Professional Documents
Culture Documents
DOI 10.1007/s11940-013-0257-2
Nonpharmacologic Treatment
of Behavioral Disorders
in Dementia
Jiska Cohen-Mansfield, PhD1,2,3
Address
1
Department of Health Promotion, School of Public Health, Sackler Faculty of
Medicine, Tel-Aviv University, Ramat Aviv, P.O.B. 39040, Tel-Aviv 69978, Israel
2
Herczeg Institute on Aging, Tel-Aviv University, Tel-Aviv, Israel
3
Minerva Center for the Interdisciplinary Study of End of Life, Tel-Aviv University,
Tel-Aviv, Israel
Email: jiska@post.tau.ac.il
Agitation
Treatment
Nonpharmacologic
Opinion statement
Dementia symptoms are often complicated by behavioral disorders such as repetitive
verbalizations, aggressive behavior, and pacing. In clinical practice, the most common responses to behavioral disorders are pharmacologic, mostly using antipsychotic medication, or
ignoring the behavior. However, multiple research studies support the notion that these behavioral disorders in dementia are related to unmet needs that can be addressed by
nonpharmacologic interventions. Persons with dementia present multiple unmet needs, most
commonly pain and discomfort, need of social contact and support, and need of stimulation that
alleviates boredom. A wide range of interventions that address these needs has been investigated, though the rigor of the investigations varied greatly depending on factors related to the behavioral disorder, setting, and resource limitations. In practice, the avenues to address the
unmet needs should depend on the persons abilities and preferences. Thus, nonpharmacologic
interventions that are individually tailored to the person with dementia comprise a superior response to behavioral disorders and should be at the frontline of treatment of these disorders.
Introduction
Conceptualization of behavioral disorders
in dementia
Behavioral disorders in dementia encompass a range
of observable behaviors displayed by Persons With
Dementia (PWDs) that are socially unusual or inap-
766
Multiple factor analyses [1] have shown that behavioral disorders can be described along behavioral
dimensions, namely: aggressive vs nonaggressive, verbal vs physical, and frequency of display (never to
all the time). Subtypes of behavioral disorders are
therefore physically aggressive behaviors, such as hitting or kicking others; physical nonaggressive behaviors, such as pacing back and forth or handling
things inappropriately; verbally aggressive behaviors,
such as cursing; and verbal nonaggressive behaviors,
such as repetitious verbalizations. Both correlational
and longitudinal studies have shown that different subtypes of behavioral disorders are associated with different demographic and health variables. For example,
verbal manifestations, but not physical nonaggressive
behaviors, have been associated with medical problems
or pain. Aggressive behaviors tend to occur in the very
last stages of dementia [2, 3].
Conceptually, it is important to determine whether
behavioral disorders are symptoms of dementia and
therefore integral to the dementia syndrome, or alternatively, represent disease symptoms that are not specific to dementia but may stem from inappropriate
treatment of dementia and reflect symptoms of discontent or other difficulties. As will be seen in the next
section, different theoretical frameworks present different answers to this question.
Treatment
Pharmacologic treatment
Drug therapy for behavioral disorders is based on the biological theoretical framework described above and therefore aims to decrease behavioral
disinhibition by changing the balance of neurotransmitters. The most
common class of drugs for pharmacologic treatment of persistent and
pervasive behavioral disorders is antipsychotic medication [1012], which
has severe side effects including increased mortality rates [10, 1315].
Consensus statements and guidelines recommend limiting the use of antipsychotics in all settings [1618]. Additional information on the use of
antipsychotics and the pharmacologic approach to treating behavioral
disorders in dementia is discussed in the accompanying topic of dementia
with Lewy bodies in this edition.
Cohen-Mansfield
767
Nonpharmacologic interventions
The goals of nonpharmacologic interventions depend on the theoretical model
used for the intervention. Most interventions aim to fulfill unmet needs,
whereas some aim to decrease the level of stimulation (eg, by playing peaceful
music) or to modify behaviors by changing contingencies or by stimulus control
(eg, by camouflaging a bothersome stimulus). Some interventions may be appropriate for multiple objectives. Regardless of the stated goal, the actual
mechanism of action is often unclear. For example, music may be provided as
sensory stimulation to combat boredom, but, if the music is presented by a
musician who provides participants with social contact, the social contact may
be the most important active ingredient of the intervention.
