Children with externalizing

Transcription

Children with externalizing
Children with
externalizing
behavior problems
Risk factors and
preventive efforts
Jill Thijssen
Children with externalizing behavior problems
Risk factors and preventive efforts
Jill Thijssen
ISBN:
Cover design: Printing: 978-94-6299-317-4
Guyon Muijres & Jill Thijssen
Ridderprint BV - www.ridderprint.nl
© Jill Thijssen, Maastricht, 2016
All rights are reserved. No part of this book may be reproduced or transmitted in
any form or by any means, without written permission from the author or, when
appropriate, the publisher of the article.
Children with externalizing behavior problems
Risk factors and preventive efforts
Proefschrift
ter verkrijging van de graad van doctor aan de Universiteit Maastricht,
op gezag van de Rector Magnificus, Prof. dr. L.L.G. Soete
volgens het besluit van het College van Decanen,
in het openbaar te verdedigen
op vrijdag 22 april 2016 om 12.00 uur.
door
Jill Thijssen
Promotores
Prof. dr. C. de Ruiter
Prof. dr. P. Muris
Beoordelingscommissie
Prof. dr. T. van Amelsvoort (voorzitter)
Prof. dr. M. Dekovic (Utrecht University)
Dr. L. van Domburgh (VUmc)
Dr. J. Roelofs
The research presented in this dissertation was supported by The Netherlands
Organisation for Health Research (ZonMw, grant number 157001022), Innovatiefonds,
VSB fonds, and Kinderpostzegels Nederland.
CONTENTS
Chapter 1 General introduction
7
PART I: Parent Management Training – Oregon model
21
Chapter 2 The effectiveness of Parent Management Training - Oregon Model 23
in clinically referred children with externalizing behavior problems
in The Netherlands
Chapter 3 Treatment fidelity as a determinant of the effectiveness of Parent 47
Management Training – Oregon model in The Netherlands
Chapter 4 Initial validation of the Dutch translation of the Caregiver Wish List, 63
an interview-based scale for measuring parenting practices
PART II: Callous-unemotional traits
81
Chapter 5 Emotional memory for central and peripheral details in children 83
with callous-unemotional traits
Chapter 6 Emotional true and false memories in children with callous-
95
unemotional traits
Chapter 7 General discussion
Summary
Samenvatting
Valorization addendum
References
Dankwoord (acknowledgments)
Curriculum Vitae
105
123
129
135
143
157
163
General
introduction
1
This chapter is a translated and extended version of the following chapter:
Thijssen, J., de Ruiter, C., & Albrecht, G. (2008). Preventie van antisociaal gedrag bij kinderen:
Parent Management Training Oregon. [Prevention of antisocial behavior in children:
Parent Management Training Oregon]. In J.R.M. Gerris en R.C.M.E. Engels (Eds.), Vernieuwingen in
jeugd en gezin: Beleidsvisies, gezinsrelaties en interventies (p. 125 - 140). Assen: Van Gorcum.
General introduction | Chapter 1
The case of Tim
Tim is eight years old and has a difficult temperament. His parents are divorced
and he lives with his mother Kate. Tim spends every other weekend with his father.
Kate does not have a lot of financial means and she feels depressed which makes her
less patient with Tim. Kate gets easily irritated by his disobedient behavior, while Tim
is frustrated because he feels he cannot do anything right. Positive behaviors do not
meet with compliments, which makes Tim use negative behaviors to attempt to obtain
what he wants. The negative behavior between parent and child becomes a standard
in the lives of Kate and Tim. Because Tim regards negative behavior as normal, he also
starts to demonstrate negative behavior at school. His friends reject him because of his
aggressive behavior, which makes Tim develop low self-esteem.
Over the years, the negative interactions between Kate and Tim become more
ingrained and the emotional distance between mother and son keeps growing. Kate
does not have a grip on Tim anymore and she has no idea what he is doing in his spare
time. At age 13, Tim starts hanging out with older delinquent boys who encourage him
to commit crimes. Because Tim does not feel supported at home and his delinquent
friends do not think school is important, he drops out of school at the age of 16.
Regularly, he gets into trouble with the police. At the age of 20, Tim gets arrested on
suspicion of dealing drugs.
