Sie sind auf Seite 1von 10

Journal of Applied Research in Intellectual Disabilities 2008, 21, 126–135

The Effect of an Attachment-Based Behaviour


Therapy for Children with Visual and Severe
Intellectual Disabilities
P. S. Sterkenburg, C. G. C. Janssen and C. Schuengel
Department of Clinical Child and Family Studies, Faculty of Psychology and Education, Vrije Universiteit Amsterdam,
Amsterdam, The Netherlands

Accepted for publication 4 April 2007

Background A combination of an attachment-based ther- Results Across clients, challenging behaviour in the resi-
apy and behaviour modification was investigated for dential home decreased during the attachment therapy
children with persistent challenging behaviour. phase. The behaviour modification sessions conducted
Method Six clients with visual and severe intellectual by the experimental therapist resulted in significantly
disabilities, severe challenging behaviour and with a more adaptive target behaviour than the sessions with
background of pathogenic care were treated. Challen- the control therapist.
ging behaviour was recorded continuously in the resi- Conclusion For these clients with a background of
dential home and during therapy sessions. Alternating attachment problems, attachment-based behaviour
treatments were given by two therapists. In phase 1, the modification treatment may have important advantages
experimental therapist attempted to build an attachment over standard behaviour modification.
relationship in sessions alternating with sessions in
which a control therapist provided positive attention Keywords: attachment-based psychotherapy, behaviour
only. In phase 2, both therapists applied the same beha- modification, challenging behaviour, evidence-based
viour modification protocol. intervention, intellectual disability, visual disability

scaffolding for the development of children’s social


Introduction
behaviour systems, but may also have beneficial effects
Some children with serious intellectual and visual dis- on affect regulation in these children. Attachment theory
abilities appear insensitive to social contingencies to their and recent theories of affect regulation point to the
behaviour, impeding the therapeutic use of social rein- important homeostatic affect-regulating function of
forcers to treat challenging behaviour and shape adap- attachment relationships (Bowlby 1984; Bradley 2000).
tive behaviour (e.g. O’Reilly et al. 2003). Insensitivity to Children without attachment relationships are more
social rewards may result from disrupted or atypical likely to develop maladaptive ways of regulating stress
attachment relationships, for which children with intel- in their daily lives (Janssen et al. 2002; Schuengel & Jans-
lectual and multiple disabilities are at risk (Schuengel & sen 2006). Using medication, it has been shown that
Janssen 2006). Intensive social contact found in attach- decreasing the level of arousal is associated with a
ment relationships is necessary for the development of decrease in challenging behaviour in people with intel-
social regulatory systems (Porges 2004), allowing social lectual disabilities (Verhoeven & Tuinier 1996). Develop-
contact to become rewarding. This study sought to test ing a therapeutic attachment relationship with children
the hypothesis that addressing the attachment needs of who are currently not involved in stable and ⁄ or organ-
children with severe intellectual and visual disabilities ized attachment relationships may in itself have benefi-
would subsequently increase the effectiveness of social cial effects on challenging behaviour through its effect
contingencies in the treatment of challenging behaviour. on affect regulation. Indeed, this may even precede its
Enabling children to build attachment relationships effects on sensitivity to social reinforcements (Bradley
with caregivers or therapists may not only provide a 2000; Fisher et al. 2000).

 2007 The Authors. Journal compilation  2007 Blackwell Publishing Ltd 10.1111/j.1468-3148.2007.00374.x
Journal of Applied Research in Intellectual Disabilities 127

