See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/12519849

Effects of Individualized Versus Classical “Relaxation” Music on

the Frequency of Agitation in Elderly Persons With Alzheimer's
Disease and Related Disorders

Article  in  International Psychogeriatrics · April 2000

DOI: 10.1017/S1041610200006190 · Source: PubMed

CITATIONS READS

341 1,735

1 author:

Linda A Gerdner
Stanford University
77 PUBLICATIONS   1,773 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Evidence-Based Guideline: Individualized Music for Persons with Dementia View project

Gerdner, L. A. (2000). Effects of individualized vs. classical "relaxation" music on the frequency of agitation in elderly persons with Alzheimer's
disease and related disorders. International Psychogeriatrics, 12(1), 49-65. DOI:10.1017/S1041610200006190 Published online: 10 January
2005 at http://dx.doi.org/10.1017/S1041610200006190 View project

All content following this page was uploaded by Linda A Gerdner on 07 April 2016.

The user has requested enhancement of the downloaded file.

International Psychogeriatrics, Vol. 12, No. 1, 2000,pp. 49-65
0 2000 International Psychogeriatric Association

Third Place
31 999 IPAIBa yer Research Awards
in Psychogeriatrics
Effects of Individualized Versus
Classical ”Relaxation’’ Music on the
-
krequency of Agitation in Elderly
Persons With Alzheimer’s Disease and
Related Disorders
LINDAA. GERDNER

ABSTRACT. Confusion and agitation in elderly patients are crucial problems. This study
tested Gerdner’s mid-range theory of individualized music intervention for agitation. An
experimental repeated measures pretest-posttest crossover design compared the immediate
and residual effects of individualized music to classical “relaxation” music relative to
baseline on the frequency of agitated behaviors in elderly persons with Alzheimer’s disease
and related disorders (ADRD). Thirty-nine subjects were recruited from six long-term-care
facilities in Iowa. The sample consisted of 30 women and 9 men (mean age 82 years) with
severe cognitive impairment. Baseline data were collected for 3 weeks. Findings from the
Modified Hartsock Music Preference Questionnaire guided the selection of individualized
music. Group A ( n = 16) received individualized music for 6 weeks followed by a 2-week
“washout” period and 6 weeks of classical ”relaxation” music. Group B (n = 23) received the
same protocol but in reverse order. Music interventions were presented for 30 minutes, two
times per week. The Modified Cohen-Mansfield Agitation Inventory measured the depen-
dent variable. A repeated measures analysis of variance with Bonferroni post hoc test
showed a significant reduction in agitation during and following individualized music
compared to classical music. This study expands science by testing and supporting a
theoretically based intervention for agitation in persons with ADRD.

From Health Services Research and Develop-
ment, Center for Mental Healthcare & Outcomes
Research, Veterans Administration, University
of Arkansas for Medical Sciences, Little Rock,
Arkansas, USA (L. A. Gerdner, PhD, RN).
Requests for offprints should be di-
Offprints.
rected to Linda A. Gerdner, PhD, RN, Little
Rock VA Medical Center, Bldg. 58, 2200 Fort
Roots Dr., North Little Rock, AR 72114, USA.
e-mail: gerdnerlindaa@exchange.uams.edu