Externalizing behavior problems
In the Netherlands, 13.6% of the children between 4 and 11 years of age show
externalizing behavior problems (de Looze et al., 2014). Externalizing behavior problems
refer to aggressive (fighting, destroying), oppositional (disobedient, running away), and
delinquent behaviors (lying, stealing, substance abuse). Aggressive behavior is part of
the normal development of children (Tremblay, 2000). Between the age of 2 and 3, most
children start to show physical aggression as a natural way of expressing anger. After
their third birthday, when they learn to talk, there is a decline in the level of physical
aggression (Alink et al., 2006). However, some children continue to show a high level
of aggression and are at risk of developing a chronic pattern of physical aggression
and delinquency (Broidy et al., 2003; Nagin & Tremblay, 1999). This can result in the
development of disruptive behavior disorders, such as Oppositional Defiant Disorder
(ODD) and Conduct Disorder (CD; American Psychiatric Association [APA], 2013).
ODD consists of a persistent pattern of negativistic, defiant, disobedient, and hostile
behavior towards authority figures. Children with ODD annoy other people, refuse to
comply with requests or rules of adults, blame others for their own mistakes, are easily
irritated and often angry (APA, 2013). CD comprises a recurrent pattern of behavior in
which the basic rights of others or major age-appropriate societal norms or rules are
9
1
Chapter 1 | General introduction
violated. Children with CD steal, vandalize, cause physical harm to other people or
animals, and lie (APA, 2013). ODD is often seen as a precursor or less severe form of CD
(Loeber & Burke, 2011). Both disorders have a high co-morbidity with Attention Deficit
Hyperactivity Disorder (ADHD; APA, 2013).
Early signs and consequences
Several longitudinal studies have demonstrated that already during toddlerhood
there are behavioral characteristics which predict later antisocial behavior (Coté,
Vaillancourt, LeBlanc, Nagin, & Tremblay, 2006; Simonoff et al., 2004). Indeed, the earlier
externalizing behavior problems begin, the higher the risk of serious problems during
adulthood (Moffitt, 1993; Patterson, DeBaryshe, & Ramsey, 1989). In toddlers, behavioral
problems are characterized by behaviors such as disobedience, temper tantrums and
hitting. During the school years, new forms of antisocial behavior develop, such as
fighting, stealing, and cruelty to animals, while substance abuse and robbery occur
in adolescence (Patterson, Dishion, & Yoerger, 2000). When externalizing behavior
problems in childhood are not addressed properly, about 50 to 75% will continue to
show serious problem behaviors up to six years later (Nixon, 2002).
Externalizing behavior problems are a major social and financial burden on the
community (Bennett et al., 1999). Scott, Knapp, Henderson, and Maughan (2001)
examined the financial costs of children with disruptive behavior problems for public
services. They followed 142 10-year-old children from inner London up to adulthood
(until age 28). Based on their childhood profile, the 142 children were divided into
three groups: children with no problems, children with some disruptive behavioral
problems, and children with conduct disorder. They found that by age 28, the costs
for individuals with conduct disorder (£70 019) were 10 times higher than for children
with no problems (£7 423) and 3.5 times higher than for children with mild behavioral
problems (£24 324). The highest costs were incurred by crime, followed by the provision
of extra education, and foster and residential care (Scott et al., 2001). Romeo, Knapp,
and Scott (2006) found that the mean annual costs for children with severe antisocial
behavior in the UK were £5 960. The National Health Service, education and volunteer
agencies were the most commonly used services. However, over three-quarters of the
total costs fell on the family itself (e.g., additional household tasks, house repairs due
to the child’s destructiveness, taking days off work because the child was sent home
from school). Thus, the financial costs of externalizing behavior problems are large and
involve several agencies (Romeo et al., 2006; Scott et al., 2001).
10
General introduction | Chapter 1
Parenting
A large number of risk factors have been reported as being associated with
the development and persistence of externalizing behavior problems, such as low
serotonin level, low resting heart rate, impulsivity, difficult temperament, associations
with deviant peers, and living in a disadvantaged neighborhood (Frick, 2006). Probably
the most studied and established risk factor for the development of externalizing
behavior problems is ineffective parenting (DeBaryshe, Patterson, & Capaldi, 1993;
Forgatch, Bullock, & Patterson, 2004; McCoy, Frick, Loney, & Ellis, 1999; Nix et al., 1999;
Patterson, Forgatch, Yoerger, & Stoolmiller, 1998). In particular, harsh and inconsistent
discipline practices, poor monitoring and supervision, and low positive involvement
have been associated with externalizing behavior problems in a number of studies
(Bierman & Smoot, 1991; Cunningham & Boyle, 2002; Ehrensaft et al., 2003; Frick et al.,
1992; Nicholson, Fox, & Johnson, 2005).