Many studies of the effectiveness of treatment for opment of an attachment relationship (Bowlby 1984).
challenging behaviours in clients with intellectual dis- Only later, in the behaviour modification phase, is the
abilities showed strong evidence that behavioural inter- attention focused on systematically placing social contin-
ventions can be effective and efficient (e.g. Sturney 2005; gencies on adaptive replacement behaviour. In this
Prout & Nowak-Drabik 2003; Didden et al. 1997; Scotti respect, the integrative treatment is also different from
et al. 1991). But for some clients, as mentioned above, approaches such as gentle teaching, which have been
behavioural interventions have shown little effect; for proposed as alternatives to behaviour modification
example, in clients who show no sensitivity to reinforce- (Jones & McCaughey 1992); this integrative treatment
ment (e.g. O’Reilly et al. 2003), and in clients with histor- seeks to complement behaviour modification. Audiovis-
ies suggesting that they may have had little opportunity ual and written materials describing the intervention in
to develop selective attachment, because of pathogenic detail are available from the authors.
care in the past (early disruptions in caregiving relation- In this study, the effect of attachment-based behaviour
ships, abuse, neglect). modification treatment on children with visual and
An integrative psychotherapeutic treatment, the severe intellectual disabilities was tested. First of all,
Attachment-based Behaviour Modification Therapy, was changes in challenging behaviour in the residential
developed for children for whom other treatments (e.g. home during the period in which the children were
standard behaviour modification) had failed. These chil- under therapy were examined. The main aim of the
dren exhibited severe challenging behaviour and had study was, however, to test whether the attachment
visual and severe intellectual disabilities. This intensive therapist was more effective than the non-attachment or
integrative treatment is partly based on Došen’s (1984, control therapist in modifying challenging into adaptive
1990, 2001, 2005) Developmental-Dynamic Relationship behaviour, with both therapists using the same beha-
Therapy. The attachment-based approach is combined viour modification protocol.
with a behavioural intervention, in order to replace
remaining maladaptive behaviour with more adaptive
Methods
behaviour. Furthermore, it is an integrative, individual
treatment with a therapist who works directly with cli-
Design
ents to effect intra-psychic changes during therapy. It
differs from Došen’s Developmental-Dynamic Relation- A series of single-case studies with alternating therapy
ship Therapy in that the latter is mediated by caregivers conditions across clients was conducted. Both treatments
(Došen 1984, 1990, 2001, 2005). For the clients in this (attachment therapy and the control treatment) were
study, however, the social isolation and level of challen- conducted concurrently and consisted of two phases,
ging behaviour were deemed to require a one-to-one attachment-based therapy (phase 1) and behaviour
approach from a skilled psychotherapist before caregiv- modification (phase 2), followed by a generalization
ers could receive guidance on building positive caregiv- phase (phase 3). The attachment therapist conducted the
ing relationships themselves and applying appropriate phase 1 attachment therapy sessions (the intervention is
social rewards for adaptive behaviour. described later in this article) and the control therapist
It is important to note that the attachment-based treat- merely positively engaged with that same client during
ment must be distinguished from ‘re-birthing’, ‘holding’ the control sessions. During the behaviour modification
or ‘attachment’ therapies which, as Zilberstein (2006) (phase 2), both therapists used the same behaviour
mentioned, focus more broadly on a forced simulation modification protocol. The alternating treatment design
of the birth process, and address aggression and non- meant that therapist 1 was the attachment therapist and
compliance instead of attachment. Attachment-based therapist 2 the control therapist for three clients, the
treatment as described in this paper is based on sensi- situation being reversed for the other three clients. In
tive and responsive interactions, on caring and empathy, this study, it was not possible to add a baseline in a
instead of coercion. The therapist positively encourages non-treatment phase preceding behaviour modification,
contact by letting the child become acquainted with him as is usually done in single-case studies. The reason was
or her as a figure that is supportive, predictable and that the attachment-based intervention preceded beha-
comfortable to be with. In addition, the integrative treat- viour modification. A baseline before the attachment-
ment also does not mainly focus on developing new based intervention would not accurately represent the
behaviour (O’Reilly et al. 1999). The attachment-based frequency of challenging behaviour approximately
behaviour therapy first follows the stages in the devel- 5 months later at the start of behaviour modification,

 2007 The Authors. Journal compilation  2007 Blackwell Publishing Ltd, 21, 126–135
128 Journal of Applied Research in Intellectual Disabilities

because of the expected positive effect of the attach- needs; frequent changes of regular caregivers), as
ment-based treatment. The manifestation of challenging assessed by an independent psychiatrist. Examinations
behaviour might thus be different at the start of the by physicians ruled out medical aspects that might be
attachment-based intervention from what it was at the associated with self-injury. There were no contraindica-
start of behaviour modification. Furthermore, adding a tions for the treatment protocol, such as aversion to
no-treatment baseline after the attachment-based inter- physical contact. The independent psychiatrist had
vention might disrupt the child’s fragile security of extensive experience with assessment and treatment of
attachment to the therapist and was therefore not con- clients with a visual and intellectual disability and was
sidered as an option. also able to identify clients with autistic spectrum disor-
ders; these clients were then excluded from this study.
Psychologists working with children with severe intel-
Participants
lectual and visual disabilities were asked to present cli-
All six clients had a severe intellectual disability and ents matching these criteria. Medical ethical approval
were blind or had a visual impairment (in accordance was obtained from the Vrije Universiteit Medical Centre
with the WHO criteria). Five clients lived in a residential Medical-Ethics Review Board. This board is licensed to
home, and one in a foster home. In the Netherlands chil- approve research by the Central Committee on Research
dren are placed in residential homes only in exceptional Involving Human Subjects (CCMO), which monitors
situations, for instance if the child displays severe chal- compliance with Dutch legislation on medical research.
lenging behaviour and ⁄ or in situations of dysfunctional Parents gave their informed consent in writing. Three
family care. Characteristics of the participants are shown clients participated during 2002–2003 and three clients
in Table 1. during 2003–2004 with a mean of 21.8 weeks for phase
All the clients persistently engaged in severe self- 1, a mean of 7.6 weeks for phase 2 and a mean of
injurious behaviour, aggression and disruption. Their 8.2 weeks for phase 3. The intervention took place in the
behaviour was noted as severe and persistent with a residential environment of the client.
high score on the ‘Severe Challenging Behaviour Con-
sensus Protocol – National Institute for Health Care
Intervention
Management’(Consensusprotocol Ernstig Probleemge-
drag – Nationaal Ziekenhuisinstituut or CEP), a com- The integrative treatment (the Attachment-based Beha-
monly used Dutch protocol for measuring challenging viour Modification Treatment) consists of three phases.
behaviour which is reliable and valid (Kramer 2001). Phase 1 is aimed at creating the conditions under which
Table 2, column 2 describes the challenging behaviour attachment may develop. First (phase 1.1), the therapist
noted at the start of the intervention for each client. provides sensitive and encouraging responses with the
The history of the clients included early pathogenic aim of making contact with the client. Contact starts
care, as defined by the criteria for attachment disorder with vocal initiatives, as clients have to get to know
(DSM-IV: persistent neglect of basic emotional needs for their therapist by voice first, because of their visual dis-
comfort and affection; persistent neglect of physical ability. Touch can be added if the client feels comfort-
able when touched. The therapist reacts sensitively to
the positive and to the negative reactions of the client
Table 1 Characteristics of participants
by verbally or non-verbally acknowledging the signal
Age in Motor Medical and adapting the interaction or the situation to suit the
Therapist Client Gender years development information client. The therapist positively engages with the client
by singing, telling stories and talking to the client. The
1 A M 17 Normal Blind, Down next step is taken when the client shows signs of recog-
syndrome nition of the therapist, when the therapist can predict
1 C M 17 Normal Blind the client’s behaviour, when the client experiences posit-
1 E F 17 Delayed Blind ive and enjoyable contact which can easily be restored
2 B F 16 Cerebral palsy, Visual
after contact is broken, and when the client initiates
delayed impairment
proximity-seeking in relation to the therapist.
2 D F 14 Normal Visual
impairment
During the next, ‘symbiotic’ phase (1.2), communica-
2 F M 10 Normal Blind tion in close contact is central. The therapist stimulates
the client to take the lead while the therapist attempts to

