Content uploaded by Jenny Chan
Author content
All content in this area was uploaded by Jenny Chan on Dec 11, 2014
Content may be subject to copyright.
Article
Massage as therapy for
persons with intellectual
disabilities: A review of
the literature
Jenny Sau-Lai Chan
Tuen Mun Hospital, Hong Kong
Sonny Hing-Min Tse
Polytechnic University, Hong Kong
Abstract
Persons with intellectual disabilities record a higher incidence of challenging behaviours than the
general population. Massage therapy has frequently been applied to such persons to induce
relaxation. The aim of this review was to evaluate the effectiveness of massage therapy on
relaxation and reduction of challenging behaviours. A literature search was conducted through
various electronic databases; a total of 64 articles was generated but only seven met all inclusion
criteria. These studies were evaluated in terms of (a) number of participants, (b) study setting,
(c) methodology, (d) intervention pattern, (e) outcome measure of the intervention, and (f) major
results. Evidence-based literature demonstrating the effectiveness of massage therapy in support-
ing clinical practice is extremely limited. Case study designs, large amounts of qualitative data and
small sample sizes meant that the therapeutic effect of massage therapy could not be substantiated.
Hence, future studies with randomized clinical trials or of experimental design are recommended.
Keywords
challenging behaviours, intellectual disability, massage, relaxation, self-stimulating behaviours
Date accepted: 2/03/2011
Introduction
The condition of intellectual disability is lifelong and irreversible. Despite the variation in their
physical conditions, persons with intellectual disabilities are all vulnerable to psychological stress
Corresponding author:
Jenny Chan, Mental Handicap Unit, Tuen Mun Hospital, Ching Chung Koon Road, Tuen Mun, NT, Hong Kong. E-mail:
csl437@ha.org.hk
Journal of Intellectual Disabilities
15(1) 47–62
ªThe Author(s) 2011
Reprints and permission:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1744629511405105
jid.sagepub.com
and anxiety in relation to appraising and processing information, suffer from inadequate social
support, possess limited behaviour repertoire and have poor coping skills (Cooray and Bakala,
2005). When stress and anxiety cannot be vented, behavioural problems or self-injurious behaviour
may arise (Moss et al., 2000). In fact, the prevalence of challenging behaviour in persons with
intellectual disabilities is approximately three to five times higher than that in the general popula-
tion; the more serious the disability, the higher the prevalence will be (Janssen et al., 2002).
Reasons for challenging behaviours include a desire for social attention, and escape from the
demands of learning new skills and acquiring experiences that persons with intellectual disabilities
feel distressing. If challenging behaviours could be reduced, the potential for social integration and
productive learning would probably increase (McEvoy et al., 1987; Smith et al., 2005).
Other than challenging behaviours, self-stimulating behaviour is also common among persons
with severe to profound intellectual disabilities. One explanation of self-stimulating behaviour is
that it fulfils the drive of sensory input which is the human basic need for stimulation. Owing to
their severe cognitive deficits, such persons are unaware of external stimuli; hence, they generate
self-stimulating behaviour to satisfy this basic sensory need. Another reason that is similar to chal-
lenging behaviour is to escape from task demands or aversive situations (Durand and Carr, 1987).
Many studies have suggested that any interventions promoting relaxation can be of therapeutic
value (Deakin, 1995; Lindsay and Baty, 1986; Schilling and Poppen, 1983). The assumption is that
muscle relaxation counteracts anxiety, challenging behaviour and stereotypic self-stimulating
behaviour because relaxation response cannot simultaneously exist in agitated and restless persons;
consequently, it replaces maladaptive behaviours (Chan et al., 2010).
The conventional treatments for challenging behaviours are psychotropic medications and
behavioural programmes. However, their effectiveness is far from conclusive, especially in
refractory and severe cases. This indicates the need for an effective relaxation method. In fact,
massage therapy has been used as a form of relaxation for many years in different countries. For
persons with challenging behaviours, the specific purpose of using massage therapy is to divert
their attention to experience pleasure and induce relaxation so as to increase positive behaviours
with which they can earn social acceptance from caregivers and community members. Another
reason to espouse massage therapy is that it does not require the client’s cognitive capacity to
understand the process (McEvoy et al., 1987).
From the perspective of direct care staff, touch is not only a basic human need, but also a way to
foster social closeness, facilitating the communication of both affection and the upcoming event
(Forster and Iacono, 2008). It is particularly useful for persons with profound intellectual dis-
abilities as they have little or even no apparent understanding of verbal language and symbolic
interaction, and rely mostly on direct care staff to carry out their daily living activities (Nakken and
Vlaskamp, 2007). Touch thus becomes the clearest way for them to interact with the environment
and to receive information from the people around them.
Although massage therapy is often applied to persons with intellectual disabilities to promote
relaxation, increase awareness of the immediate environment, enhance mood and develop communication
through a trusting relationship, these positive assertions of massage therapy are mostly based on personal
clinical experiences or are incorporated into sensory programmes, and they lack independent assessment.
The objectives of this literature review are twofold. The first is to do a systematic review of the
research-based evidence on the use of massage therapy for persons with intellectual disabilities to sup-
port clinical practice. The second is to compare and contrast the work of related studies by tabulating
the detailsof each study. In the following review, the effectiveness and the physiological mechanisms
of massage therapy were examined closely in an effort to identify research gaps for future study.