Persistent ineffective parenting can result in the development of overt forms of
externalizing behavior problems, such as hitting, temper tantrums and noncompliance.
Children learn and maintain the overt forms of externalizing behavior by negative
reinforcement of antisocial behavior by family members (Snyder & Patterson, 1995).
In a well-functioning family, the child learns to use prosocial skills during conflicts.
However, in a dysfunctional family, the payoffs are higher for coercive (negative)
responses than for prosocial reactions (see the case of Tim presented above). Coercive
processes usually become automatic with little or no cognitive awareness (Forgatch et
al., 2004). Prosocial behaviors in such families remain unnoticed and are not rewarded.
The antisocial behaviors learned at home can generalize to other settings, such as
school and sports clubs (DeBaryshe et al., 1993). Noncompliance with teachers puts
these children at risk of poor school achievement (Ledingham & Schwartzman, 1984;
Prior, Smart, Sanson, & Oberklaid, 1999). With peers, the antisocial behavior of the child
can lead to rejection by prosocial peers (DeGarmo & Forgatch, 2005; Patterson, 1986).
School failure and rejection by peers can lead to low self-esteem and depressed mood,
but also to associations with deviant peer groups (DeGarmo & Forgatch, 2005; Snyder,
2002). Through interaction with deviant peers, children learn covert forms of antisocial
behavior, such as lying, stealing and substance abuse (Forgatch, Patterson, DeGarmo,
& Beldavs, 2009). In deviant peer groups it pays off to show deviant behavior. However,
parents also contribute to covert antisocial behaviors by coercive parenting and poor
monitoring. The latter enables the youngsters to wander away from home and to
engage in undesirable activities. Eventually, this can lead to delinquency. Thus, overt
antisocial behavior is maintained through negative reinforcement by family members,
while covert antisocial behavior is mainly maintained by positive reinforcement by
deviant peer group members (Forgatch & Patterson, 2010). Youngsters who combine
11
1
Chapter 1 | General introduction
overt and covert forms of antisocial behavior problems are at highest risk to become
career offenders (Patterson, Forgatch, & DeGarmo, 2010).
As outlined above, externalizing behavior problems in childhood can have major
social and financial consequences. Therefore, it is important to intervene as soon as the
child shows signs of externalizing behavior problems. Many interventions for childhood
behavior problems focus on increasing effective parenting practices and decreasing the
use of ineffective rearing behaviors. Results of several studies support the theory that an
improvement of parenting practices is the most important mechanism that contributes
to more adaptive behavior of the child (DeGarmo & Forgatch, 2005, 2007; Forgatch &
DeGarmo, 1999; Martinez & Forgatch, 2001; Ogden & Amlund-Hagen, 2008).
Social Interaction Learning theory
The Social Interaction Learning (SIL) theory describes the mechanisms that could
stimulate or prevent antisocial behavior. Figure 1 shows that the child is surrounded by
two layers of context. The inner layer represents the parents, while the outer layer refers
to the background context. On the left side of the inner layer are the effective (positive)
parenting practices (discipline, monitoring, positive involvement, skill encouragement,
and problem solving). On the right side are the coercive (negative) parenting practices
(negative reciprocity, escalation, and negative reinforcement), which hinder a healthy
social-emotional development of the child (Forgatch et al., 2004). The layer of the
parenting practices is surrounded by contextual factors. These contextual factors
influence the quality of parenting. For example, when a father experiences a lot of
stress from work, he may become more irritable and less patient with his child at home.
Therefore, these contextual factors indirectly also have an effect on the development of
the child. On the other hand, parents also influence the wider social context, for instance,
they can make sure the child grows up in a safe environment (Forgatch et al., 2004).