49
50
Healthcare providers identify chronic
confusion and agitation in persons with
dementia as foremost patient manage-
ment problems (Cohen-Mansfield, 1986).
These behavioral problems have detri-
mental effects not only on the physical
and psychological status of patients
but also on other individuals within
the vicinity (e.g., staff, family, visitors,
other residents). The management of
chronically confused and agitated patients
traditionally included chemical and phys-
ical restraints. However, recent federal
laws have mandated a reduction in re-
straint use. Further, research has linked
increased falls (Bradley et al., 1995;Tinet-
ti et al., 1992), injuries, and aggression
(Bradley et al., 1995) to restraint use.
Interventions such as maintenance of
a high caretaker-to-elderly ratio and spe-
cial environmental designs have been
used as a means of alleviating the use of
restraints. However, the acute shortage
of nursing personnel in long-term-care
settings makes it difficult to implement
the high staff/patient ratios necessary
to ensure safe management of problem-
atic behavior, and little conclusive evi-
dence exists on the effectiveness of
special environmental designs for man-
Cognitive Lowered
Impairment Stress
Threshold
1
7
Individualized
Music Intervention
Figure 1.Mid-range theory of individualized music intervention for agitation (Gerdner, 1997).
International Psychogeriatrics, 12(1), March 2000
L. A . Gerdner
agement of confusion and agitation (Co-
hen-Mansfield et al., 1989; Sloane et al.,
1991; Swanson et al., 1992). Such man-
agement limitations and concerns sug-
gest the need to investigate alternative
interventions such as music. The pur-
pose of this study was to compare the
immediate and 30-minute residual ef-
fects of individualized music to those of
classical “relaxation” music as an inter-
vention for agitated residents with
Alzheimer’s disease and related disor-
ders (ADRD) (Gerdner, 1998).
THEORETICAL FRAMEWORK
The mid-rangetheory of individualizedmu-
sic intervention for agitation (IMIA)(Gerd-
ner, 1997)provides the theoretical basis for
this study. Elements of the mid-range theo-
ry/ including cognitive impairment, pro-
gressively lowered stress threshold,
agitation, and individualized music inter-
vention, are presented in Figure 1.
Cognitive impairment, as found in per-
sons with ADRD, is a key antecedent to
agitation (Beck et al., 1998; Cohen-Mans-
field et al., 1995;Cohen-Mansfield&Marx,
1989; Deutsch & Rovner, 1991). Hall and
Decreased
-+ Agitation
Agitation
Effects of Music on Agitation
Buckwalter (1987) attributed this to a de-
creased ability to receive and process sen-
sory stimuli, resulting in a progressive
decline in the person’s stress threshold
and a heightened potential for anxiety.
Assessment over a 24-hour period usually
reveals that the person with ADRD experi-
ences a relatively low level of stress in the
early morning. Without intervention,stres-
sors begin to accumulate throughout the
day until they exceed the stress threshold,
usually by early afternoon. This results in
dysfunctional behaviors such as agitation
(Hall & Buckwalter, 1987).
Agitation is defined as the “inappro-
priate verbal, vocal, or motor activity
that is not explained by needs or confu-
sion of the individual per se” (Cohen-
Mansfield & Billig, 1986, p. 712).
Interventions for agitated behaviors, in-
cluding individualized music, are most
effective when implemented before the
peak level of agitation is reached (Hall &
Buckwalter, 1987).
INDIVIDUALIZED MUSIC
Patients with advanced ADRDmay have a
decreased ability or an inability to under-
stand verbal language.However, anecdot-
al data indicate that receptive and
expressive musical abilities remain intact
even in persons with severe cognitive de-
cline. The cognitive processing of music
and language appears to be conducted
independently. Experts disagree on the
hypothesized means by which this occurs
(Aldridge, 1993; Gates & Bradshaw, 1977;
Petsche et al., 1988).Swartz and colleagues
(1989) hypothesized that vocal and motor
responses and expressionsof music should
persist in persons with ADRD until the
advanced stages. Consequently, music
may be an effective alternative method of
51
communication in patients with ADRD
whose cognitive ability to receive and ex-
press language has deteriorated.
In addition to affecting verbal language,
Alzheimer’s disease affects the hippo-
campus, an essential component of the
human memory system (Walton et al.,
1988). Memory loss usually concerns re-
cent events, whereas past events or re-
mote memory remain surprisingly intact
(Burnside, 1988). A stimulus associated
with the recollection of memories may be
more pleasing to the individual than stim-
uli in their present environment.
Music serves as a powerful catalyst
for reminiscence, precipitating either
pleasant or unpleasant memories (Bright,
1982). Thaut (1990) stated that the expe-
rience associated with music is not em-
bodied in its content, but is reflective of
emotions, memories, and private imag-
es that have a learned association with
music stimuli. For music to elicit posi-
tive memories, an assumption of this
theory is that music must be individual-
ized (Gerdner, 1997). This is defined as
music that has been integrated into the
person’s life and is based on personal
preference. The musical selection must
have specific meaning to the person’s
life. For this reason, advance assessment
is critical to the selection of music.
In summary, the mid-range theory of
IMIA (Gerdner, 1997) predicts that the
presentation of carefully selected music
will provide an opportunity for the per-
son with ADRD to connect with the past.
Music changes the focus of attention and
provides an interpretable stimulus, over-
riding stimuli in the environment that
are meaningless or confusing. The elici-
tation of memories associated with pos-
itive feelings will have a soothing effect
on the person with ADRD, which should
prevent or alleviate agitation.
52
PROPOSITIONS OF THE
MID-RANGE THEORY OF IMIA
The mid-range theory of IMIA is based
on the following propositions:
1. The temporal patterning of agitated
behaviors in persons with ADRD is
often predictable based on applica-
tion of the progressively lowered
stress threshold (PLST)model (Hall
& Buckwalter, 1987).
2. Music evokes an individualized emo-
tional response within the listener
that is associated with personal
memories.
3. Response to personal memory is en-
hanced when music selection is
based on past personal preference
by the patient.
4. The presentation of an individualized
music intervention alleviates agita-
tion in the person with ADRD.
5 . The degree of significance that music
had in the person’s life prior to the
onset of cognitive impairment posi-
tively correlates with the effective-
ness of the intervention.
6 . Individualized music intervention is
most effective when the interven-
tion is implemented approximately
30 minutes prior to the peak level of
agitation.
PRELIMINARY STUDY
This theory was developed after exten-
sive work with persons with ADRD and
was further refined through a prelimi-
nary study. Gerdner (1992)used a preex-
perimental one-group pretest-posttest
design to investigate the immediate and
1-hour residual effects of individualized
music on the frequency of agitated be-
haviors in five elderly female subjects