 2007 The Authors. Journal compilation  2007 Blackwell Publishing Ltd, 21, 126–135
Journal of Applied Research in Intellectual Disabilities 129

Table 2 All challenging behaviour noted at the start of the intervention, and the target challenging behaviour and appropriate
behaviour which were the focus of treatment during phase 2 (behaviour modification)

All challenging behaviour noted Target challenging behaviour focused on Functional alternative appropriate behaviour
Client at the start of the intervention during phase 2 (behaviour modification) focused on during phase 2 (behaviour modification)

Therapist 1
A Pinching ⁄ scratching Scratching (himself and others) Using his hand to ask for music
Hitting ⁄ kicking Turning away or putting his hand under his legs
Spitting Using his hand to ask for sweets
Screaming
Other challenging behaviour
C Pinching ⁄ scratching Pinching ⁄ scratching Getting up from his chair without showing
Hitting ⁄ kicking Hitting ⁄ kicking challenging behaviour
Biting Biting Sighing
Hair pulling Screaming Using his hands to play with toys
Screaming Other challenging behaviour
Other challenging behaviour
E Pinching ⁄ scratching Pinching ⁄ scratching Using her hand to ask for music
Hitting ⁄ kicking Hitting ⁄ kicking Turning away or pushing the therapist away
Biting Screaming Walking to another chair without challenging
Screaming behaviour
Other challenging behaviour
Therapist 2
B Pinching ⁄ scratching Pinching ⁄ scratching Using her hand to ask for fruit
Hitting ⁄ kicking Screaming loudly Using her hands to play with toys
Biting Making low noises
Screaming loudly
Other challenging behaviour
D Hitting ⁄ kicking Hitting ⁄ kicking Using her hands to hold a handkerchief
Hair pulling Stroking her ears or turning her head towards
Screaming shoulder
Other challenging behaviour Using her hands to play with toys
F Pinching ⁄ scratching Hitting ⁄ kicking Using his voice to say ‘no’
Hitting ⁄ kicking Screaming Putting the toys on the table when finished
Screaming Using his hands to play with toys
Crying
Other challenging behaviour

achieve synchronicity in the interaction by anticipating verbal comfort when exploration results in anger or
the client’s actions as closely as possible. For example, anxiety. Seeking contact, when unpleasant emotions are
when the client and therapist rock sideways to the experienced, is thus rewarded by giving positive atten-
rhythm of music, the therapist gradually leaves the ini- tion. When the client (i) actively explores the environ-
tiative to continue this activity to the client, anticipating ment, (ii) not only enjoys the closeness of the therapist
each intended ‘rock’. The therapist may start with phase but also enjoys playing together and playing next to the
1.3 when the therapist and the client can easily establish therapist, and (iii) when the client continues exploration
synchronous interaction, when the client takes the ini- when the therapist leaves the room, then phase 2 of the
tiative to indicate that he ⁄ she wants to have pleasurable intervention may be introduced.
contact with the therapist, and when the client focuses Phase 2 is aimed at replacing remaining maladaptive
not only on him or herself but also on the contact with behaviours by teaching socially acceptable alternative
the therapist. behaviour. Using functional behaviour analysis (e.g.
Subsequently (phase 1.3), the therapist stimulates Northup et al. 1991, 1994; Sigafoos & Meikle 1996),
exploration of the environment, offering verbal or non- socially acceptable behaviour is identified, and through