48 Journal of Intellectual Disabilities 15(1)
Method
The review addressed the major findings of the studies in relation to relaxation and positive
behavioural changes after receiving massage therapy, e.g. reduced frequency of challenging
behaviours after massage therapy. In order to develop a better understanding of different behaviour
terms, typical patterns are described. Positive behaviours refer to adaptive behaviour, initiation of
communicative attempts, rapport building, engagement behaviour and concentration. Frequency of
on-task behaviour is greater than that of off-task behaviour (Chan et al., 2010). Challenging beha-
viours refer to those behaviours affecting the social order of the environment, violating others’
rights, and ignoring the safety of others and oneself (Emerson, 1995). Challenging behaviours
comprise self-injurious, stereotypic self-stimulating, and aggressive or destructive behaviours.
Self-injurious behaviour is defined as behaviour that causes damage to the person’s own body, for
instance, biting, slapping, eye/anal poking, head banging (Lambrechts and Maes, 2009; Singh
et al., 2004). Stereotypic self-stimulating behaviours are behaviours that look unusual, strange
or repetitive and have no purposeful function for daily living, such as body rocking or hand
flapping (Durand and Carr, 1987; Shapiro et al., 1997). Aggressive or destructive behaviours are
offensive actions or deliberate overt attacks directed towards other individuals or objects, such as
kicking, punching and hitting (Lambrechts and Maes, 2009; Singh et al., 2004). The use of mas-
sage therapy involves rhythmical stroking and mechanical manipulation of body tissues (Harris
and Lewis, 1994). Only articles that met the pre-set inclusion criteria were analysed and organized
into six areas, namely nature of participant, study setting, research design, intervention pattern,
assessment methods, and major findings.
Search procedures
A literature search was conducted through the Research Pro system and various electronic data-
bases, namely Academic Search Premier, EMBASE, ERIC, CINAHL, MEDLINE and PsycINFO.
No limitation was set on year of publication provided that massage use was reported. Synonymous
with intellectual disability are mental deficiency, mental handicap, mental retardation, learning
disability and developmental disability. Massage or touch therapy was used in the literature search
in combination with intellectual disability and its synonyms. The term ‘intellectual disability’ is
used in this review unless direct quotations from articles are presented.
Inclusion and exclusion criteria
Massage therapy is widely applied in the field of intellectual disabilities, but most seemingly
relevant articles simply report experience sharing or do not present new data. Only research-based
articles were selected to weigh against the inclusion criteria. The focus of the review targeted on
relaxation and behavioural changes following massage therapy. Outcomes involving medical prob-
lems were excluded from the review, for instance, massage therapy for the treatment of constipa-
tion (Moss et al., 2007) or bruxism (Rudrud and Halaszyn, 1981). The following selection criteria
were formulated.
Inclusion criteria.
The primary intervention of the study was massage therapy.
Massage therapy was conducted/taught by a trained massage therapist.
Subjects must have intellectual disability or its associated diagnosis, e.g. Down syndrome,
Cornelia de Lange syndrome.
Chan and Tse 49
Outcome measures focused on relaxation or behavioural change, for instance, increasing
positive behaviours or decreasing challenging behaviours.
At least one assessment method was used to evaluate the effectiveness of massage therapy.
Exclusion criteria.
Articles focused on description of massage therapy, personal clinical experience or medical
dimensions, e.g. treatment of constipation.
No assessment method was administered to validate the massage effect.
Intervention did not involve direct manipulation of soft tissue, like hugging or gentle touching.
Data extraction
The basic or preliminary keyword search was ‘massage’ or ‘touch therapy’. The term ‘intellectual
disability’ or its synonyms was then added at theadvance search level.The keywords of the articles had
to include synonyms of both massage therapy and intellectual disabilities. A total of 64 articles was
generatedfrom the databases; an overview of the article retrieval is shown in Table 1.In fact, no article
could be retrieved from some databases, e.g. the British Nursing Index and the Cochrane Library.
Some articles overlapped among different databases. Each identified article was assessed based on the
inclusion and exclusion criteria. Only seven articles were found that met all the inclusion criteria.
Each included article was organized in terms of (a) sample, including number of participants,
(b) study setting, (c) methodology, (d) intervention pattern, (e) outcome measure of the inter-
vention, and (f) major results. The characteristics of the selected articles are shown in Table 2.
Evidence of the effectiveness of massage therapy was evaluated by considering the major results of
each article in light of the study design and methods of assessment. For instance, randomized
clinical trial with control group would be considered to yield stronger evidence of the efficacy of
the studied intervention than other research methods.
Given the diversity of massage techniques in clinical practice, evaluation of the effectiveness of
different massage techniques is impossible in the present review as some articles did not reveal the
massage protocols used.
Results
Samples
Among the seven qualifying studies, a total of 20 subjects (13 females and seven males, aged from
5 to 62 years) was involved. Sample size of individual studies ranged from one to eight subjects.
Table 1. Number of articles retrieved from different databases
Name of database No. of articles retrieved
Academic search premier (1980 to June 2010) 10
All EBM Reviews (1991 to June 2010) 1
CINAHL (1982 to June 2010) 17
EMBASE (1980 to June 2010) 22
ERIC (1966 to June 2010) 1
Ovid MEDLINE (1950 to June 2010) 10
PsycINFO (1806 to June 2010) 3
50 Journal of Intellectual Disabilities 15(1)
Table 2. Overview of the characteristics of the selected studies
Author(s) Sample Setting
Study
design Intervention pattern Outcome measure Major results
Croghan
(2009)
One girl with profound
intellectual and
multiple disabilities,
aged 13
Hospice setting Case study
report
Three sessions of hand
massage were offered.