International Psychogeriatrics, 12(1), March 2000
L. A . Gerdner
who were confused. Findings indicated
a clinically significant reduction in agi-
tation in four of the five subjects with a
statistically significant reduction in be-
haviors in the hour immediately follow-
ing the presentation of individualized
music. These findings were supported
in a modified replication study conduct-
ed by Devereaux (1997).
Knowledge of the strengths and limi-
tations of this preliminary work was syn-
thesized and incorporated into the design
of the study reported herein to strength-
en its methodology. This included the
following: (a) an increasing sample size
based on power calculations; (b) in-
creased recruitment efforts to diversify
the sample with regard to gender, race,
and ethnicity; (c) individualizing the time
of intervention to precede the subject’s
”peak level’’ of agitation based on appli-
cation of the PLST model; and (d)adding
classical music as a comparison inter-
vention based on previous research in
this area (Denny, 1997; Goddaer & Abra-
ham, 1994;Ragneskog et al., 1996;Tablos-
ki et al., 1995).
METHODS
This study compared the immediate and
30-minute residual effects of individual-
ized music to those of classical ”relax-
ation” music as an intervention for
agitated residents with ADRD, serving
as an initial effort to test the mid-range
theory of IMIA. Specifically, this study
investigated the following hypotheses:
1. Relative to baseline, a larger decrease
in frequency of agitated behaviors
will occur during the 30-minute pre-
sentation of individualized music
than during the 30-minute presen-
tation of classical music.
Effects of Music on Agitation
2. Relative to baseline, a larger decrease
in frequency of agitated behaviors
will occur during the 30 minutes
immediately following the presen-
tation of individualized music than
during the 30 minutes immediately
following the presentation of classi-
cal music.
Sample
Residents with ADRD were recruited
from six participating long-term-care
facilities in Iowa. A signed consent doc-
ument from legally authorized repre-
sentatives was obtained on 45 subjects.
However, six of these subjects were ex-
cluded from the analyses; four died, one
developed a urinary tract infection that
drastically increased agitated behaviors,
and one was transferred to another long-
term-care facility.
Consequently, 39 subjects who met the
following criteria completed the study
and were included in data analysis: (a)
between the ages of 70 and 99; (b) exhib-
ited agitation as defined by Cohen-Mans-
field and Billig (1986);(c) scored between
3 and 7 on the Global Deterioration Scale
(GDS) (Reisberget al., 1982),correspond-
ing to mild to severe cognitive decline;
(d) were able to hear a normal speaking
voice at a distance of l l / z feet; (e) had no
obvious signs of pain or infection; (f) were
a resident for a minimum of 6 weeks on
the currently assigned unit; and (g) had
available information regarding person-
al music preference.
The sample consisted of 30 women
and 9 men with a mean age of 82.6 years.
Thirty-eight subjects were White and 1
subject was Hispanic. Of the White sub-
jects, two were of Czech heritage, one
was of German heritage, and one was of
Italian descent. The majority ( n = 27,
53
69.2%) of subjects were widowed. With
regard to functional ability, the majority
( n = 19, 48%) were ambulatory, had a
weak gait ( n = 15, 38.5%), had total de-
pendence with use of toilet ( n = 18,
46.2%), had bladder ( n = 16, 41.0%) and
bowel incontinence ( M = 14, 35.9%), and
had a GDS score of 6 ( M = 34, 87.270)’
representing severe cognitive decline.
Measures
A demographic data form was complet-
ed on each subject. Information was col-
lected from documentation on the
patient’s chart (i.e., admission sheet,
minimum data set, progress notes, and
medication profile sheet).
The GDS was used for inclusionary
purposes as an assessment for cognitive
impairment. The GDS is divided into
seven categories corresponding to dis-
tinct, clinically identifiable stages of
the disease and has been correlated sig-
nificantly ( r = .31-.64, p < .05) with inde-
pendent psychometric assessments
(Reisberg et al., 1982).
The Modified Hartsock Music Prefer-
ence Questionnaire (Hartsock, 1982)was
designed to obtain specific information
on the subject’s musical preference and
to identify the importance of music in
the subject’s life during independent liv-
ing. This information was used to indi-
vidualize the selection of music that was
presented to the subject. Because the
subjects’ severe degree of cognitive im-
pairment precluded them from complet-
ing the form, a family member provided
the information. Examples of subjects’
preferred music included The Glenn
Miller Orchestra - In The Digital Mood,
Perry Como - Pure Gold, Patsy Cline’s Great-
est Hits, Czech Time with Leo Greco, and
All The Best From Germany.
54
The Temporal Patterning Assessment
of Agitation (TPAA) is a modification of
the Cohen-Mansfield Agitation Invento-
ry. It was used to estimate each subject’s
“peak” level of agitation. Certified nurs-
ing assistants from each long-term-care
facility were instructed in the use of the
TPAA. Raters, who were certified nurs-
ing assistants from the day shift (7 a.m.-
3 p.m.) and the evening shift (3-11p.m.),
assessed the subject’s behavior through-
out the assigned shift on the specified 5-
day period. Each rater then estimated
the time that the subject exhibited the
greatest number of agitated behaviors.
The time selected for intervention pre-
ceded this identified time period by 30
minutes.
The dependent variable was measured
using a modified version of the Cohen-
Mansfield Agitation Inventory (MCMAI)
(Cohen-Mansfield, 1986). The original
instrument is an observational checklist
designed to assess the frequency of 29
agitated behaviors over a 2-week period
of time and has well-established psycho-
metric properties. To provide a more
definitive assessment of the immediate
and residual effects of individualized
music on the level of agitation, the in-
strument was modified to accommodate
the 60-minute observation period using
10-minute increments.
Two videos of an elderly woman diag-
nosed with dementia were developed to
train research assistants in the proper
use of the MCMAI. The first video was
used for demonstration purposes; the
second was used for return demonstra-
tion and calculation of interrater agree-
ment, with a predetermined minimum
competency of .87. Results ranged
from 88.1 to 96.5 with a mean of 92.7.
Interrater agreement was periodically
checked to ensure the maintenance of
International Psychogeriatrics, 12(1),March 2000
L. A . Gerdner
minimum competency. Results ranged
from 87.5 to 100 with a mean of 92.4.
The research assistants completed the
anecdotal notes form in conjunction with
the MCMAI. This form was used to doc-
ument extraneous variables that may
have affected the subject’s behavior be-
yond the music intervention. It was also
used to document positive responses that
the subject expressed during the obser-
vation period that were not captured by
the MCMAI.
Procedure
Subjects were followed for an 18-week
period of time. During Week 1,data were
obtained by trained research assistants
via the demographic data form and the
GDS. In addition, certified nursing as-
sistants completed the TPAA over a