 2007 The Authors. Journal compilation  2007 Blackwell Publishing Ltd, 21, 126–135
130 Journal of Applied Research in Intellectual Disabilities

systematic ‘chaining’ and ‘fading’ new appropri- the opportunity to complete the observation lists. Obser-
ate behaviour is taught. Verbal and nonverbal social vation results were reported and evaluated every
reinforcements such as affectionate pats, hugs, approval month, ensuring the subsequent continuation of their
and praise are given. More information on the beha- use. Challenging behaviour was recorded every waking
viour modification therapy is provided later on in this hour of each day during the intervention over a 12-
article under the section Procedure. month period. Frequency was rated by the number of
The last phase (phase 3: generalization) starts when times the challenging behaviour had occurred during
challenging behaviour has diminished and the client the past hour. These frequencies were summed to pro-
uses socially acceptable, appropriate behaviour instead. vide a total frequency for the data. The day totals were
The therapist facilitates contact between the client and combined into weekly average scores, which were used
the regular caregivers by giving feedback on the care- in the data analysis. Intraclass correlation coefficients
givers’ sensitive responses to the client during video- were computed for each client to gauge the reliability of
training and team discussions. Over time, the number of the weekly averages. Because the weekly averages were
sessions is slowly reduced until the therapy stops. The determined on the basis of scores from different days
caregivers continue to invest in the bond with the client from different raters (caregivers), the criterion for reliab-
through being sensitive and responsive towards him or ility should be based on the product of acceptable inter-
her, as taught by the therapist. The therapy ends when rater reliability and short-term stability. Taking 0.80 as a
the transfer of the principles of sensitivity and respon- criterion for both, we therefore examined whether the
sivity to the caregivers is completed, when the care- intraclass coefficients for the weekly averages were 0.64
givers can sooth the child when distressed and when (0.80 · 0.80) or above. This was the case for four of the
the caregivers appropriately stimulate the client’s explora- six clients (range 0.64–0.81). Apparently, challenging
tion of objects and the environment. behaviour either occurred and ⁄ or was reported in
highly inconsistent patterns in the remaining two cases.
These cases were therefore dropped from the analyses
Instruments
of challenging behaviour in the residential setting.
Challenging behaviour in the residential home
Challenging and adaptive behaviour in the therapy sessions
Standardized instruments were used to assess the chal-
lenging behaviour shown by the client at the start of the During the behaviour therapy, one of the two therapy
intervention as well as after its completion. These instru- sessions by the attachment therapist and by the control
ments were the ‘Severe Challenging Behaviour Con- therapist was videotaped each week. A weekly fre-
sensus Protocol – National Institute for Health Care quency for video analysis was chosen with the expecta-
Management (CEP) with a Cohen’s kappa of 0.91 and tion that change would occur over weeks, not days.
with a demonstrated high external validity (Kramer Observers who were blind to therapist status and phase
2001), and the ‘Challenging Behaviour Scale for People of therapy independently coded the sessions in random
with an Intellectual Disability’ (Storend Gedragsschaal order using Noldus computer software (Noldus Infor-
voor Zwakzinnigen or SGZ) with a Cohen’s kappa mation Technology, Wageningen, The Netherlands). The
of between r = 0.75 and 0.79 and adequate validity observers recorded the occurrence of the challenging
(Kraijer & Kema 1994). Two caregivers under the behaviour and the adaptive replacement behaviour. The
supervision of a psychologist scored the instruments mean duration of these behaviours was calculated for
independently. each session. Interrater reliability was analysed in 15
randomly selected sessions. Twenty-minute checks were
carried out for each of the 15 videotapes evenly divided
Residential observation lists for challenging behaviour
over the two therapists and over different clients. For
The professional caregivers in the clients’ residential the frequency of target challenging behaviour, interob-
homes were instructed to record the frequency of the server agreement was 78% and Cohen’s kappa was 0.74;
challenging behaviours listed in Table 2. The observa- and for duration it was 93% and 0.92, respectively.
tion lists were easy to score and fitted into the caregiv- For the frequency of adaptive replacement behaviour,
er’s daily reporting routine. Because these clients had interobserver agreement was 81% and Cohen’s
severe challenging behaviour, they received individual kappa was 0.78; and for duration it was 91% and 0.90,
care or participated in a small group, giving caregivers respectively.