The parenting skills will be briefly explained. Discipline decreases the problem
behavior by appropriate and consistent use of mild sanctions, such as giving a time
out or taking away privileges (Patterson, 1986). Monitoring involves knowing the
child’s friends and keeping track of the child’s activities to protect the child against
associations with deviant peers (Snyder, 2002). Positive involvement concerns the many
ways in which parents give their child loving attention, spending time together and
engaging in fun activities (Forgatch & DeGarmo, 1999; Forgatch, Patterson, & DeGarmo,
2005b). Skill encouragement stimulates the development of prosocial behaviors by
means of scaffolding techniques (e.g., breaking behavior into small steps, prompting
appropriate behavior) and positive reinforcement (e.g., giving compliments when the
child behaves well; Forgatch & DeGarmo, 1999). Finally, the parenting practice problem
solving helps parents to negotiate during arguments and to clearly determine rules and
consequences when the rules are violated (Forgatch et al., 2005b).
12
General introduction | Chapter 1
1
䘀愀洀椀氀礀
猀琀爀甀挀琀甀爀攀
䔀猀挀愀氀愀愀漀渀
匀漀挀椀愀氀
猀甀瀀瀀漀爀琀
一攀最愀愀瘀攀
爀攀椀渀昀漀爀挀攀ⴀ
洀攀渀琀
Figure 1 | The Social Interaction Learning model (Forgatch et al., 2004)
Coercion is a pattern of social interaction which hinders the development of
prosocial behaviors in the child and stimulates problem behavior. Coercive parenting
is characterized by irritable commands and negative attributions about others, but also
by other ineffective techniques such as inconsistency, threatening but not following
through, and high emotional intensity sanctions (Martinez & Forgatch, 2001; Patterson
et al., 2010). Negative reciprocity arises when parents pay more attention to what the
child is doing wrong than to what the child is doing right and when parents frequently
start arguing with their child. This often leads to escalation. When the parent does not
stop the argument in time, emotions will get the upper hand. Negative reciprocity
creates irritation in the parent and anger in the child. At the moment the escalation
reaches its peak, the child starts screaming and throwing things around, for example,
and the parent yells to the child to go upstairs. After the peak, the escalation decreases
and everything becomes relatively calm again. However, the confrontation has not
been solved in a satisfactory way, which means there is high risk of a new negative
reciprocity and escalation.
Through negative reinforcement children learn that negative behavior pays off in
the short term. For example, a child and the parent go to the supermarket and the child
wants some candy but does not get it. However, the child has learned from previous
experiences that when he screams long and hard, the parent will eventually give in and
buy what the child wants. In this way, parents develop the feeling that they do not have
13
Chapter 1 | General introduction
any control over their child’s negative behavior (Snyder, Edwards, McGraw, Kilgore, &
Holton, 1994).
Negative contextual factors, such as an unsafe neighborhood, low socio-economic
status, stressful life events (e.g., divorce), substance abuse, and mental health problems
can cause an imbalance between coercive and positive parenting practices. For example,
when the father loses his job, he may become more irritable which in turn can lead to
increased coercive parenting. Biological factors are also included among the contextual
factors. Several studies have shown that, for example, having a difficult temperament or
ADHD can lead to externalizing behavior problems (Bates, Pettit, Dodge, & Ridge, 1998;
Chronis et al., 2007; Johnston & Jassy, 2007; Simonoff et al., 2004). When families have
to deal with such negative contextual factors, adequate parenting practices become
even more important. Positive parenting practices ensure that the child is able to cope
with the negative influence of negative contextual factors, thereby acting as a buffer
(Forgatch et al., 2004).
PMTO
Parent Management Training - Oregon model (PMTO) is a training for parents of
children with serious externalizing behavior problems such as ODD and CD, but also
ADHD. It is based on the SIL model. PMTO focuses on children between 4 and 12 years
of age, because this type of preventive parent training is most effective with young
children. Reasons for this are that their behavior is still malleable and for children of
this age parents are still the main socializing agents (McCart, Priester, Davies, & Azen,
2006). The basic assumption of PMTO is that not only the child is the problem, but the
social environment of the child is involved as well. Changing the behavior of aggressive
children means that the social environment in which the child is living also needs to be
changed (Patterson, Reid, & Eddy, 2002). The parents or caretakers play a very important
role in maintaining the quality of the social environment of their child.