5-day period and family members com-
pleted the Modified Hartsock Music Pref-
erence Questionnaire (Hartsock, 1982).
During Weeks 2 through 4, trained
research assistants used the MCMAI to
assess the frequency of agitated behav-
ior over a 60-minute period that incor-
porated the subject’s peak level of
agitation. This baseline assessment was
conducted 2 days per week for 3 weeks.
A crossover design was used to ran-
domly assign subjects to either Group A
( n = 16) or Group B ( n = 23). Groups were
constructed by matching subjects on the
variables of age, gender, and degree of
cognitive impairment. Subjects in Group
A received 6 weeks of individualized
music, followed by 2 weeks of no music
(”washout” period). The purpose of the
”washout” period was to nullify possi-
ble cumulative effects of the music in-
tervention. To detect cumulative effects
prior to implementation of the second in-
tervention, research assistants completed
Effects of Music on Agitation
an agitation assessment during this 2-
week interval. In addition, during the
first week of the “washout” period, cog-
nitive assessment was repeated using
the GDS, because a deterioration in cog-
nitive status may enhance agitation.
After the ”washout” period, subjects
in Group A received 6 weeks of music
entitled Meditation - Classical Relaxation
Vol. 3. This is an anthology of classical
music by a variety of composers (i.e.,
Grieg, Beethoven, Schubert), judged by
a team of music therapists to be soothing
in effect and sedative in quality. A pref-
erence for classical music by one subject
warranted an alternative selection that
included an anthology of instrumental
Lawrence Welk music.
Group B received the same protocol in
reverse order. Group B initially received
6 weeks of music entitled Meditation -
ClassicaI ReZaxafion VoZ. 3 followed by a
2-week ”washout” period with assess-
ment of agitation using the same proto-
col as stated above. Group B then
received 6 weeks of individualized
music.
Each session of music intervention was
presented ”free field” on an RCA porta-
ble audio cassette player (RP-7668) for
30 minutes, 2 days per week (i.e., Mon-
day/Thursday, Tuesday/Friday). The
intervention occurred in an area of the
long-term-care facility where the subject
spent the majority of his/her time. Re-
search assistants observed and docu-
mented behaviors of subjects during the
presentation of music and for the one-
half hour immediately following the in-
tervention using the MCMAI. While
research assistants recorded the frequen-
cy of agitated behaviors, they also took
anecdotal notes on extraneous variables
that may affect the subject’s behavior
beyond the music intervention.
55
RESULTS
Data were analyzed using the Statistical
Analysis System (SAS),version 6.12, for
the personal computer. Because of the
design of this study, a number of prelim-
inary analyses were necessary prior to
hypotheses testing.
Because a crossover design was used,
the independence between subjects in
Group A and Group B was first deter-
mined. An alpha of 2 0 was established
to control for a Type I error and to ensure
that similarities between comparisons
would be found. A Fisher’s Exact Test
was used to compare categorical data,
an independent samples t test for the
continuous variable of age, and the Wil-
coxon Rank Sums Test for ordinal data
from the GDS. The two groups did not
differ significantly on 14 of the 16 demo-
graphic variables measured. Although
significant differences emerged on uri-
nary incontinence and assistance nued-
ed with ambulation, these differences
were not associated with the variables of
gait and toileting. On the basis of this
maintenance of functionality (gait and
activity), the groups were determined to
be essentially similar with regard to de-
mographic profile.
Subsequent analyses for this study
involved data from the MCMAI. On
occasion, data were missing from the
MCMAI. This usually was limited to a
10-minute period when, for example,
staff assisted the subject to the bath-
room, which prevented direct observa-
tion and documentation of agitated
behaviors. Consequently, least square
means were calculated using individu-
al actual values to predict missing
values. Calculations were based on pre-
dicted values of the fitted mixed effects
model.
56
Least square means were used to con-
duct a repeated measures analysis of
variance (ANOVA) (mixed procedure)
with Bonferroni post hoc analysis to
determine if a significant difference oc-
curred between the baseline and "wash-
out" periods. An alpha of .20 was
established for reasons previously not-
ed. No statistically significant differenc-
es occurred between the baseline and
"washout" periods in any of the six 10-
minute intervals in either Group A or
Group B. On the basis of this finding,
only baseline measures were used for
hypotheses testing.
In addition, a repeated measures
ANOVA (mixed procedure) with Bon-
ferroni post hoc analysis was used to
determine if an order effect occurred
regarding the presentation of music for
subjects in Group A compared to that in
Group B. No significant difference was
found between groups within each of
the 10-minute increments using an al-
pha of .20. Therefore, subjects in Group
A were combined with subjects in Group
B for the purpose of hypotheses testing.
The least square means of the frequen-
cy of agitation for subjects during base-
line, classical music, and individualized
music are numerically presented in
Table 1and graphically displayed in Fig-
ure 2. The first 30 minutes (0-30) repre-
sents the period in which music was
TABLE 1. Least Square Means of the Frequency of Agitation for Subjects
in Group A and Group B ( N = 39) During Baseline, Classical Music, and
Individualized Music
Minutes (Per 10-Minute Increments)
Data Collection Phases 0-10
Baseline ( S E = 2.1) 14.0
Classical ( S E = 1.3) 12.2
Individualized ( S E = 1.3) 6.4
L. A . Gerdner
played and the corresponding baseline
assessment. The second 30 minutes (30-
60) represents the 30-minute period fol-
lowing the presentation of music and the
corresponding baseline assessment.
Hypothesis testing was conducted using
a repeated measures ANOVA (mixed pro-
cedure) with the following factors: minute
(per 10-minute increments), phase (base-
line, individualized music, classical music),
and week (1through 6).An alpha of .05was
established for all hypotheses testing. A
significant two-way interaction occurred
between phase and minute. The main ef-
fectsof phase and minute were also identi-
fied as significant (refer to Table 2).
A Bonferroni post hoc analysis (see
Table 3) revealed that individualized
music was associated with a significant-
ly greater decrease in the frequency of
agitated behaviors compared to classical
music during each of the three 10-minute
intervals in which music was played (0-
30 minutes) and the 30-minute postin-
tervention period (30-60 minutes) (refer
to Table 3). Therefore, Hypotheses 1and
2 were both accepted.
To completely explore the phase/
minute interaction, a Bonferroni post hoc
analysis was conducted to compare base-
line agitation measure to the immediate
effects (0-30 minutes) of classical and
individualized music and the 30-minute
postintervention period (30-60minutes).
10-20 20-30 30-40 40-50 50-60
~~ ~ ~~
16.2 17.3 17.7 18.5 17.8
12.5 12.2 13.8 14.7 15.5
6.3 5.8 7.8 9.7 10.7
Ejfects of Music on Agitation 57
4