In a 2001 review [19] the following nomenclature for nonpharmacologic
interventions was set forth: sensory, structured activities, social contact - real
or simulated, medical/nursing care interventions, environmental interventions, behavior therapy, staff training, and combination therapies. While the
goal and the theoretical formulation may vary among interventions, general
principles can be ascertained. The first four intervention categories are based
on the unmet needs model. Specifically, sensory interventions and activities
are geared to ameliorate boredom and sensory deprivation (Fig. 1); social
contact interventions aim to alleviate loneliness (Fig. 2); and medical and
nursing interventions aim to address pain and discomfort (Fig. 3). In con-
Figure 1. Unmet needs of boredom, sensory deprivation or relaxation: examples of nonpharmacologic interventions.
(Copyright 2013, Jiska Cohen-Mansfield, PhD).
768
Figure 2. Unmet need of loneliness: examples of nonpharmacologic interventions. (Copyright 2013, Jiska Cohen-Mansfield, PhD).
Cohen-Mansfield
769
Fig. 4. Examples of environmental interventions for behavioral disorders in dementia. (Copyright 2013, Jiska CohenMansfield, PhD).
770
Aromatherapy
There are relatively few studies of aromatherapy interventions. A review of
11 randomized controlled trials [26] concluded that aromatherapy is potentially useful for treating behavioral disorders. A meta-analysis of two
studies [27, 28] in this review found a significant effect for aromatherapy
treatment. Interestingly, a replication of one of these studies, which included an improved experimental design (ie, blinding of treatment arms)
and a different assessment, did not find Melissa aromatherapy to be superior to placebo [29]. Overall, despite some positive results, current evidence
is inconclusive due to the small number of studies and methodological
limitations [26, 30].
Massage
A review of massage interventions for behavioral disorders in persons with
dementia [31] found only one study with adequate methodology [32] out of
13 studies that met review criteria. This study demonstrated a significant
reduction in behavioral disorders following massage in a sample of 52 persons with dementia. There was no comparison group. Several individual
studies support the impact of massage on behavioral disorders. Foot massage
has been associated with a significant decrease in behavioral disorders in 22
persons with dementia [33]. Massage was also associated with a significant
decrease in aggressiveness, but not in other behavioral disorders, in a group
of 20 persons with dementia. No significant changes in behavioral disorders
were found in the control group (n=20) [34]. In a study of 65 nursing home
residents with dementia, therapeutic touch significantly decreased restlessness, but not other behavioral disorders, in comparison to a control group
receiving routine care, and there was no significant reduction of behavioral
disorders in comparison with a placebo mimic treatment [35]. Taken together, massage intervention studies all suffer from methodological shortcomings or inconclusive results, often referring to a single behavioral
symptom. Nonetheless, they support the promise of massage and touch in
decreasing behavioral disorders in persons with dementia.
Multisensory interventions
Conflicting results have been found in studies of Snoezelen among persons
with dementia, with some studies reporting some effect on behavior but
others finding no effect [30].
Structured activities
The positive impact of activities, such as arranging flowers or arts and crafts,
has been demonstrated by several studies. In a comparison of different
Cohen-Mansfield
771
772
Human contact
One-on-one human contact with a PWD by a trained individual has been
repeatedly found to be a highly potent intervention for behavioral disorders
[37, 45]. Humor therapy can serve as a source of social interaction, and it
was shown to significantly decrease agitation in a cluster randomized trial
[46] that utilized both trained staff and professional performers to deliver a
humor therapy intervention.
Animal-assisted interventions
A review of 18 studies of animal assisted interventions with persons with
dementia [47] shows that these interventions may have the potential to reduce behavioral disorders. Ten studies in this review examined the impact of
animal assisted therapy on behavioral disorders in dementia, of which three
reported statistically significant decreases in behavioral disorders [48, 49] or
aggressiveness [50]. The remaining studies either did not report a statistical
analysis or else reported nonsignificant results. While this review [47] suggested that robotic pets may be as effective as real live pets in terms of animal
assisted therapy, the evidence for this is still lacking [47, 51].