A core principle of the PMTO training is the balance of five positive reinforces
against one negative one (Forgatch et al., 2004). This means that for every limit set, five
compliments or encouragements need to be given. Parents who have children with
externalizing behavior problems generally pay attention to the problem behaviors,
while ignoring the positive, prosocial behaviors. Teaching parents to encourage their
children forces them to focus more on their child’s positive behavior. The 5:1 principle
teaches parents to recognize and reinforce positive behaviors, which makes the child
better able to accept limit setting (Forgatch et al., 2004).
14
General introduction | Chapter 1
Content of the PMTO intervention
PMTO is an individual therapy of 15 to 25 weekly one-hour sessions in which the
therapist works with the parent(s) of one family. The children are not present during
these sessions. The PMTO therapy aims to teach the parents how to reduce coercive
parenting practices and to replace them by the five effective parenting practices.
Usually, the therapy takes place at the site where the therapist works, but sometimes at
the parents’ home as well.
Role play is an important mechanism in the PMTO sessions to teach the parents
effective parenting practices. Simply telling parents what they should do, does not
lead to change. The new parenting practices have to be practiced repeatedly until they
become automatic (Forgatch et al., 2004). From other parenting therapies it was learned
that there was a lot of resistance among parents to change their behaviors, but by means
of the frequent use of role play this resistance tends to decrease (Patterson, 2005). The
therapist uses the role play to demonstrate good and bad examples of parenting. One
of the parents plays the role of the child and directly experiences the positive and
negative effect of these behaviors. Afterwards, the therapist asks the parents to indicate
which specific behaviors stimulated or hindered a positive outcome. The role play can
also be used as a diagnostic instrument to examine the strengths and weaknesses of
the parents, to determine what the parents have already learned, and to see which skills
need extra attention (Forgatch et al., 2004).
As soon as the parents have sufficiently mastered the pertinent parenting behavior,
the therapist and parent select a specific situation for the upcoming week in which the
parent will try to apply the newly acquired skill at home. This increases the chance that
the parent will actually use the skill in daily life. During that week, the therapist calls
the parent to offer support and to answer questions. In the subsequent session, the
therapist evaluates the success of the newly acquired parenting skill and discusses how
this could be further improved (van Leeuwen & Albrecht, 2008).
One of the parenting practices parents learn with PMTO is stimulating their children
by encouragement. The therapist uses this skill during the sessions with the parents as
well. When the parent says or does something right, the therapist gives a compliment
or a reward in the form of Scooby Loops. Scooby loops are elastic wrist bands which the
parents put on their wrist. The effect of the wrist bands for the parents is the same as
for the child: the confidence and the appropriate behaviors, in this case the adequate
parenting practices, will increase. Parents also use the wrist bands to remind themselves
to compliment their child. They start the day with, for example, five wrist bands and
each wrist band represents a compliment to the child. At the end of the day, all wrist
bands must have been used. The earned wrist bands can also be exchanged for a bigger
reward for the child, such as playing on the iPad for half an hour.
15
1
Chapter 1 | General introduction
Integrity of PMTO
The training to become a PMTO therapist encompasses 18 days. After the first nine
days of training, the therapist-in-training starts with his first PMTO-family. All therapy
sessions are recorded on DVD. The supervisor uses the DVD’s to give feedback, but also
for evaluating treatment fidelity. During the last nine days of the training, the therapistin-training brings the DVD’s of the therapy sessions for further coaching. A therapist has
to treat at least three families before the certification procedure can start. The therapist
is being rated on the basis of four DVD’s. When the therapist performs the PMTO
principles correctly, he receives the PMTO license. The duration of the licensing process
varies per therapist, because it is fitted to individual learning processes. In general, the
training takes about 13 to 15 months (Berger & van Everdingen, 2006). After finishing
the PMTO training, there will still be regular supervision meetings and the therapist has
to recertify every year to retain the PMTO license.
The DVD’s of the therapy sessions are judged by using the Fidelity of IMPlementation
rating system (FIMP; Knutson, Forgatch, Rains, & Sigmarsdóttir, 2009). This rating system
determines to what extent the therapist conducted the PMTO intervention according
to its principles. Research has shown that high FIMP scores are good predictors of
improvements in parenting practices. Therefore, being loyal to the PMTO principles
predicts a positive therapy outcome (Forgatch et al., 2005b). During supervision the
FIMP is used to see which FIMP elements the therapist is performing well and which
elements could be used more and/or better. With certification and recertification, the
FIMP can also help to make a decision. Hence, the FIMP is an important part of the PMTO
training.