.. .. .r
..... . ..I. . . . . . . . .... . . . . . . . . . _ . . . . . . .I
E n

+Baseline
-0- Classical
-0-Individualized

U i I 1 ' 1 ' 1 ~ 1 ~ 1 ~ 1 ' 1

0 10 20 30 40 SO 60
(Minutes)
Figure 2. Least square means of frequency of agitation for subjects in Group A and Group B
(N = 39) during baseline, classical music, and individualized music.

TABLE 2. Repeated Measures ANOVA (Mixed

Procedure) on the Immediate and Residual Effects of
Individualized vs. Classical Music Compared to Baseline
Factor NDF DDF F p Value
~ ~~~~

Phase 2 74 32.92 .0001"

Minute 5 2763 53.28 .0001*
Phase/week 12 442 0.74 .7090
Phase/ minute 10 2763 4.77 .0001*
Phase/week/minute 60 2763 0.64 ,9852
Note. ANOVA = analysis of variance; DDF = denominator degrees of free-
dom; NDF = numerator degrees of freedom.
*p < .05.

As shown in Table 4, the frequency of following the presentation of individu-

agitated behaviors was significantlyless alized music compared to the corre-
during each of the three 10-minute incre- sponding three 10-minute intervals of
ments in which individualized music was baseline assessment.
presented. In addition, the frequency of As shown in Table 5, no significant
agitated behaviors was significantly difference occurred in the frequency of
less during the 30 minutes immediately agitation between baseline and the first
58 L. A. Gerdner
TABLE 3. Comparison of Individualized Music to
Classical Music per 10-Minute Increments Using
Bonferroni Post Hoc Analysis
Mean Difference
Between
Individualized and Bonferroni
Classical Music Adjusted
Minutes ( S E = 1.155) t p Value
0-10 -5.823 -5.04 < .0001*
10-20 -6.197 -5.37 < .0001*
20-30 -6.419 -5.56 < .0001*
30-40 -5.996 -5.19 < .0001*
40-50 -5.042 4.37 < .0002*
50-60 4.754 4.12 < .0001*
Note. ddf = 2763.
*p < .05.

20 minutes of classical music. A signifi-
cant decrease in agitated behaviors did
occur during the final 10 minutes of clas-
sical music compared to baseline. This
significant reduction continued only
during the first 10 minutes after the pre-
sentation of classical music.
CASE STUDY
Anecdotal notes provided qualitative
data that extended beyond the MCMAI
measurements. When used to supple-
ment quantitative findings, the qualita-
tive data provided a more holistic view
of the subjects' response to each music
intervention. The following case study is
provided to evaluate the clinical signifi-
cance of individualized music in addi-
tion to the statistical significance. It is
accompanied by a graphic display of the
mean frequencyof agitation during base-
line, individualized music, "washout"
period, and classical music.
Subject 1was a 75-year-old widowed
White woman residing in a special care
unit. She had a medical diagnosisof prob-
able Alzheimer's disease with a GDS
score of 6. Although English was the
subject's primary language, she also
spoke German and had previously lived
in a predominantly German ethnic com-
munity. The subject's daughter stated
that her mother enjoyed listening to
German polka music. Favorite songs
included the following: "Do Do Liest
Mier Inhansen" and "Tanta Anna." An
audio cassette entitled All The Best From
Germany was selected for the subject.
Prior to the presentation of music, the
subject exhibited frequent episodes of
crying and looking for family members,
and occasionally told her roommate to
"shut up" in a loud aggressive voice.
Verbalizations included a combination
of German and English. In addition,
Subject 1 exhibited frequent wander-
ing, general restlessness, and repeti-
tive mannerisms.
Upon initiation of individualized mu-
sic, the subject would frequently smile
and make statements such as, "My, isn't
this nice music we are blessed with,"

International Psychogeriatrics, 12(1),March 2000

Effects of Music on Agitafion 59
TABLE 4. Comparison of Individualized Music to
Baseline per 10-Minute Increments
Mean Difference
Between
Individualized Bonferroni
Music and Baseline Adjusted
Minutes (SE = 1.264) t p Value
0-10 -7.579 -5.99 < .0001*
10-20 -9.868 -7.81 < .0001*
20-30 -11.515 -9.11 < .0001*
30-40 -9.956 -7.88 < .0001*
40-50 -8.783 -6.95 < .0001*
50-60 -4.754 -5.63 < .0001*
Note. ddf = 2763.
*p < .05.

TABLE 5. Comparison of Classical Music to

Baseline per 10-Minute Increments
Mean Difference
Between Classical Bonferroni
Music and Baseline Adjusted
Minutes ( S E = 1.267) f p Value
0-10 -1.757 -1.39 > .9999
10-20 -3.671 -2.90 .0666
20-30 -5.096 -4.03 .0010*
30-40 -3.960 -3.13 .0324*
40-50 -3.740 -2.96 .0558
50-60 -2.361 -1.87 > .9999
Note. ddf = 2763.
p < .05.

and ”It’s so beautiful ...I could listen to
this all day and night.” In addition, she
would frequently sing along with the
German lyrics and clap in rhythm to the
melody. The most remarkable response
to individualized music occurred on a
day when the subject was extremely ag-
itated. She was asking God to take her to
heaven and repeatedly asked about her
parents and siblings. She also exhibited
verbally aggressive behavior toward her
roommate. However, as soon as the
music started, she appeared visiblycalm-
er, stopped crying, and actually began
smiling.
As depicted in Figure 3, the frequency
of agitated behaviors dramatically de-
creased during the presentation of indi-
vidualized music, but began to increase
following the presentation of music.
During the 2-week “washout,” the subject
exhibited many of the same behaviors that
were observed during the baseline assess-
ment. The subject also exhibited similar
agitated behaviors during the presentation
of classical ”relaxation”music.
60 L. A . Gerdner

30

E
2
w
25
.-
erl
Y

7 20
cw
0 +Baseline
h
15
- 0 - Individualized
-A- “Washout”
1

ks 10
E
Q

$ 2 5

Time
Figure 3. Mean frequency of agitation for Subject 1during baseline, classical music, ”wash-
out,’’ and individualized music.