Cohen-Mansfield
773
774
Environmental interventions
A previous review on nonpharmacologic interventions [19] included
studies showing some support for the impact of enhanced environments,
such as those simulating natural conditions (eg, pictures of outdoors,
bird sounds) or a home environment in decreasing behavioral disorders.
Access to an outdoor garden, access to refreshments, and a quiet unit
have all been reported to decrease behavioral disorders [19]. Much of the
more recent literature on the impact of environmental change on PWDs
has not focused directly on behavioral disorders, but rather on constructs
that may either trigger behavioral symptoms or be incompatible with
them. For example, specific environmental features, such as small number of residents per unit, no change in direction within the unit layout,
and provision of only one living or dining room were related to better
way finding by residents [68]. Better way finding is likely to result in
more autonomy and less frustration and thereby the manifestation of
fewer behavioral symptoms. Similarly, design features, such as small unit
size, that are related to improved quality of life have been identified
[69]. Improved quality of life is generally incongruent with behavioral
disorders [70]. The same review [69] reported that decreasing the institutional characteristics of residential facilities was associated with fewer
behavioral disorders and that a wall mural over an exit door decreased
exit attempts. Finally, bright light therapy, used in a study of 66 older
PWDs in two residential care settings, showed no benefit and a potential
exacerbating effect and remains a controversial modality for behavioral
disorders in dementia [71].
Behavior therapy
Previous studies have showed the efficacy of differential reinforcement and
stimulus control in treating behavior disorders yet the evidence is limited
because most used a single subject design or very small samples [19]. In
recent years there seems to be a dearth of studies using behavior therapy
compared with those investigating other approaches. Yet, the few studies
conducted have generally supported behavior therapy for treating behavior
disorders. For example, a single subject design demonstrated the impact of
both stimulus control and differential reinforcement on wandering behavior
[72], and a pilot study using a pre-post design reported significant improvement in behavioral disorders [73].
Cohen-Mansfield
775
776
Cohen-Mansfield
777
ferent needs and even the needs of a single PWD will vary by time. Moreover,
the proper response to a need is unique and is a function of the persons
abilities, habits, and preferences. Therefore, any study that uses the same
specific intervention for all participants is likely to have a moderate impact at
best. An effort, instead, is needed to provide a proof of concept or feasibility
for a wide range of potential solutions. Efficacy should be examined for research of algorithms utilizing these solutions or for research of change of
culture of practice style, which also likely would utilize these solutions.
From a methodological point of view, the determination of whether a trial
shows a significant effect often seems to depend as much on the procedures
used as on the results. The label of a nonpharmacologic intervention does
not itself indicate the active ingredient(s) in that intervention. Clarifying the result of an intervention is therefore confounded by several factors. The quality
and frequency of interactions with the person administering the intervention
may have a major effect on its outcome. The impact of the intervention on
the caregiver is also important, as the intervention may require extra effort from
the caregiver, it may have a direct impact on the caregiver (eg, music) or, alternately, provide tools to care for the PWD or result in the PWD becoming more
amenable to care. Prior attitudes toward the intervention modality (music, pet,
manipulative object, game) can also affect the response to the intervention. The
setting of the intervention and dosage (eg, duration, timing, and frequency) can
all affect the response. It is extrinsic factors such as these that can lead different
studies and different reviews to reach different conclusions for ostensibly the
same intervention. Further clouding the issue are inadequate reviews that repeat
the results of other inadequate reviews to provide either a thoroughly vague and
ambiguous picture or to recite the need for additional research and reiterate the
known limitations of studies in this field. This approach is neither informative
nor useful for practice or research in this area.
A practical point of view would indicate that, as can be seen in Figs. 1, 2, and
3, nonpharmacologic approaches encompass a multitude of techniques and
procedures and, therefore, it is unlikely that sufficient resources will be available
to perform properly powered randomized controlled studies on all of these interventions. Moreover, each intervention encompasses many particular exemplars, such that a doll may be small or baby size, soft or plastic, baby-like or
not, of the same race as the PWD or a different one, etc, or a physical exercise
may differ by type of activity, by person providing it and the amount and type
of communication and support provided, by the number of PWDs included in
the activity, and other attributes. It is definitely impossible to study each of these variants in a large study. Studies comparing attributes of interventions to clarify the crucial dimensions, which impact efficacy may ultimately be more
useful. However, those studies are often not considered for inclusion in reviews
because they do not fit the current devotion to the RCT design [Cohen-Mansfield et al., unpublished manuscript].