Each session lasted for
10 minutes. No inter-
vention period stated
Pulseratewastaken
before, during and after
each hand massage
session to evaluate the
relaxation level
‘‘Affective
Communication
Assessment’’ tool was
used in behaviour
observation to evaluate
the response before
and after the
intervention
Pulse rate decreased
after hand massage
with an average of
7 beats lower than the
baseline measure. The
response of the subject
to hand massage was
relaxed and
comfortable
Dossetor
et al.
(1991)
One female with severe
intellectual disabilities
and Cornelia de Lange
syndrome, aged 14,
exhibited severe self-
injurious behaviour for
over 10 years
Community residential
home
Case study
report
Daily massage after bath.
Areas involved trunk
and back. Each session
lasted for 30 minutes.
As massage therapy
was integrated into
treatment regime, it
continued for at least
18 months
Frequency chart for
maladaptive
behaviours, e.g.
self-injurious
behaviours, aggression,
temper tantrum. Other
observations included
wound healing,
duration of wearing
splints and helmet
The frequency of self-
injurious behaviours
was progressively
decreased after
3 weeks of
intervention. Such
improvement had been
reviewed at 6 months
and maintained for a
minimum of 18 months
(continued)
51
Table 2 (continued)
Author(s) Sample Setting
Study
design Intervention pattern Outcome measure Major results
Hegarty
and
Gale
(1996)
One female with severe
intellectual disabilities,
aged 23, exhibited
challenging behaviours,
including aggression
Female ward of a
long-stay hospital
Case study
report
Weekly Swedish massage
was conducted in a
quiet room for
3 months. A total of
12 sessions was
provided with
45 minutes per session,
of which 10–25 minutes
was for massage
therapy, remainder for
counselling and rapport
building. Areas involved
head, temples, neck,
shoulders, hands and
lower arms
Pulse and respiration rates
were taken before,
during and after each
session to evaluate the
relaxation level. Written
information: 6 month
nursing records were
analysed prior to and
after the intervention
period. Verbal
information: any
behavioural change from
the staff perspective was
collected
Both pulse and respiration
rates decreased after
massage therapy
indicating relaxation.
Frequency of problem
behaviour decreased as
showninthenursing
record. From the staff
perspective, she
became more sociable
and initiated
interaction. She
tolerated outdoor
activities well after the
interventions
Lindsay
et al.
(1997)
Six females and two
males with profound
intellectual and
multiple disabilities,
aged 23 to 62
(M ¼38.6 years),
exhibited challenging
behaviours
Single ward of a hospital.
Both sensory room
and activity area were
occupied to administer
four treatments one by
one
Crossover
design of
a single
group
Four different treatments
administered in
sequence, including
behavioural relaxation
therapy, hand massage
or aromatherapy, active
therapy, and Snoezelen.
Each subject received
20 consecutive sessions
of each therapy, and
each session lasted for
20 minutes. Each week
delivered three ses-
sions; the intervention
period was around 27
weeks
Concentration was
evaluated through 5-
minute direct observa-
tions focusing on
engagement of mean-
ingful movement in an
occupational task. Four
sessions ofeach therapy
were videotaped to rate
the behaviours. Inter-
rater reliability r ¼0.96,
inter-rater agreement
¼93%.Afour-point
Likert scale was used to
indicate responsiveness
to relaxation/
enjoyment level. Inter-
rater reliability r ¼0.82,
inter-rater
agreement ¼86%
Concentration and
responsiveness to
relaxation/engagement
were not significantly
improved following
hand massage and
active therapy. Only
Snoezelen and
behavioural relaxation
therapy had a positive
effect on concentration
and relaxation
(continued)
52
Table 2 (continued)
Author(s) Sample Setting
Study
design Intervention pattern Outcome measure Major results
Lindsay
et al.
(2001)
Eight subjects with
profound intellectual
and multiple
disabilities, aged 23 to
62 (M ¼38.6 years),
same sample in Lindsay
et al. (1997)
Single ward of a hospital.
Both sensory room
and activity area were
occupied to administer
four treatments one by
one
Crossover
design of
a single
group
Same intervention pat-
tern as Lindsay et al.
(1997)
Communication was
evaluated in 12 sessions
(three from each
therapy) through a
five-point Likert scale
which consisted of five
positive variables
(friendly vocalization,
soft touch,
non-threatening gaze,
laughter and overall
positive
responsiveness) and
five negative variables
(screaming, self-injury,
aggression to others,
pulling away or leaving,
and overall negative
responsiveness).
Inter-rater agreement
over 95%
Snoezelen and
behavioural relaxation
therapy enhanced
positive
communication and
decreased negative
communication. Active
therapy and hand
massage showed little
or no effect on
communication
(continued)
53
Table 2 (continued)
Author(s) Sample Setting
Study
design Intervention pattern Outcome measure Major results
McEvoy
et al.
(1987)
Three females and two
males with severe
intellectual disabilities,
aged from 14 to 24 (M
¼19.6 years), all
exhibited challenging
behaviours over many
years
Special school Case study
report
Each week received two
or three times 45
minute massage
therapies for 16 weeks.
Areas involved head to
body. Acupressure was
applied. All subjects
were on behaviour
programs and
psychotropic
medications
throughout the study
period
Daily behaviour
observation checklist
and frequency of
challenging behaviour
were recorded by
school staff. Inter-rater
agreement ranged
from 85%to 100%.