 2007 The Authors. Journal compilation  2007 Blackwell Publishing Ltd, 21, 126–135
Journal of Applied Research in Intellectual Disabilities 131

effects was performed in accordance with Nugent’s


Procedure
(2000) method of visual data analysis using the weigh-
During the attachment-based therapy (phase 1), the ted mean trend line for the intervention. In addition,
attachment therapist conducted three systematic 1-h Nugent’s weighted average trend across the sessions, an
sessions every week. During phase 1 the control therapist, index of the average rate of change in challenging beha-
who only positively engaged with that same client by viour was used.
for example reading aloud, singing or sitting next to the
child listening to music, visited the client every week
Meta-analytic combination of treatment effects across
for one 1-h session. The main goal for the control ther-
single-case studies
apist was to become a familiar figure with whom the cli-
ent experienced positive engagement. It was expected The results of the six single cases were meta-analytically
that one 1-h weekly session would be sufficient to combined into an overall average effect size, using the
achieve this goal. P-values. For every P-value a natural logarithm was
After having completed the attachment-based inter- determined. The sum of the natural logarithms was then
vention (phase 1), the remaining challenging behav- multiplied by –2. The result was a chi-squared devi-
iours were screened and the behaviours the caregivers ation, with twice the number of P-values as the degree
considered to be the most disturbing were selected as of freedom. In cases in which the client had P-values of
target behaviours. Staff members who routinely worked <0.005, the possibility that significance was based solely
with the clients were interviewed and asked about the on that one client was excluded by replacing this value
situations that typically evoked challenging behaviour. with 0.01. If the change was in a non-expected direction,
Antecedents–behaviours–consequences (ABC) data were a P-value of 0.5 was used, irrespective of the actual
collected during the selected situations within which P-value (de Weerth & Van Geert 2002).
the challenging behaviour occurred (e.g. Northup et al.
1991, 1994; Sigafoos & Meikle 1996), for example dur-
Results
ing free time, social interaction or daily care. The iden-
tification of appropriate replacement behaviours was
Changes in challenging behaviour during the integrative
aided by separate functional analyses for each challen-
therapy
ging behaviour (Mace et al. 1986). For each client, three
socially acceptable adaptive replacement behaviours The non-parametric sign test for two related samples
that could serve as functional alternatives were defined. showed that for the six clients the CEP score was signifi-
The target challenging behaviour and the adaptive cantly lower at the end compared with that the start of
replacement behaviour that were focused on during the intervention (mean before = 3.17, mean after = 2.17,
behaviour modification (phase 2 of the treatment) are Pone-sided < 0.034). All the clients except client B dropped
shown in Table 2, columns 3 and 4. During behaviour from clinical to sub-clinical levels. The increase in the
modification (phase 2), both therapists conducted total SGZ score failed to reach significance (mean
two 1-h therapy sessions every week, using the same before = 5.29, mean after = 5.75, Pone-sided < 0.059). All
behaviour modification protocol. The treatment was except clients B and C showed a higher score.
completed with the last phase of the intervention Figure 1 shows the weekly aggregated frequencies of
(phase 3 – generalization) being conducted by the the challenging behaviour that were recorded by the
attachment therapist. caregivers in the residential home, in all phases and
sub-phases for the four clients with sufficiently reliable
scores. Visual inspection aided by trend lines (Nugent
Data analysis
2000) showed a decrease in challenging behaviour from
phase 1.1 onwards, with a slight increase in phase 2
Visual inspection
when behaviour therapy commenced and a further
The effects of the intervention on the client’s challenging decrease in phase 3 after completion of the therapy and
behaviour in the residential setting were studied using generalization of its effects. The weighted mean trend
the weekly mean frequency of all challenging behaviour statistics were: for client A ) 0.090, client B ) 0.014, cli-
scored on the residential lists. We expected a decrease ent C ) 0.050, and for client F ) 0.048. These values
in challenging behaviour to occur as early as the attach- show the smallest change for client B. This client also
ment-based treatment phase. The visual inspection of had most of her scores above the weighted mean trend

 2007 The Authors. Journal compilation  2007 Blackwell Publishing Ltd, 21, 126–135
132 Journal of Applied Research in Intellectual Disabilities

Therapist 1 Therapist 2
3.0 3.0

2.5

Mean frequency of all


2.5

challeging behavior
Mean frequency of all
challeging behavior

2.0 2.0

1.5 1.5

1.0 1.0

0.5
0.5
0.0
0.0 1 5 9 14 18 22 26 30
2 4 6 8 10 12 14 16 18 20 22 24 26 3 7 12 16 20 24 28 32
Period preceding the therapy session Period preceding the therapy session
Client A Client B

3.0 3.0

2.5 2.5

Mean frequency of all


Mean frequency of all

challeging behavior
challeging behavior

2.0 2.0

1.5 1.5

1.0 1.0

0.5 0.5

0.0 0.0
1 5 9 13 17 21 25 29 33 1 5 9 13 17 21 25 29 33
3 7 11 15 19 23 27 31 3 7 11 15 19 23 27 31
Period preceding the therapy session Period preceding the therapy session
Client C Client F

Figure 1 Frequency of all challenging behaviour scored on the residential observation lists and mean trend lines (solid arrows).