778
Cost/cost-effectiveness
The dearth of information concerning cost-effectiveness of
nonpharmacologic interventions in dementia is due to the complexity of the
calculations and to the underdeveloped nature of the interventions themselves. Calculating the cost effectiveness of the interventions would need to
take into account any system changes (eg, training and mentoring caregivers,
providing necessary materials, etc) required to make nonpharmacologic interventions a part of daily life in the care of the PWD. Necessary funds could
potentially be achieved through savings in costs associated with pharmacologic care, currently the common practice in the clinical management of
behavioral disorders, such as medicine aids, physician calls, handling of
adverse events, etc. One study that did examine the cost-effectiveness [96] of
home visits by occupational therapists to develop customized activities and
train families in their use concluded that the program was cost-effective.
Cohen-Mansfield
779
780
Acknowledgment
This work was supported in part by the Minerva-Stiftung Foundation grant no. 31583295000.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
Cohen-Mansfield
781
782
40.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
Cohen-Mansfield
783
784
76.
Testad I, Ballard C, Bronnick K, Aarsland D. The effect of staff training on agitation and use of restraint
in nursing home residents with dementia: a singleblind, randomized controlled trial. J Clin Psychiatry.
2010;71:806.
This study describes the components of a successful staff
training program.
77.
Deudon A, Maubourguet N, Gervais X, Leone E,
Brocker P, Carcaillon L, et al. Nonpharmacological
management of behavioural symptoms in nursing
homes. Int J Geriatr Psychiatry. 2009;24:138695.
78.
Reuther S, Dichter MN, Buscher I, Vollmar HC, Holle
D, Bartholomeyczik S, et al. Case conferences as interventions dealing with the challenging behavior of
people with dementia in nursing homes: a systematic
review. Cambridge J. 2011.
79.
Jeon Y, Govett J, Low L, Chenoweth L, McNeill G,
Hoolahan A, et al. Care planning practices for behavioural and psychological symptoms of dementia
(BPSD) in residential aged care: a pilot of an education tool kit informed by the Aged Care Funding
Instrument (ACFI). Contemp Nurs. 2013:320949.
80.
Clare L, Whitaker R, Woods RT, Quinn C, Jelley H,
Hoare Z, et al. AwareCare: a pilot randomized controlled trial of an awareness-based staff training intervention to improve quality of life for residents
with severe dementia in long-term care settings. Int
Psychogeriatr. 2012:12839.
81.
Beer C, Horner B, Flicker L, Scherer S, Lautenschlager
NT, Bretland N, et al. A cluster-randomised trial of
staff education to improve the quality of life of
people with dementia living in residential care: the
DIRECT study. PloS one. 2011;6:e28155.
82. Burack OR, Weiner AS, Reinhardt JP. The impact of
culture change on elders behavioral symptoms: a
longitudinal study. J Am Med Dir Assoc.
2012;13:522528.
Although the results of this study are not clear cut, the paper
provides a useful overview of the components of the culture
change undertaking.
83. Johnson DK, Niedens M, Wilson JR, Swartzendruber
L, Yeager A, Jones K. Treatment outcomes of a crisis
intervention program for dementia with severe psychiatric complications: the Kansas Bridge Project.
Gerontologist. 2013;53:10212.
This paper describes a successful system change to address
crises in the care of PWD in the community.
84.
Cohen-Mansfield J, Parpura-Gill A, Golander H.
Utilization of self-identity roles for designing interventions for persons with dementia. J Gerontol B
Psychol Sci Soc Sci. 2006;61:20212.
85.
Cohen-Mansfield J, Jensen B. Self-maintenance
Habits and Preferences in Elderly (SHAPE): reliability
of reports of self-care preferences in older persons.
Aging Clin Exp Res. 2007;19:618.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
99.
100.
101.
Cohen-Mansfield
785