Descriptive
observation of massage
therapy was recorded
by massage therapist
Behaviour change noted in
three subjects, who
showed increased
verbalizations for social
communication,
improved physical
condition due to
decreased self-injurious
behaviour and agitation,
increased functional
hand use in skill acquisi-
tion, increased on-task
behaviour and
decreased challenging
behaviours. However,
two autistic subjects
were withdrawn from
massage therapy
because of severe touch
aversion
Solomons
(2005)
Three boys and one girl
with autistic spectrum
disorders (ASD) and
severe intellectual dis-
abilities, aged 5 to 6 (M
¼5.5 years), exhibited
various types of chal-
lenging behaviours, e.g.
extreme anxiety, social
withdrawal, ritualistic
and obsessive
behaviours
Sensory and aromather-
apy rooms of a special
school
Case study
report
Each subject received
aromatherapy massage
twice a week for eight
months and each
session lasted for
maximum 10 minutes,
depending on child’s
request in
aromatherapy room.
Areas involved hands
and feet
Once a month direct
observation to look for
shared attention
behaviours for
consecutive 8 months
to compare the sub-
ject’s behaviours in the
sensory and
aromatherapy rooms.
Semi-structured
interviews were
conducted with school
staff and parents at the
beginning and the end of
the study
The frequency of shared
attention behaviours
was higher during
aromatherapy massage
than sensory activity.
The frequency of
challenging and
avoidance behaviours
decreased, while
responsiveness,
tolerance to physical
touch, and emotional
engagement increased
in massage sessions
54
Most subjects were classified as severe to profound intellectual disabilities and frequently displayed
various types of challenging behaviours, like self-injury, aggression and stereotypic self-stimulation.
Settings
Most studies were carried out in special schools, residential homes and hospital wards. Of these
settings, the multisensory room was commonly utilized for comparing the effect of massage
therapy (Lindsay et al., 1997; 2001; Solomons, 2005). One study was carried out in a hospice care
setting (Croghan, 2009); the author was a qualified massage therapist, teaching and demonstrating
massage skills to the care staff, who in turn offered hand massage to their clients. In fact, the
hospice care unit had previously provided massage therapy to their clients. After the com-
plementary therapist left, the service stopped. This was the only study and setting in which clients
had experienced massage therapy. Though individual experience with massage therapy was dis-
cussed in McEvoy et al.’s (1987) study, most recruited subjects of other studies seemed to be
receiving massage therapy for the first time.
Study designs
Five studies adopted a case study design and two used crossover designs, i.e. all subjects received a
sequence of different treatments and were evaluated individually (Lindsay et al., 1997; 2001). No
study employed an experimental design or randomized clinical trial to demonstrate the cause and
effect consequence.
The use of a case study approach gives the advantages of improving existing clinical practice
and exploring the relationship between the research question and the conceptual framework of the
study (Solomons, 2005). The in-depth study of individual responses can identify the common
grounds for further analysis in theory building.
Some studies utilized the same group of subjects as a control for the comparison of different
interventions (Lindsay, 1997; 2001; Solomons, 2005). The crossover design is especially suitable
for a study with a small sample size where the order effect has to be addressed. Lindsay and his
colleagues (1997; 2001) counteracted this problem by rotating the treatment sequence so that each
treatment sequence had two receiving subjects for comparison.
Intervention procedures
All studies involved delivering a series of massage therapy sessions to the subjects. Description of
massage intervention varied from simple depiction (Dossetor et al., 1991) to general account of
massage techniques (McEvoy et al., 1987). The duration of each massage session ranged from 10
minutes up to a maximum of 45 minutes. The shortest treatment period was 3 months (Hegarty and
Gale, 1996) and the longest up to 8 months (Solomons, 2005). In one study, massage therapy was
integrated into a regular treatment regime and maintained for at least 18 months to evaluate its
effectiveness (Dossetor et al., 1991). The frequency of massage therapy ranged from daily to
weekly. One study did not mention the treatment period (Croghan, 2009). Hands were the most
frequent massage location (Croghan, 2009; Hegarty and Gale, 1996; Lindsay et al., 1997; 2001;
Solomons, 2005). Most authors seldom explain explicitly the rationale of the treatment schedule.
Such rationale is important because effective massage needs to consider duration, frequency,
location, intensity of pressure, type of technique, and speed of movement (Fraser and Ross Kerr,
1993). Regarding the massage techniques, only the study conducted by McEvoy and his colleagues
(1987) supplied a relatively detailed explanation, such as the use of effleurage, petrissage and
Chan and Tse 55
acupressure. In two studies, authors described the hand movement on the massage location
(Hegarty and Gale, 1996; Lindsay et al., 1997). As a result, it is difficult to evaluate the efficacy of
the massage techniques used in the studies. Other than massage therapy, multisensory activity,
behavioural relaxation therapy, active therapy (Lindsay et al., 1997; 2001), counselling (Hegarty
and Gale, 1996) and aromatherapy (Solomons, 2005) were concurrently employed as interventions
in the studies.
Outcome measures
Narrative description and qualitative data were abundant in the case study reports. Quantitative
data mainly came from studies presented by Lindsay and his colleagues (1997; 2001). Pre- and
post-treatment assessments were common for evaluating the differences in terms of behavioural
changes and physiological data on relaxation level. Measurements included observations, physio-
logical monitoring and staff interviews. All studies used behaviour observations as part of their
outcome measures. Physiological monitoring of heart rate and respiration rate was employed to
evaluate the relaxation level (Croghan, 2009; Hegarty and Gale, 1996), the lower the better in the
post-treatment reading. Formal or informal interviews with caregivers (Hegarty and Gale, 1996;
Solomons, 2005) were also held to validate the collected data from behaviour observations and
physiological readings.