line. Overall, these results indicate that challenging


behaviour in the residential home decreased from the Table 3 Mean and standard deviation (SD) of the duration
start to the end, and for the four clients studied in detail of the appropriate behaviour during behaviour modification
we consistently found a downward trend line. (phase 2)

Attachment therapist, Control therapist,


Differences in effect of behaviour modification between Client mean (SD) [n] mean (SD) [n] F (d.f. 1, d.f. 2)
attachment and control therapist
A 8.98 (12.06) 1.61 (0.89) 2.60 (1, 12)
Observations of behaviour modification sessions con- [7] [7]
ducted by the control therapist were compared with B 36.14 (6.03) * 26.44 (10.76) 4.33 (1, 12)
the sessions conducted by the attachment therapist to [7] [7]
test the hypothesis that the experimental therapist, hav- C 11.96 (6.8) 16.00 (10.63) 0.72 (1, 12)
ing stimulated an attachment relationship, would be [7] [7]
more effective in applying behaviour modification than D 41.00 (4.65)*** 26.90 (10.38) 10.76 (1, 12)
a familiar control therapist. The duration of the shaped [7] [7]
adaptive behaviour and the target challenging beha- E 3.51 (2.16) 2.67 (0.58) 0.99 (1, 12)
[7] [7]
viour in this phase were analysed. Table 3 shows the
F 1.10 (0.92) 0.92 (0.96) 0.14 (1, 13)
mean duration of the shaped adaptive behaviour. A
[8] [7]
longer duration of adaptive behaviour was found in
the sessions conducted by the attachment therapist n = number of videotaped behaviour modification sessions.
except for client C. The Fisher’s combination of *P < 0.06, **P < 0.05, ***P < 0.01

 2007 The Authors. Journal compilation  2007 Blackwell Publishing Ltd, 21, 126–135
Journal of Applied Research in Intellectual Disabilities 133

Table 4 Mean and standard deviation (SD) of the duration of ment relationships, the severe levels of challenging
target challenging behaviour that was the focus of the behaviour had already dropped drastically. All in all,
behaviour modification (phase 2) the integrative attachment-based behaviour modifica-
tion therapy showed promise as a therapeutic method
Attachment therapist, Control therapist,
to treat children with severe intellectual and visual
Client mean (SD) [n] mean (SD) [n] F (d.f. 1, d.f. 2)
disabilities who exhibit severe behaviour problems
A 0.42 (0.18) 0.40 (0.08) 0.05 (1, 12)
and are withdrawn and unresponsive to social stimu-
[7] [7] lation.
B 0.09 (0.23) 0.10 (0.18) 0.01 (1, 12) Our visual analyses of the daily scored residential
[7] [7] observation lists and results from standardized ratings
C 0.05 (0.06) 0.08 (0.05) 0.81 (1, 12) suggested that the integrative psychotherapeutic inter-
[7] [7] vention, a combination of an attachment-based treat-
D 0.34 (0.41) 2.87 (7.08) 0.90 (1, 12) ment and behaviour modification, was effective in
[7] [7] reducing challenging behaviour. It is important to note
E 2.28 (1.72) 4.18 (3.40) 1.74 (1,12) that these results were found in clients for whom in the
[7] [7]
past no other therapies and interventions had proven
F 0.09 (0.14) 0.60 (0.98) 2.09 (1, 13)
successful.
[8] [7]
No significant difference was found between the
n = number of videotaped behaviour modification sessions attachment and the control therapists in the duration of
the target challenging behaviour during behaviour
modification. Testing for differences was difficult
P-values for the one-way anova for the six clients indi- because the frequency and duration of the challenging
cated a significant overall longer duration of adaptive behaviour during these sessions were already quite low
behaviour during behaviour modification given by the at this point in the therapy. Future experimental studies
attachment therapist (combined v2 deviation = 25.34, are needed to establish whether attachment-based ther-
P < 0.025). Table 4 shows the mean duration of the apy by itself reduces challenging behaviour in clients
target challenging behaviour, the focus of behaviour with severe multiple disabilities.
modification. For all clients, except for client A, Table 4 The results of the part of the therapy aimed at estab-
shows less target challenging behaviour during beha- lishing a relationship with the client were generalized to
viour modification in the sessions conducted by the the caregivers in phase 3. By adding a generalization
attachment therapist compared with those conducted phase, during which contact between caregivers and cli-
by the control therapist, but the absolute values were ent was stimulated and the therapeutic sessions were
very low. Probably as a result, we found no significant gradually reduced, clients were protected from experi-
difference. encing the end of the intervention as a loss. In this
In conclusion, the effects for the six clients combined phase the clients had the opportunity to build new,
showed that during behaviour modification the attach- promising and long-lasting relationships with their care-
ment therapist was significantly more effective in stimu- givers.
lating adaptive behaviour. No significant differences The results indicated that for client B the effect of
were found between the therapists in their effectiveness the integrative treatment was limited. She was the cli-
with respect to reducing the fairly low levels that had ent with the lowest cognitive and physical skills (she
remained of target challenging behaviour during the had cerebral palsy and was confined to a wheelchair).
behaviour therapy sessions. Because of her severe disabilities there was limited
scope for teaching this client alternative behaviour. For
example, the focus of behaviour modification was on
Conclusion
making low noises instead of screaming loudly,
The clients in this study appeared to learn replace- because no other alternative could be found for her tar-
ment behaviours more easily from therapists who had get challenging behaviour. It is however important to
previously attempted to build a therapeutic relation- note that even for this client the attachment therapist
ship based on attachment principles, compared with was more effective than the control therapist in modi-
therapists who were just familiar. By the time the fying her behaviour. She did show longer-lasting
therapists had concluded the phase of building attach- appropriate behaviour (although it was a statistical