Effectiveness of PMTO
The past four decades, a lot of research has been conducted on the effectiveness
of PMTO. In the 1970’s, a few small randomized controlled trials were performed which
showed that parent training is more effective with aggressive boys than a placebo
condition (Walter & Gilmore, 1973) and more effective than a waiting list control
condition (Wiltz & Patterson, 1974). Later, larger-scale trials were carried out which
replicated these results (Bank, Marlowe, Reid, & Patterson, 1991; Patterson, Chamberlain,
& Reid, 1982). PMTO appeared not only to be effective in complete families with two
biological parents. The effectiveness of PMTO has also been established in divorced
families (Forgatch & DeGarmo, 1999). A sample of 238 recently divorced mothers were
randomly assigned to PMTO or a no intervention control condition. After 12 months,
the researchers found that in the PMTO condition the effective parenting practices had
improved compared to mothers in the control condition. In a follow-up study, it was
found that the parenting practices of these mothers changed within the first 12 months,
16
General introduction | Chapter 1
followed by changes in the behavior of the children and eventually a significant decrease
in the depression level of the mother after 30 months (DeGarmo, Patterson, & Forgatch,
2004). Even at 9 years follow-up, there was a significant difference in delinquency and
the number of police arrests between the boys in the PMTO condition compared to the
control group (Forgatch et al., 2009).
The effectiveness of PMTO has also been investigated in the context of stepfamilies
(Forgatch, DeGarmo, & Beldavs, 2005a). The study consisted of 110 remarried biological
mothers and stepfathers who were randomly assigned to either PMTO or a no intervention
control condition. After 12 months, PMTO was found to produce clear improvements in
parenting practices and a significant reduction of externalizing behavior problems of
the child (Forgatch et al., 2005a). However, this effect was less clear-cut at 24 months.
This was mainly due to the fact that the quality of the parenting practices in the control
group had increased. Thus, at first, a remarriage can be problematic for the parenting
of the stepfather, but eventually his skills also improve during the natural course of
stepfamily formation (DeGarmo & Forgatch, 2007).
Finally, the effectiveness of PMTO was also examined in foster families. A permanent
placement is the best choice for foster children. To examine if PMTO is effective in
increasing the chance of a permanent placing, 90 children between 3 and 6 years old
were randomly assigned to either PMTO or treatment as usual. After 24 months, the
researchers found success rates of permanent placings of 64% for treatment as usual
and 90% for PMTO. Therefore, it can be concluded that PMTO increases the chance of
success of a permanent placing (Fisher, Burraston, & Pears, 2005).
The first randomized controlled trial (RCT) on the effectiveness of PMTO conducted
outside of the US was completed in Norway, in which 112 clinically referred boys and
girls aged between 4 and 12 years were randomly assigned to either PMTO or CAU
(Ogden & Amlund-Hagen, 2008). Results indicated that PMTO was superior to CAU on
the post-treatment outcome measures relating to effective discipline, obedience of the
child, child-initiated negative behaviors and externalizing behavior problems. The effect
of PMTO appeared to be moderated by the age of the child: that is, the intervention
proved to be more effective in children below 8 years of age than among older children
(Ogden & Amlund-Hagen, 2008). However, at one-year follow-up, the differences
between PMTO and CAU on child behavior problems and parental discipline were no
longer significant (Amlund-Hagen, Ogden, & Bjørnebekk, 2011). A highly similar RCT
was conducted in Iceland by Sigmarsdóttir, Thorlacius, Guŏmundsdóttir and DeGarmo
(2014), who also allocated clinically referred children with behavior problems aged 5
to 12 years (N = 102) to either PMTO or CAU. PMTO was found to be more effective
than CAU in improving general child adjustment at post-treatment, although the only
significant effect was documented for social skills. Surprisingly, this study did not obtain
17
1
Chapter 1 | General introduction
support for the idea that PMTO has a positive effect on parenting skills (Sigmarsdóttir,
DeGarmo, Forgatch, & Guðmundsdóttir, 2013).