Some researchers used a standardized measuring tool, i.e. affective/early communication
assessment (Croghan, 2009; Solomons, 2005), to evaluate the behavioural changes, while others
devised their own tools, such as behaviour frequency charts and Likert scales. McEvoy and his
colleagues (1987) revealed that the inter-rater agreement of direct observations ranged from 85 to
100 percent. Nursing record was also used to examine the behavioural changes in the pre- and post-
intervention stages (Hegarty and Gale, 1996). Only Lindsay and his colleagues (1997; 2001)
adopted a quantitative method to analyse the observation data with inter-rater reliability ranging
from 0.82 to 0.95. The two-way analysis of variance (ANOVA) statistical method was applied
to the obtained scores based on observation data.
The typical measure outcomes included relaxation level, frequency of positive or challenging
behaviours, attention or concentration span, and communication intention.
Major results
Effectiveness in inducing relaxation. A study confirmed that reductions in physiological signs of heart
rate and respiration rate were associated with biological relaxation state (Labyak and Metzger,
1997). Studies using physiological data to indicate relaxation effect yielded positive results
(Croghan, 2009; Hegarty and Gale, 1996), both pulse and respiration rates decreased compared
with baseline measures after subjects received interventions. However, the statistical significance
of the results is unknown as the researchers did not report levels of significance in the articles.
Based on behaviour observations, the responses of most subjects reached a certain level of
relaxation and enjoyment during and after the massage therapy. For instance, the subjects took the
initiative to ask for massage (Dossetor et al., 1991), showed increased social awareness (McEvoy
et al., 1987), and appeared relaxed and calm (Croghan, 2009). On the whole, the relaxation effect
was prevalent in most studies, though it might not be explicitly specified in the major findings; the
decrease in challenging behaviours and increase in positive behaviours could be an indicator of
relaxation, except in the studies of Lindsay and his colleagues (1997; 2001). The results of hand
56 Journal of Intellectual Disabilities 15(1)
massage showed a limited effect on relaxation when compared with Snoezelen and behavioural
relaxation training (Lindsay et al., 1997; 2001).
Effectiveness on challenging behaviours. Subjects with severe challenging behaviours showed a
decreased frequency of such behaviours after the interventions (Dossetor et al., 1991; McEvoy
et al., 1987; Hegarty and Gale, 1996). The result was very obvious in Dossetor et al.’s (1991) study:
the frequency of self-injurious behaviours was progressively decreased after a 3-week intervention.
The improvement was maintained for over 18 months. Similar positive results were also seen by
Hegarty and Gale (1996). The incidence of challenging behaviours dropped to 0–4 from a maxi-
mum of 15 incidents after a 3-month intervention (Hegarty and Gale, 1996). Although two autistic
subjects withdrew from the intervention because of severe touch aversion, the results of McEvoy
et al.’s (1987) study showed a decrease in self-injurious behaviours. This triggered a maintenance
programme to make progress with problem behaviours.
Effectiveness on positive behaviours. As defined earlier, positive behaviours involve communication
intention and concentration span (Chan et al., 2010). Two studies revealed that hand massage had
no effect on concentration and attempts at communication (Lindsay et al., 1997; 2001). Only
Snoezelen and behavioural relaxation training had positive effects on concentration and com-
munication capacity. Subjects from McEvoy et al.’s (1987) study recorded an increased number of
verbalizations and engagement behaviours following massage therapy. One subject became more
sociable and initiated communication with care staff, and was able to tolerate outdoor activities
that she had never tried before (Hegarty and Gale, 1996). In a study for children with autistic
spectrum disorders (ASD), during interactive aromatherapy massage sessions, the frequency of
avoidance behaviours decreased, and at the same time, social engagement increased (Solomons,
2005). The duration of interactive massage depended on the demand of the subjects, with a
maximum of 10 minutes. Based on narrative data from caregivers and behaviour observations,
positive behaviours generally increased following massage therapy.
Other findings. Despite the fact that children with ASD were characterized by short attention span
and resistance to relationship building, Solomons (2005) contended that the nature of the activity
was far more important than the diagnosis for the manipulation of engagement level. For instance,
aromatherapy massage provided non-threatening communication and emotional bonding between
the therapist and the child subjects. Subsequently, the duration of physical engagement and atten-
tion increased. As a result, the shared attention behaviours, which emphasize eye contact and fin-
ger pointing for social exchange, were found to be more frequently enhanced during aromatherapy
massage than during sensory activity. The study also showed an increase in subjects’ tolerance to
physical touch through the use of massage therapy. With increasing on-task behaviour during day-
time, the sleeping problem was relieved (McEvoy et al., 1987; Solomons, 2005). Such behavioural
change could enhance subjects’ learning at school and prevent them engaging in stereotypic self-
stimulating behaviours (Solomons, 2005).
Thus, massage therapy could promote positive emotional and physiological changes; it was
exceptionally important for staff to provide such comfort measures for clients at the end of their
lives. From the staff perspective, hand massage could facilitate emotional connections and
effective communication between caregivers and the clients. More importantly, the staff found that
they had done something meaningful for the terminally ill (Croghan, 2009).
Chan and Tse 57
Discussion
Issues in intervention pattern
As discussed earlier, few studies have explained the reason behind the treatment schedule. The
level of tactile arousal indeed is determined by six factors: duration, location of touch, rate of
massage movement, type of massage technique, intensity of pressure, and movement direction
(Fraser and Ross Kerr, 1993). In the selected studies, most addressed the intervention duration and
areas massaged, but few studies discussed the other determinants. Hence, it is difficult to evaluate
the effectiveness of the intervention pattern.