DISCUSSION within the first 10 minutes, and contin-

Findings of this study are discussed as
they relate to theory testing with empha-
sis on Propositions 4 and 2. This is fol-
lowed by a section on limitations and the
implications for research and practice.
The fourth proposition contends that
individualized music alleviates agitation
in persons with ADRD. This was ad-
dressed by Hypotheses 1and 2. Individ-
ualized music resulted in a significant
reduction in the frequency of agitated
behaviors compared to classical music
(relative to baseline).
In addition, a Bonferroni post hoc com-
parison revealed that individualized
music resulted in a statistically signifi-
cant reduction in the frequency of agita-
tion (compared to baseline), which began
ued throughout the presentation of indi-
vidualized music as well as during the
30 minutes immediately following the
presentation of individualized music. In
contrast, classical music did not begin to
have a statistically significant reduction
in agitation (compared to baseline) until
after 20 minutes of the intervention. A
significant reduction in agitation with
classical music was sustained only
through the first 10 minutes following
this intervention.
The second proposition states that
music evokes an individualized emo-
tional response within the listener that is
associated with personal memories.
None of the identified hypotheses spe-
cificallyaddressed this proposition. How-
ever, anecdotal notes provide qualitative

International Psychogeriatrics, 12(1), March 2000

Effects of Music on Agitation
data regarding emotional response in
support of this proposition. For exam-
ple, in the above case study, music elic-
ited reminiscence as evidenced by Subject
1’s conversation with other residents
during the presentation of music. On
one occasion, she fondly recalled a gen-
tleman who sang, played the piano, and
taught German songs to children in her
community while she was growing up.
A second example involved another
subject whose preferred music (Golden
Organ Favorites Featuring Bob Rals ton/fer-
ry Burke and the Magic Organ) came di-
rectly from the subject’s personal library,
as supplied by a family member. The
presentation of this music appeared to
evoke reminiscence as evidenced by the
statement “I used to play the organ-I
love this music!”
Another subject had frequently attend-
ed dance contests prior to the onset of
cognitive impairment. She preferred
music from the ”Big Band era.” An audio
cassette of music performed by the Glenn
Miller Orchestra, In the Digital Mood, was
selected. On one occasion, the subject
began dancing with an “imaginary part-
ner” when she heard the music. On an-
other occasion, she sang the words to
“Gal from Kalamazoo” while swaying
and snapping her fingers in rhythm to
the music and stated, ”I just gotta dance.”
After the completion of the intervention
on that day, she said, ”My, my, I guess
this dance is over.”
Although anecdotal notes supported
Proposition 2, rigorous testing is needed
that includes quantitative as well as qual-
itative data analysis. This might include
use of the Philadelphia Geriatric Center
Affect Rating Scale (Lawton et al., 1996),
a six-item instrument that uses direct
observation of nonverbal communica-
tion to measure both positive affect
61
(pleasure, interest, contentment) and
negative affect (anger, anxiety/fear, sad-
ness). Psychometric properties have been
established through initial testing.
Limitations
A nonprobability convenience sample
limits generalizability, even though ef-
forts were made to include diversity with
respect to gender and ethnicity. Women
constituted 77% of the sample. Five of
the nine male subjects enrolled in this
study resided in a veterans‘ home. Be-
cause the majority of residents in the
nation’s long-term-care facilities are fe-
male, this study’s sample is fairly
representative of the usual gender dis-
tribution. The majority of subjects in this
sample were White, reflecting the rural
midwestern setting of this study.
The risk of rater bias was present be-
cause the person who conducted the
intervention also observed and docu-
mented the frequency of agitated behav-
iors. However, the seven paid research
assistantswho collected data had no vest-
ed interest in the results of this study. To
help neutralize the risk of rater bias, an
interrater agreement of .87 was reached
during initial training. Periodic checks
were conducted to ensure maintenance
of this level of agreement.
This study addressed only the frequen-
cy of agitation and not the degree of
disruptiveness caused by these behav-
iors. Consequently, individual behaviors
were weighted equally (e.g., hitting ver-
sus negativism) even though the out-
comes may have differed in severity. A
more comprehensive method of evalua-
tion would include the establishment of
a set of criteria to rank these behaviors
based on their degree of severity. This
is a difficult component to measure
62
because of the inherent subjectivity and
the variability within each individual
behavior.
However, since the start of this study,
the Disruptive Behavior Scale was de-
veloped and addresses many of these
concerns (Beck et al., 1997). Beck and
colleagues (1997) conceptually define
disruptive behavior as that which re-
sults in negative consequences for the
resident, caregiver, or other residents.
Efforts were made to operationally de-
fine behaviors. Initial research has es-
tablished psychometric properties. This
instrument could be used to measure the
dependent variable in future studies that
further explore the effects of individual-
ized music.
Research Implications
Aspects of the mid-range theory of IMIA