 2007 The Authors. Journal compilation  2007 Blackwell Publishing Ltd, 21, 126–135
134 Journal of Applied Research in Intellectual Disabilities

trend; P < 0.060) and overall the challenging behaviour tocol, in order to find out whether children and care-
significantly less challenging behaviour during the givers are able to capitalize on a newly established
behaviour modification (F = 8.14, d.f. = 1, 12, P < 0.015) sensitivity to social rewards and interactions to
conducted by the attachment therapist (mean = 1.66, develop trusting relationships within the context of
SD = 2.10, n = 7) compared with that given by the residential care, given the challenges of staff turnover
control therapist (mean = 5.91, SD = 3.34, n = 7). and workloads. Ultimately, the most important effect
Despite the gains during the treatment sessions, the of the therapy may be the increased reactivity of chil-
scores on the standardized instruments, the CEP and dren with severe disabilities to their social environ-
the SGZ, remained unchanged. It is possible that in the ment. The therapy may have long-term effects if the
case of such a dependent client with multi-impairments caregiving environment sustains this social reactivity
an extra intervention is needed for the transference of and continues to stimulate the development of adap-
the newly learnt behaviour as shown during the ther- tive behaviour using positive and social means of
apy sessions to the residential setting. Client D dis- communication.
played considerable fluctuations in her behaviour, with
many outliers and extremes, revealing a highly unrelia-
Acknowledgments
ble pattern on the residential observation lists. How-
ever, the standardized instruments did show lower This study received funding from InSight, a Dutch
post-test scores. association promoting application-oriented research to
The integrative treatment was developed for clients support the needs of people with a visual impairment.
for whom social contingencies failed to reinforce adap- We are grateful for the cooperation and assistance of
tive behaviour. Our hypothesis was that if it is possible parents, caregivers and observers. The authors acknow-
for clients to develop a certain sensitivity to social ledge the invaluable contributions of Francien Dekker,
rewards, this would not only have a therapeutic effect, psychotherapist, as co-attachment and control therapist.
that is a reduction in challenging behaviour but they
would also be able to learn new, appropriate behaviour,
Correspondence
at least from the therapists for whom they developed
this sensitivity. This hypothesis was confirmed by the Any correspondence should be directed to P.S. Sterken-
study results. This might be explained by the higher burg, Vrije Universiteit Amsterdam, Faculty of Psycho-
reward value of positive social contingencies on beha- logy and Education, Department of Clinical Child and
viour, as well as by better affect regulation during the Family Studies, Van der Boechorststraat 1, 1081 BT
presence of the attachment therapists, allowing clients to Amsterdam, the Netherlands (e-mail: ps.sterkenburg@
concentrate better on learning. psy.vu.nl)

Study limitations and directions for further research References


In this study we focused on the effect of an integrative Bowlby J. (1984). Attachment and Loss. Volume 1: Attachment.
intervention. The quality and characteristics of the Penguin Books, London (first published in 1969).
therapeutic relationship as attachment relationship were Bradley S. J. (2000) Affect Regulation and the Development of Psy-
not examined in this report, nor was the actual affect- chopathology. Psychology Press, London.
Didden R., Duker P.C. & Korzilius H. (1997) Meta-analytic
regulating function of the attachment relationship in
study on treatment effectiveness for problem behaviors with
these children analysed. Such data will be important to
individuals who have mental retardation. American Journal on
further explain the effects of the therapy. Another limi- Mental Retardation 101, 387–399.
tation was that the design did not allow for examination Došen A. (1984) Psychische stoornissen bij zwakzinnige kinderen.
of the independent effect of the attachment-based ther- Swets & Zeitlinger, Lisse.
apy phase. A reversal design (ABAB) was deemed Došen A. (1990) Psychische en gedragsstoornissen bij zwakzinnigen.
undesirable, because that would have disrupted the Boom, Meppel.
developing relationship. Finally, the number of sessions Došen A. (2001) Developmental-Dynamic Relationship Therapy:
in the behaviour modification phase limited statistical an approach to more severely mentally retarded children. In:
power in testing treatment effects. Treating Mental Illness and Behavior Disorders in Children and
Further research is needed to examine the effect of Adults with Mental Retardation (eds A. Došen & K. Day),
pp. 415–427. American Psychiatric, London.
the generalization component of the intervention pro-

