on the
sensory component (inducing counterstimulation of af-
ferent fibres, namely effective in the treatment of pain),
the cognitive component (stimulating memory encoding,
evoking images and memories), the affective component
(modifying mood associated with states such as depres-
sion or anxiety, and reducing tension and feelings of anx-
iety) and the behavioural component (acting on agita-
tion, muscular hypertonia and psychomotor function).
Lastly, only more in-depth neurobiological, functional
(electrophysiological, positron emission tomography,
functional MRI) or morphological (cerebral MRI) stud-
ies will be able to provide greater insight into the physi-
ological mechanisms brought into play during this type
of non-medicinal-based therapy.
Conclusion
This randomised, controlled study, conducted in a
population of patients suffering from AD, confirms the
efficacy of music therapy on anxiety and depression.
Music therapy modifies the components of the disease
through sensory, cognitive, affective and behavioural ef-
fects. Receptive music therapy encourages cognitive
stimulation, allowing patients to recall autobiographical
memories and images.
This method fits perfectly into a global multidisci-
plinary care approach. Music therapy, a method which is
easy to apply, contributes to the treatment of anxiety dis-
orders and depressive syndrome in patients suffering
from Alzheimer’s disease.
Acknowledgements
Th is research could be c arried out than ks to support from Cen-
tres Mémoire de Ressources et de Recherches, Les Violettes nurs-
ing home, Université René Descartes – Pa ris V, I nstitut Al zheimer,
the Rotary Club and La Fondation Médéric Alzheimer.
Guétin et al.
Dement Geriatr Cogn Disord 2009;28:36–46
46
References
1 Ferri CP, Prince M, Brayne C, Brodaty H,
Fratiglioni L, Ganguli M, Hall K, Hasegawa
K, Hendrie H, Huang Y, Jorm A, Mathers C,
Menezes PR, Rimmer E, Scazufca M; Alz-
heimer’s Disease International: Global prev-
alence of dementia: a Delphi consensus
study. Lancet 2005;
366: 2112–2117.
2 Bierma n EJ, Comijs HC, Jonker C, Beekman
AT: Symptoms of anxiety and depression in
the course of cognitive decline. Dement
Geriatr Cogn Disord 2007;
24: 213–219.
3 Starr JM, Lonie J: Relationship between be-
havioural and psychological symptoms of
dementia and cognition in Alzheimer’s dis-
ease. Dement Geriatr Cogn Disord 2007;
24:
343–347.
4 Koger SM, Chapin K, Brotons M: Is music
therapy an effective intervention for demen-
tia? A meta-analytic review of literature. J
Music Ther 1999;
36: 2–15.
5 Lou MF: The use of music to decrease agi-
tated behaviour of the demented elderly: the
state of the science. Scand J Caring Sci 2001;
15: 165 –173.
6 Munro S, Mount B: Music therapy in pallia-
tive care. Can Med Assoc J 1978;
119: 1029 –
1034 .
7 Biley F: Use of music in therapeutic care. Br
J Nurs 1992;
1: 178–179.
8 Guetin S, Coudey re E, Picot MC, Ginies P,
Graber-Duvernay B, Ratsimba D, Vanbierv-
liet W, Blayac JP, Hérisson C: Effect of music
therapy among hospitalized patients with
chronic low bac k pain: a controll ed, random-
ized trial. Ann Readapt Med Phys 2005;
48:
217–224.
9 Jaber S, Balhoul H, Guetin S, Chanques G,
Sebbane M , Eledjam JJ: Effe cts of music ther-
apy in intens ive care unit wit hout sedation in
weaning patients versus non-ventilated pa-
tients. Ann Fr Anesth Reanim 2007;
26: 30–
38.
10 Verdeau-Pailles J: Aspects of psychothera-
pies: music therapy and its specif icity. En-
cephale 1991;
17: 43–49.
11 Kneafsey R: The therapeutic use of music in
a care of the elderly setting: a literature re-
view. J Clin Nurs 1997;
6: 341–346.