Callous-unemotional traits
A subset of children with externalizing behavior problems is characterized by socalled callous-unemotional (CU) traits. CU traits resemble the emotional detachment
component of psychopathy in adult forensic samples (Frick, 2006; Frick, Cornell, Barry,
Bodin, & Dane, 2003a). Children with CU traits display low fear in combination with high
impulsivity, are not so much concerned with others’ feelings, and typically do not feel
bad or guilty when breaking rules. Furthermore, children who are characterized by CU
traits have an earlier onset of offending and higher levels of aggression, delinquency,
and police contacts compared with antisocial children without these traits (Andershed,
Gustafson, Kerr, & Stattin, 2002; Frick et al., 2003a; Frick, Stickle, Dandreaux, Farrell, &
Kimonis, 2005; Frick & White, 2008; Lawing, Frick, & Cruise, 2010). McMahon, Witkiewitz,
and Kotler (2010), for example, found that CU traits in young adolescents with
conduct disorder improved the prediction of antisocial outcomes, in terms of general
delinquency and police arrests, with very low false-positive rates. A number of studies
have demonstrated that CU traits are relatively stable over time, either assessed by selfreport or parent-report (Burke, Loeber, & Lahey, 2007; Dadds, Fraser, Frost, & Hawes, 2005;
Frick, Kimonis, Dandreaux, & Farell, 2003b; Obradovic, Pardini, Long, & Loeber, 2007).
Therefore, the presence of CU traits seems to designate a distinct group of children
with conduct disorder who show a more severe and chronic pattern of externalizing
behavior problems and delinquency.
As outlined above, studies demonstrate that improving parenting practices is the
most important mechanism contributing to the decrease of externalizing behavior in
children (DeBaryshe et al., 1993; DeGarmo & Forgatch, 2005; McCoy et al., 1999; Nix et al.,
1999). However, in children with CU traits, previous research has shown that ineffective
parenting seems to be unrelated to the frequency and severity of antisocial behavior
problems (Hipwell et al., 2007; Oxford, Cavell, & Hughes, 2003; Viding, Fontaine, Oliver,
& Plomin, 2009; Vitacco, Neumann, Ramos, & Roberts, 2003; Wootton, Frick, Shelton, &
Silverthorn, 1997). For example, Wootton and colleagues (1997) found that the relation
between ineffective parenting and antisocial behavior problems was moderated by
the presence of CU traits. Ineffective parenting was related to behavior problems only
for children without CU traits, while children with CU traits showed antisocial behavior
independent of the quality of parenting.
One of the main reasons why children with CU traits are less responsive to effective
parenting is that these children tend to be less sensitive to punishment (e.g., house arrest,
privilege removal), while previous studies found that effective discipline is the most
18
General introduction | Chapter 1
important parenting practice in reducing behavior problems in the child (DeBaryshe
et al., 1993; Nix et al., 1999; Ogden & Amlund-Hagen, 2008; Oliver, Guerin, & Coffman,
2009). According to the operant conditioning principle, children should learn to inhibit
undesirable behaviors in order to avoid the punishment (Dadds & Salmon, 2003).
However, children with CU traits have a specific deficit in processing negative stimuli
(Frick, 2006). For example, they have more difficulty in identifying fearful and sad facial
expressions (Blair, Colledge, Murray, & Mitchell, 2001; Stevens, Charman, & Blair, 2001;
Woodworth & Waschbusch, 2008) and are less physiologically responsive to distressing
and threatening pictures, films and words (Anastassiou-Hadjicharalambous & Warden,
2008; Blair, 1999; Kimonis, Frick, Fazekas, & Loney, 2006; Loney, Frick, Clements, Ellis, &
Kerlin, 2003). It appears that children with CU traits respond with insufficient anxiety
when parents confront them with their misbehaviors.
This deficit in processing negative emotional stimuli could also affect memory.
In general, emotional events are remembered better than neutral ones (Buchanan,
2007; LaBar & Cabeza, 2006). However, because children with CU traits have difficulties
processing negative emotional material, they might not benefit from an enhanced
memory for emotional events. This effect has already been demonstrated in adults
and adolescents with psychopathic traits (Christianson et al., 1996; Dolan & Fullam,
2005, 2010). Therefore, children with CU traits are less likely to learn from a negative
emotional experience. As they are less capable of remembering the experience of being
punished after their misbehavior, they are less inclined to inhibit their behavior in future
situations in order to avoid the punishment.