Because massage therapy is usually applied to a localized body region, short duration is suf-
ficient to bring about positive physiological parameters (Moraska et al., 2008). A study examined
the duration of massage to trigger physiological changes; a 3-minute effleurage backrub was able
to lower blood pressure, heart rate and respiration rate. Such physiological effects reached their
maximum decrease at 10 minutes of massage, and then regressed. Since the length of massage
therapy was independent of subjective experience of pleasure and relaxation, the ideal duration of
massage therapy was still uncertain (Labyak and Metzger, 1997). However, 10- to 30-minute
intervention duration appears appropriate with reference to previous literature, possibly because of
short attention span and inadequate social exchange among persons with intellectual disabilities.
Massage therapy with a longer than 30-minute duration was not common; only one study offered a
45-minute therapy (McEvoy et al., 1987). For persons with ASD, 10-minute massage therapy may
be appropriate for them to experience physical proximity, touch and social engagement with the
therapist (Solomons, 2005). A longer duration than 10 minutes may result in a high attrition rate
(McEvoy et al., 1987).
The technique of backrub massage promoted rest and relaxation as evidenced by a reduction of
physiological parameters (Fraser and Ross Kerr, 1993; Labyak and Metzger, 1997); however, only
two selected studies involved back massage (Dossetor et al., 1991; McEvoy et al., 1987), while the
rest of the studies comprised hand massage only. The widespread occurrence of chest infection,
kyphosis and scoliosis, and spine deformity among persons with severe to profound intellectual
disabilities may be the reason for choosing the hands rather than the back as the site for massage.
A study demonstrated that moderate pressure massage was best for relaxation and stress
reduction, as evidenced by changes in electroencephalogram (EEG) and electrocardiogram (ECG)
that were consistent with the results obtained through questionnaires concerning anxiety, stress and
relaxation (Diego et al., 2005). Moreover, massage oil was found to reduce the skin friction between
therapist and client, making the massage movement more rhythmic and smoother than massage
without oil. Subsequently, the therapeutic effect of massage therapy was enhanced, as supported by
reduced salivary cortisol level and increased parasympathetic activity (Field et al., 1996).
Despite the insignificant results of massage therapy on relaxation level (Dunn et al., 1995;
Fraser and Ross Kerr, 1993; Lindsay et al., 1997; 2001), the use of massage therapy is still rec-
ommended because it communicates a caring attitude to persons with intellectual disabilities.
Based on the stress–attachment model (Janssen et al., 2002), the frequency of challenging beha-
viour is reduced if subjects feel relaxed.
Mechanisms of massage therapy
Possible mechanisms to explain the benefits of massage therapy include four dimensions, i.e.
biomechanical, physiological, neurological and psychological aspects (Weerapong et al., 2005).
58 Journal of Intellectual Disabilities 15(1)
Biomechanical effects are produced by mechanical pressure on body tissues, where muscle
compliance and range of joint motion are increased; hence muscle and joint stiffness, and tissue
adhesion are decreased.
Physiological benefits are derived from the relaxation effect. It is hypothesized that relaxation is
a state of parasympathetic activation by the vagal nerve, suppressing the activity of the sympathetic
nervous system. This assumption is based on the outcomes of physiological measures, i.e. heart
rate, blood pressure and respiration rate (Fraser and Ross Kerr, 1993; Moraska et al., 2008). All
these physiological parameters gradually return to baseline level after massage therapy, resulting
in a state of relaxation (Moraska et al., 2008). However, it is not easy to identify the physiological
differences in pre- and post-intervention comparison if the physiological or psychological arousals
are low or sustained at baseline level, where blood pressure, heart rate, respiration rate and cortisol
level are relatively stable. It would be difficult to generate significant results in these cases.
Neurological effects result from reflex stimulation of nerve impulse or neurotransmitter
transmission. For instance, EEG patterns change from active and busy positions to resting or
sleeping states, and the secretion of relaxation hormones, e.g. endorphin, increase while stress
hormones, like cortisol, decrease (Weerapong et al., 2005). Lastly, psychological effects
involve relationship building between body and mind, promoting relaxation and decreasing
anxiety.
At present, mechanisms that can fully explain the relaxation and therapeutic effects of massage
therapy remain unknown. More research is needed to identify the chains of action sequence when
massage therapy is administered.
Research gaps and further studies
The research designs of the case study report and the crossover study have similar methodological
limitations, such as small sample size, the absence of a control group, lack of baseline and follow-
up data, and lack of reliable indications of the measurements. All these limitations lead to incon-
clusive results. The effectiveness of hand massage on concentration, responsiveness to relaxation
and communication potential would be different if a control group and follow-up assessment were
offered.
Despite behaviour observations being employed in outcome measures, most case study reports
did not report the inter-rater reliability or agreement, except for McEvoy et al. (1987). However,
this unpublished manuscript harbours lots of confounding variables; for instance, all subjects were
on behaviour programmes and psychotropic medications. During the intervention period, some
behaviour programmes and medication regimes were revised. It is difficult therefore to decide
which variable(s) made a difference to outcomes.
Apart from physiological data, qualitative data from behaviour observations and interviews
were dominant in the literature. This form of data could be a bias for the effectiveness of the
massage therapy, e.g. Hawthorn effect from parents, care staff, and massage therapist (McEvoy
et al., 1987). EEG, electromyography (EMG) and salivary cortisol are objective measures for
evaluating the biological relaxation state or stress level. These measures, which are non-invasive
and painless to subjects, can be used in future studies to evaluate the effectiveness of massage
therapy.