 2007 The Authors. Journal compilation  2007 Blackwell Publishing Ltd, 21, 126–135
Journal of Applied Research in Intellectual Disabilities 135

Došen A. (2005). Psychische stoornissen, gedragsproblemen en vers- O’Reilly M., Lancioni G. & Taylor I. (1999) An empirical analy-
tandelijke handicap. Een integratieve benadering bij kinderen en sis of two forms of extinction to treat aggression. Research in
volwassenen. Van Gorcum, Assen. Developmental Disabilities, 20, 315–325.
Fisher P. A., Gunnar M. R., Chamberlain P. & Reid J. B. O’Reilly M. F., Murray N., Lancioni G. E., Sigafoos J. & Lacey
(2000) Preventive intervention for maltreated preschool C. (2003) Functional analysis and intervention to reduce self-
children: impact on children’s behavior, neuroendocrine injurious and agitated behavior when removing protective
activity, and foster parent functioning. Journal of the Ameri- equipment for brief time periods. Behavior Modification, 27,
can Academy of Child and Adolescent Psychiatry, 39, 1356– 538–559.
1364. Porges S.W. (2004) Social engagement and attachment: a phylo-
Janssen C. G. C., Schuengel C. & Stolk J. (2002) Understanding genetic perspective. Annals of the New York Academy of Sci-
challenging behavior in people with severe and profound ences, 1008, 31–47.
intellectual disabilities: a stress attachment model. Journal Prout H. T. & Nowak-Drabik K. M. (2003) Psychotherapy with
of Intellectual Disability Research, 46, 445–453. persons who have mental retardation: an evaluation of effec-
Jones R. S. P. & McCaughey R. E. (1992) Gentle teaching and tiveness. American Journal on Mental Retardation, 108, 82–93.
applied behavior analysis: a critical review. Journal of Applied Schuengel C. & Janssen C. G. C. (2006) People with mental
Behavior Analysis, 25, 853–867. retardation and psychopathology. Stress, affect regulation
Kraijer D.W. & Kema G. N. (1994) SGZ, Storend Gedragsschaal and attachment. A review. International Review of Research in
voor Zwakzinnigen. Handleiding. Swets & Zeitlinger, Lisse. Mental Retardation, 32, 229–260.
Kramer G. J. A. (ed.). (2001). Consensusprotocol Ernstig Prob- Scotti J. R., Evans I. M., Meyer L. H. & Walker P. (1991) A
leemgedrag. Handleiding voor het beschrijven en beoordelen van meta-analysis of intervention research with problem beha-
probleemsituaties rond cliënten van de gehandicaptenzorg. [Severe vior: Treatment validity and standards of practice. American
challenging behaviour consensus protocol. Manual for the Journal on Mental Retardation, 96, 233–256.
description and assessment of problem situations in the care Sigafoos J. & Meikle B. (1996) Functional communication train-
of clients with an intellectual disability]. Vereniging Gehandi- ing for the treatment of multiply determined challenging
captenzorg Nederland, Utrecht. behavior in two boys with autism. Behavior Modification, 20,
Mace F. C., Page T. J., Ivancic M. T. & O’Brien S. (1986) Analy- 60–84.
sis of environmental determinants of aggression and disrup- Sturney P. (2005) Against psychotherapy with people who have
tion in mentally retarded children. Applied Research in Mental mental retardation. Mental Retardation, 43, 55–57.
Retardation, 7, 203–221. Verhoeven W. M. A. & Tuinier S. (1996) The effect of buspirone
Northup J., Wacker D., Sasso G., Steege M., Cigrand K., Cook J. on challenging behaviour in mentally retarded patients: an
& Raad de A. (1991) A brief functional analysis of aggressive open prospective multiple-case study. Journal of Intellectual
and alternative behavior in an outclinic setting. Journal Disability Research, 40, 502–8.
of Applied Behavior Analysis, 24, 509–522. de Weerth C. & Van Geert P. (2002) Changing patterns of infant
Northup J., Wacker D. P., Berg W. K., Kelly L., Sasso G. & Raad behavior and mother-infant interaction: intra- and interindivi-
de A. (1994) The treatment of severe behavior problems in dual variability. Infant Behaior & Development, 24, 347–371.
school settings using a technical assistance model. Journal Zilberstein K. (2006) Clarifying core characteristics of attach-
of Applied Behavior Analysis, 27, 33–47. ment disorders: a review of current research and theory.
Nugent W. R. (2000) Single case design visual analysis proce- American Journal of Orthopsychiatry, 76, 55–64.
dures for use in practice evaluation. Journal of Social Service
Research, 27, 39–75.

 2007 The Authors. Journal compilation  2007 Blackwell Publishing Ltd, 21, 126–135

Das könnte Ihnen auch gefallen