were supported by the preliminary find-
ings. No one test will definitively refute
or substantiate a theory. Theoretical test-
ing is rather the weight of accumulated
knowledge from a variety of studies.
The mid-range theory of IMIA provides
a framework for future research. It is
presented as a heuristic mechanism to
encourage additional testing and refine-
ment and thereby as a means of building
knowledge and improving practice.
Additional work is under way in this
effort. A graduate nursing student at the
University of Texas is developing a re-
search proposal for additional testing of
the mid-range theory of IMIA. She is
planning to utilize a long-term-care fa-
cility whose patient population is large-
ly Hispanic. In addition, a doctoral
student from The Netherlands is using
the mid-range theory of IMIA as the
theoretical framework for his disserta-
tion. When completed, these studies will
International Psychogeriatrics, 12(1),March 2000
L. A . Gerdner
provide additional theoretical testing
with culturally diverse populations.
Research utilization is another ave-
nue for additional testing of mid-range
theories (Blegen & Tripp-Reimer, 1997).
Outcome measures could include not
only the direct effects of individualized
music on the frequency of patient agita-
tion but also the indirect effect on the
incidence of falls and use of chemical
and physical restraints. The evaluation
of research utilization also involves staff
and family outcomes regarding their
perception of the effectiveness of the in-
tervention. In addition, an intervention
that promotes the effective management
of problematic behaviors may result in a
decrease in scheduled staffing hours and
overtime.
Implications for Practice
The findings of this study support the
use of individualized music as an alter-
native approach to the management of
elderly patients who are confused and
agitated. The decision to implement this
intervention into clinical practice in-
volves weighing the potential risks and
benefits. If the patient exhibits a nega-
tive response (i.e., increased agitation),
the music should be stopped. In addi-
tion, the clinician should monitor other
patients in the immediate area who may
respond negatively or find the music
offensive(Gerdner, 1999).Potential ben-
efits for the use of individualized music,
as presented in this study, far exceed the
risks. Since this study, the nursing staff
at the participating facilities have imple-
mented individualized music on a regu-
lar basis with a number of patients.
Music preference may be incorporat-
ed into the initial assessment of each
patient upon admission to a long-term-
Effects of Music on Agitation
care facility. Information can be en-
hanced by collaboration from family
members. A key factor in the success of
this intervention is the identification of
music preference. Information should be
as specific as possible to include song
titles, performers, and preference for
vocal or instrumental music (piano, or-
chestra, guitar, etc.) (Gerdner, 1997,
1999). Family members may be able to
provide a favorite album from the pa-
tient’s personal collection, which can
then be transferred to audio tape and
returned. Music may also be obtained
from libraries and philanthropic groups
(Gerdner & Buckwalter, 1999).
A cassette player can be placed at the
patient’s bedside or checked out from a
central location, such as the nurses’ sta-
tion, for use as needed (Gerdner & Buck-
Walter, 1999).This approach is relatively
inexpensive and requires minimal time
expenditure. After instruction by nurs-
ing staff, music may be implemented by
nursing assistants, activity staff, volun-
teers, and family members. An ongoing
assessment should be conducted to de-
termine the patient’s response. Positive
outcomes are anticipated to reinforce the
use of music as one alternative approach
to managing patients who are agitated
or confused.
REFERENCES
Aldridge, D. (1993). Music and Alzheimer’s
diseaseassessment and therapy: Discus-
sion paper. Journal of the Royal Society of
Medicine, 86/93-95.
Beck, C., Frank, L., Chumbler, N. R.,
O’Sullivan, P., Vogelpohl, T. S., et al.
(1998). Correlates of disruptive behav-
ior in severely cognitively-impaired
nursing home residents. The Gerontolo-
gist, 38, 189-198.
63
Beck, C., Heithhoff, B., Baldwin, B., Cuffel, B.,
OSullivan, P., et al. (1997). Assessing
disruptive behavior in older adults: The
disruptive behavior scale. Aging and
Mental Health, 1(1),71-79.
Blegen, M. A., & Tripp-Reimer,T. (1997).Nurs-
ing theory, nursing research and nursing
practice: Connected or separate? In J. C.
McCloskey & H. K. Grace (Eds.), Current
issues in nursing (5th ed., pp. 68-74).
St. Louis, MO: Mosby Year Book.
Bradley, L., Siddique,C.M., & I o n , B. (1995).
Reducing the use of physical restraints in
long-term care facilities. Journal of Geronto-
logical Nursing, 21(9), 21-34.
Bright, R. (1982). Music geriatric care. USA:
Musicgraphics.
Burnside, I. (1988). Dementia and delirium.
In I. Burnside (Ed.), Nursing and the aged
(pp. 732-794).New York McGraw-Hill.
Cohen-Mansfield, J. (1986). Agitated behav-
iors in the elderly 11. Preliminary results
in the cognitively deteriorated. Journal of
the American Geriatrics Society, 34,722-727.
Cohen-Mansfield, J., & Billig, N. (1986). Agi-
tated behaviors in the elderly I. A
conceptual review. Journal of the American
Geriatrics Society, 34,711-721.
Cohen-Mansfield, J., Culpepper, W. J., &
Werner, P. (1995). The relationship be-
tween cognitive function and agitation
in senior day care participants. Inferna-
tional Journal of Geriatric Psychiatry, 20,
585-595.
Cohen-Mansfield,J., Marx, M., & Rosenthal, A.
S. (1989). A description of agitation in a