Massage has long been used as a nursing modality. With the advancement of medical
technologies and complex nursing procedures, massage therapy is seldom adopted in acute
specialties (Labyak and Metzger, 1997). Even when massage is used, most studies provide largely
Chan and Tse 59
qualitative data and subjective reports. Without rigorous scientific method, the substantive effects
of massage therapy are difficult to evaluate. Hence, a randomized clinical trial or experimental
study design is required in future studies to evaluate the efficacy of massage therapy, using
objective measuring tools, including neurophysiological data and validated behaviour observation
scales. The intervention pattern needs to pay particular attention to the six determinants of tactile
arousal. The longer-term efficacy of massage therapy can then be assessed after the intervention
schedule has been completed. Moreover, documentation of massage protocols would be beneficial
to replicate and evaluate the effectiveness of different massage techniques.
Clinically, functional or task-related touch shares major physical contact between caregivers
and persons with intellectual disabilities. The emotional exchange of functional touch is scarce.
In fact, the frequency of touch decreases when there is busy routine or staff shortages (Gale and
Hegarty, 2000). It seems that a structured massage programme with staff participation is a promis-
ing combination for generating positive therapeutic effects on relaxation and positive behavioural
changes.
The existing database of massage therapy for persons with intellectual disabilities is very limited.
Only seven studies met the pre-set criteria in the review. The effectiveness of massage therapy is
inconclusive because of the paucity of studies and the small number of participants (N¼20). Com-
pared with other target populations, massage therapy for persons with intellectual disabilities is not a
popular topic among researchers. The reasons may be the difficulty in obtaining valid consent,
administering self-report questionnaires, and receiving verbal feedback from subjects.
With increasing longevity and vulnerability to challenging behaviours, experimental designs
or randomized clinical trials are feasible to evaluate the short-term and longer-term efficacy of
massage therapy among persons with intellectual disabilities to bridge the research gap in this
area. Other than effective interventions, non-judgemental acceptance and reciprocal communi-
cation skills are also important to handle children with emotional and behavioural problems
(Dossetor et al., 1991).
Conclusion
It is believed that massage therapy increases client–staff interaction through physical touch and
attention. Such interaction may bring the clients with intellectual disabilities to an increased
awareness of the immediate environment and a decreased frequency of self-stimulating behaviour.
Effective massage is determined by six factors: duration, location of touch, massage movement,
massage techniques, pressure type, and movement direction (Fraser and Ross Kerr, 1993). The
mechanisms of explaining the effects of massage are still under experiment, although the activa-
tion of the parasympathetic nervous system is understandable as it can be indicated by physiolo-
gical parameters. Yet, the philosophy of using massage therapy should be upheld for promoting the
psychological wellbeing of the clients.
Massage claims to be effective in inducing relaxation and reducing challenging behaviours in
persons with intellectual disabilities. However, there is a research gap in affirming such relaxation
effects in clinical practice. Moreover, no sustainable long-term effects from massage therapy are
reported in the literature. A number of methodological questions prevent drawing solid conclusions
about the beneficial effects of massage therapy because most reported positive results were largely
based on case studies and anecdotal clinical experience. More rigorous research designs, e.g.
experimental or randomized clinical trials, are indicated to evaluate accurately the therapeutic
effects of massage therapy.
60 Journal of Intellectual Disabilities 15(1)
References
Chan, S. W. C., Thompson, D. R., Chau, J. P. C., Tam, W. W. S., Chiu, I. W. S. & Lo, S. H. S. (2010) ‘The
Effects of Multisensory Therapy on Behaviour of Adult Clients with Developmental Disabilities:
A Systematic Review’, International Journal of Nursing Studies, 47: 108–22.
Cooray, S. E. & Bakala, A. (2005) ‘Anxiety Disorders in People with Learning Disabilities’, Advances in
Psychiatric Treatment, 11: 355–61.
Croghan, P. (2009) ‘The Therapeutic Effects of Hand Massage’, Learning Disability Practice,12(5):
29–32.
Deakin, M. (1995) ‘Using Relaxation Techniques to Manage Disruptive Behaviour’, Nursing Times, 91 (17):
40–1.
Diego, M., Field, T., Sanders, C. & Hernandez-Reif, M. (2005) ‘Moderate Pressure Massage Is Best for
Relaxation and Reducing Stress’, Massage Therapy Review, 11 (4): 181–2.
Dossetor, D. R., Couryer, S. & Nicol, A. R. (1991) ‘Massage for Very Severe Self-Injurious Behaviour
in a Girl with Cornelia de Lange Syndrome’, Developmental Medicine and Child Neurology,33:
636–44.
Dunn, C., Sleep, J. & Collett, D. (1995) ‘Sensing an Improvement: An Experimental Study to Evaluate the
Use of Aromatherapy, Massage and Periods of Rest in an Intensive Care Unit’, 21: 34–40.
Durand, V. M. & Carr, E. G. (1987) ‘Social Influences on ‘‘Self-Stimulatory’’ Behaviour: Analysis and
Treatment Application’, Journal of Applied Behavior Analysis, 20 (2): 119–32.
Emerson, E. (1995) Challenging Behaviour: Analysis and Intervention in People with Learning Disabilities.
New York: Cambridge University Press.
Field, T., Schanberg, S., Davalos, M. & Malphurs, J. (1996) ‘Massage with Oil Has More Positive Effects on
Normal Infants’, Pre- and Perinatal Psychology Journal, 11 (2): 75–80.