12 Magill L: Music therapy in pain symptom
management. J Palliat Care 1993;
9: 42–48.
13 O’Callaghan C: Pain, music creativity and
music therapy in palliative care. J Palliat
Care 1996;
3: 43–49.
14 Ashida S: The effect of reminiscence music
therapy session on changes in depressive
symptoms in elderly persons with dementia.
J Music Ther 2000;
37: 170–182.
15 Kuma r AM, Tims F, Cruess D G, Mintzer MJ,
Ironson G, Loewenstein D, Cattan R, Fer-
nandez JB, Eisdorfer C, Kumar M: Music
therapy increases serum melatonin levels in
patients with Alzheimer’s diseases. Altern
Ther Health Med 1999;
5: 49–57.
16 Guetin S, Portet F, Picot MC, Defez C, Pose
C, Blayac JP, Touchon J: Impact of music
therapy on anxiety and depression for pa-
tients with Alzheimer’s disease and on the
burden felt by the main ca regiver (feasibility
study). Encephale 2009;
35: 57–65.
17 Folstein M F, Folstein SE , McHugh PR: Min i-
Mental State: a practical method for grading
the state of patients for the clinician. J Psy-
chiatr Res 1975;
12: 189–198.
18 McKhann G, Drachman D, Folstein M,
Katzman R, Price D, Stadlan EM: Clinical
diagnosis of Alzheimer’s disease: report of
the NINCDS-ADRDA Work Group under
the auspices of Depar tment of Hea lth and
Human Ser vices Task Force on Alzheimer’s
Disease. Neurology 1984;
34: 939–944.
19 Gerdner LA : Effects of ind ividuali zed versus
classical ‘relaxation’ music on the frequency
of agitat ion with Alzhei mer’s disease and re-
lated disorders. Int Psychogeriatr 2000;
12:
49–65.
20 Gerdner L A, Swanson EA: E ffects of ind ivid-
ualized music on confused and agitated el-
derly patients. Arch Psychiatr Nurs 1993;
7:
284–291.
21 Predescu V, Ciurezu T, Romila A, Pirée S,
Ionescu G, Roman I, Brasla N, Florescu D,
Damian N: The ‘double-blind’ procedure
in study of the anxiolytic effects of the prep-
aration Wy 3498 (Oxazepam): evaluation of
anxiety states with the Hamilton scale. Neu-
rol Psihiatr Neurochir 1969;
14: 153–165.
22 Hamilton M: Development of a rating sca le
for primary depressive illness. Br J Soc Clin
Psychol 1967;
6: 278–296.
23 Sheikh JI, Yesavage JA, Brooks JO III, Fried-
man LF, Gratzinger P, Hill RD, Zadeik A,
Crook T: Proposed factor structure of the
Geriatric Depression Scale. Int Psychogeri-
atr 1991;
3: 23–28.
24 She rrat t K, T hornton A, Hatt on C: Music i n-
terventions for people wit h dementia: a re-
view of the literature. Aging Ment Health
2004;
8: 3–12.
25 Clark ME , Lipe AW, Bilbrey M: Use of music
to decrease aggressive behaviors in people
with dementia. J Gerontol Nurs 1998;
24: 10–
17.
26 Irish M, Cu nningha m CJ, Walsh J B, Coakley
D, Lawlor BA, Robertson IH, Coen RF: In-
vestigating the enhancing effect of music on
autobiographical memory in mild Alzhei-
mer’s disease. Dement Geriatr Cogn Disord
2006;
22: 108–120.
27 Thompson RG, Moulin CJ, Hayre S, Jones
RW: Music enhances category f luency in
healthy older adults and Alzheimer’s disease
patients. Exp Aging Res 2005;
31: 91–99.
28 Zatorre RJ, Krumhansl CL: Neuroscience:
mental models and musical minds. Science
2002;
298: 2138–2139.
29 Blood AJ, Zatorre RJ: Intensely pleasurable
responses to music correlate with activity in
brain reg ions implicated in re ward and emo-
tion. Proc Nat l Acad Sci USA 2001;
98: 11818–
1182 3.