nursing home. Journal of Gerontology, 44,
77-84.
Cohen-Mansfield, J., & Mam, M. S. (1984). Do
past experiences predict agitation in nurs-
ing home residents? International Journal of
Agingand Human Development, 28,285-293.
Denny, A. (1997).Quiet music an intervention
for mealtime agitation. Journal of Geronto-
logical Nursing, 23(7), 16-23.
Deutsch, L. H., & Rovner, B. W. (1991).Agita-
tion and other noncognitiveabnormalities
in Alzheimer’s disease. The Psychiatric
Clinics of North America, 14,341-351.
64
Devereaux, M. A. (1997). The effects of indi-
vidualized music on cognitively impaired
nursing home residents exhibiting agita-
tion. Master’s thesis, College of St.
Catherine, St. Paul, MN.
Gates, A., & Bradshaw, J. L. (1977).The role of
the cerebral hemispheres in music. Brain
and Language, 4,403-431.
Gerdner, L. (1997). An individualized music
intervention for agitation. Journal of the
American Psychiatric Nurses Association, 3,
177-184.
Gerdner, L. A. (1992).Theefects ofindividualized
music on elderly patients who are confused
and agitated. Master’s thesis, The Univer-
sity of Iowa, Iowa City, IA.
Gerdner, L. A. (1998).Individualized U S . classical
music on agitation in Alzheimer’s disease
and rela ted disorders. Doctoral dissertation,
The University of Iowa, Iowa City, IA.
Gerdner, L. A. (1999). Individualized music
intervention protocol. Journal of Geronto-
logical Nursing, 25(10), 10-16.
Gerdner, L. A., & Buckwalter, K. C. (1999).
Music therapy. In G. M. Bulechek & J. C.
McCloskey (Eds.), Nursing interventions:
Efective nursing treatments (3rd ed., pp.
451-468). Philadelphia: Saunders.
Goddaer, J., & Abraham, I. L. (1994).Effects of
relaxing music on agitation during meals
among nursing home residents with se-
vere cognitive impairment. Archives of
Psychiatric Nursing, 8, 150-158.
Hall, G . R., & Buckwalter, K. C. (1987). Prog-
ressively lowered stress threshold: A
conceptual model for care of adults with
Alzheimer’s disease.Archives ofPsychia tric
Nursing, 1,399-406.
Hartsock, J. W. (1982). The effects of music on
levels of depression in orthopedic patients on
prolonged bed rest. Master’s thesis, The Uni-
versity of Iowa, Iowa City, IA.
Lawton, M. P., Van Haitsma, K., & Klapper, J.
(1996). Observed affect in nursing home
residents with Alzheimer’s disease. Jour-
nal of Gerontology, 52B, P3-Pl4.
Petsche, H., Lindner, K., Rappelsberger, P., &
Gruber, G. (1988). The EEG: An adequate
International Psychogeriatrics, 12(1), March 2000
L. A . Gerdner
method to concretize brain processes
elicited by music. Music Perceptions, 6,
133-159.
Ragneskog, H., Kihlgren, M., Karlsson, I., &
Norberg, A. (1996). Dinner music for
demented patients: Analysis of video-
recorded observations. Clinical Nursing
Research, 5,262-277.
Reisberg, B., Ferris, S. H., deleon, M. J., &
Crook, T. (1982). The global deterioration
scale for assessment of primary de-
generative dementia. American Journal of
Psychiatry, 139,1136-1139.
Sloane, P. D., Mathew, L. J., Scarborough, M.,
Desai, J. R., Koch, G. G., etal. (1991).Physi-
cal and pharmacologic restraint of
nursing home patients with dementia.Jour-
nal of the American Medical Association, 265,
1278-1282.
Swanson,E.,Maas,M.,&Buckwalter,K.(1992).
Catastrophic reactions of Alzheimer’s
patients: Special unit compared to tradi-
tional units. Archives ofPsychiatric Nursing,
7,292-299.
Swartz, K. P., Hantz, E. C., Crummer, G. C.,
Walton, J. P., & Frisina, R. D. (1989).Does
the melody longer on? Music cognition
in Alzheimer’s disease. Seminars in
Neurology, 9, 152-157.
Tabloski, P. A., McKinnon-Howe, L., &
Remington, R. (1995). Effects of calming
music on the level of agitationincognitively
impaired nursing home residents. The
American Journal of Alzheimer’s Care and
Related Disorders and Research, 10(1),10-15.
Thaut, M. H. (1990).Neurophysiological pro-
cesses in music perception and their
relevance in music therapy. In R. F. Unkefer
(Ed.), Music therapy in the treatment of
adults with mental disorders (pp. 3-32).New
York: Schimer Books.
Tinetti, M. E., Liu, W., & Ginter, S. F. (1992).
Mechanical restraint use and fall-related
injuriesamong residents of skillednursing
facilities. Annals of Internal Medicine, 116,
369-374.
Walton,J. P., Frisina, R. D.,Swartz, K. P., Hantz,
E., & Crummer, G. C. (1988). Neural basis
Effects of Music on Agitation
for music cognition: Future directions and
biomedical implications.PsychornusicoZogy,
7, 127-138.
Acknowledgments.This work was supported
in part by grants from the National Insti-
tute for Nursing Research (F31NR07090,
IR15NR03728); the 1996-1997 American
Association of Retired Persons Andrus
65
Foundation Fellowship in Gerontology;
and a grant from the University of Iowa
Student Government. The author would
like to thank the following individuals for
their support and guidance in this en-
deavor: Kathleen C. Buckwalter, PhD, RN,
FAAN, Toni Tripp-Reimer, PhD, RN,
FAAN, Elizabeth Swanson, PhD, RN, and
Cornelia Beck, PhD, RN, FAAN.

Are you interested in sending a paper

to International Psychogeriatrics?
IPAs journal is read and respected by leading
researchers, clinicians, and educators in psychogeriatrics.
Submission of original, previously unpublished material
is invited in two main areas:
Research and Reviews
Clinical Practice and Service Development

Guidelines for contributing to lntemational Psychogeriatrics

can be found inside the back cover of this issue and at
IPAs web site, http://www.ipa-online.org

All submissions should be sent to:

Dr. Robin Eastwood, Editor-in-Chief
International Psychogeriatrics
Department of Psychiatry, St. Louis University
1221 S. Grand Boulevard
St. Louis, Missouri 63104 USA

View publication stats