Forster, S. & Iacono, T. (2008) ‘Disability Support Workers’ Experience of Interaction with a Person with
Profound Intellectual Disability’, Journal of Intellectual & Developmental Disability, 33 (2): 137–47.
Fraser, J. & Ross Kerr, J. (1993) ‘Psychophysiological Effects of Back Massage on Elderly Institutionalized
Patients’, Journal of Advanced Nursing, 18: 238–45.
Gale, E. & Hegarty, R. (2000) ‘The Use of Touch in Caring for People with Learning Disability’, The British
Journal of Developmental Disabilities, 46 (91): 97–108.
Harris, B. & Lewis, R. (1994) ‘Physiological Effects of Massage’, International Journal of Alternative and
Complementary Medicine, 12 (2): 16–17.
Hegarty, J. R. & Gale, E. (1996) ‘Touch as a Therapeutic Medium for People with Challenging Behaviours’,
British Journal of Learning Disabilities, 24: 26–32.
Janssen, C. G. C., Schuengel, C. & Stolk, J. (2002) ‘Understanding Challenging Behaviour in People with
Severe and Profound Intellectual Disability: A Stress–Attachment Model’, Journal of Intellectual Dis-
ability Research, 46: 445–53.
Labyak, S. E. & Metzger, B.L. (1997) ‘The Effects of Effleurage Backrub on the Physiological Components
of Relaxation: A Meta-Analysis’, Nursing Research, 46 (1): 59–62.
Lambrechts, G. & Maes, B. (2009) ‘Analysis of Staff Reports on the Frequency of Challenging Behaviour in
People with Severe or Profound Intellectual Disabilities’, Research in Developmental Disabilities, 30:
863–72.
Lindsay, W. R. & Baty, F. (1986) ‘Behavioural Relaxation Training: Explorations with Adults Who Are
Mentally Handicapped’, Mental Handicap, 14: 160–2.
Lindsay, W. R., Pitcaithly, D., Geelen, N., Buntin, L., Broxholme, S. & Ashby, M. A. (1997) ‘A Comparison
of the Effects of Four Therapy Procedures on Concentration and Responsiveness in People with Profound
Learning Disabilities’, Journal of Intellectual Disability Research, 41: 201–7.
Lindsay, W. R., Black, E., Broxholme, S., Pitcaithly, D. & Hornsby, N. (2001) ‘Effects of Four Therapy
Procedures on Communication in People with Profound Intellectual Disabilities’, Journal of Applied
Research in Intellectual Disabilities, 14: 110–19.
Chan and Tse 61
McEvoy, C., Perrault, A. & Graetz, J. (1987) ‘The Use of Massage Therapy in the Treatment of Self-Injurious
Behavior’. Unpublished manuscript, Wayne County Intermediate School District.
Moraska, A., Pollini, R. A., Boulanger, K., Brooks, M. Z. & Teitlebaum, L. (2008) ‘Physiological Adjust-
ments to Stress Measures following Massage Therapy: A Review of the Literature’. Available from http://
ecam.oxfordjournals.org.
Moss, S., Emerson, E. & Kiernan, C. (2000) ‘Psychiatric Symptoms in Adults with Learning Disability and
Challenging Behaviour’, British Journal of Psychiatry, 17: 452–6.
Moss, L., Smith, M., Wharton, S. & Hames, A. (2007) ‘Abdominal Massage for the Treatment of Idiopathic
Constipation in Children with Profound Learning Disabilities: A Single Case Study Design’, British
Journal of Learning Disabilities, 36: 102–8.
Nakken, H. & Vlaskamp, C. (2007) ‘A Need for a Taxonomy for Profound Intellectual and Multiple Dis-
abilities’, Journal of Policy and Practice in Intellectual Disabilities, 4 (2): 83–7.
Rudrud, E. & Halaszyn, J. (1981) ‘Reduction of Bruxism by Contingent Massage’, Special Care in Dentistry,
1 (3): 122–4.
Schilling, D. J. & Poppen, R. (1983) ‘Behavioral Relaxation Training and Assessment’, Journal of Behavior
Therapy & Experimental Psychiatry, 14 (2): 99–107.
Shapiro, M., Parush, S., Green, M. & Roth, D. (1997) ‘The Efficacy of the ‘‘Snoezelen’’ in the Management of
Children with Mental Retardation Who Exhibit Maladaptive Behaviours’, British Journal of Develop-
mental Disability, 43 (Part 2): 140–55.
Singh, N. N., Lancioni, G. E., Winton, A. S. W., Molina, E. J., Sage, M., Brown, S. & Groeneweg, J. (2004)
‘Effects of Snoezelen Room, Activities of Daily Living Skills Training, and Vocational Skills Training on
Aggression and Self-Injury by Adults with Mental Retardation and Mental Illness’, Research in Develop-
mental Disabilities, 25: 285–93.
Smith, S. A., Press, B., Koenig, K. P. & Kinnealey, M. (2005) ‘Effects of Sensory Integration Intervention on
Self-Stimulating and Self-Injurious Behaviors’, The American Journal of Occupational Therapy, 59 (4):
418–25.
Solomons, S. (2005) ‘Using Aromatherapy Massage to Increase Shared Attention Behaviours in Children
with Autistic Spectrum Disorders and Severe Learning Difficulties’, British Journal of Special Education,
32 (3): 127–37.
Weerapong, P., Hume, P. A. & Kolt, G. S. (2005) ‘The Mechanisms of Massage and Effects on Performance,
Muscle Recovery and Injury Prevention’, Sports Medicine, 35 (3): 235–56.
62 Journal of Intellectual Disabilities 15(1)