Aim and outline of the thesis
The aim of this thesis is to examine the effectiveness of PMTO in The Netherlands and
to identify possible risk factors for the effectiveness of parent management trainings in
general, and PMTO in particular. It consists of two parts. Part 1 concerns the effectiveness
of PMTO in The Netherlands in reducing externalizing problem behavior in children.
Part 2 concerns emotional memory in children with callous-unemotional traits.
PART I: PARENT MANAGEMENT TRAINING – OREGON MODEL
Improving effective parenting is seen as an important mechanism in reducing
externalizing behavior problems in children (McCart et al., 2006). PMTO had been
proven effective in America, Norway and Iceland (e.g., Forgatch et al., 2009; Ogden &
Amlund-Hagen, 2008; Sigmarsdóttir et al., 2014). However, this does not guarantee
that PMTO also works in the Netherlands. Therefore, the effectiveness of PMTO in
the Netherlands was examined. Families of clinically referred children (N = 146) with
19
1
Chapter 1 | General introduction
externalizing behavior problems aged between 4 and 11 years were assigned to either
PMTO or CAU. In Chapter 2, the results on behavior problems, parenting practices, and
parental psychopathology are presented.
Research has shown that there is a positive association between treatment
fidelity and the effectiveness of PMTO (Forgatch et al., 2005b). The aim of Chapter 3
is to examine whether this relation is also true for Dutch PMTO therapists. Hence, the
treatment outcomes of the 86 PMTO families from the effectiveness study as described
in Chapter 2 were related to the FIMP-scores of the PMTO therapists.
Research has revealed the crucial importance of adequate parenting for a healthy
social-emotional development of children. Thus, the need for valid assessment
of parenting quality cannot be overestimated, particularly in clinical settings. The
Caregiver Wish List (CWL; Hodges, 2002) is a new 53-item interview scheme with Likerttype scales for measuring parenting practices as defined by the SIL-model. Chapter 4
describes a validation study of the Dutch-language version of the CWL. We examined
the factor structure, reliability, and validity of this instrument in a sample of 348 parents
of children aged between 4 and 11 years.
PART II: CALLOUS-UNEMOTIONAL TRAITS
Children with callous-unemotional (CU) traits are less responsive to parenting than
children without these traits. Children with CU traits have a specific deficit in processing
negative stimuli, which make them respond with insufficient anxiety when confronted
with their misbehavior (Frick, 2006). Therefore, we examined emotional memory of
children with CU traits. In Chapter 5, we used a self-developed peripheral memory test
to test the memory for central and peripheral components of neutral and emotionally
negative pictures in 8-to-12-year-old non-clinical children.
The deficiency in emotional processing in children with CU traits could also target
incorrect remembering (i.e., false memories). The study described in Chapter 6
examined false memories in children with CU traits using neutral and negative word
lists of the Deese-Roediger-McDermott (DRM) paradigm (Deese, 1959; Roediger &
McDermott, 1995).
Finally, in Chapter 7 the findings of the studies in the preceding chapters are
summarized and discussed. Implications for clinical practice and recommendations for
future research are presented.
20
P a r e n t
Management
Training
- Oregon model
PART
I
The effectiveness of
Parent Management
Training-Oregon
model
in clinically referred
children with externalizing
behavior problems in
The Netherlands
2
Thijssen, J., Vink, G., Muris, P., & de Ruiter, C. (2015). The effectiveness of Parent
Management Training - Oregon Model in clinically referred children with externalizing
behavior problems in The Netherlands. Manuscript accepted with revisions
Chapter 2 | Effectiveness of PMTO
ABSTRACT
The present study examined the effectiveness of Parent Management Training
- Oregon model (PMTO) as a treatment for children with externalizing behavior
problems in The Netherlands. Clinically referred children (N = 146) aged 4 to 11 years
and their parents were included in this research of which 91 received PMTO and 55
Care As Usual (CAU). Families were assessed at four time points: at pretreatment, and
after 6, 12, and 18 months. Results showed that both PMTO and CAU were effective in
reducing child externalizing behavior, parenting stress and parental psychopathology,
with no significant differences between the two treatment conditions. PMTO and CAU
interventions also produced some improvements in self-reported parenting skills, but
not in observed parenting skills. Finally, the effect size of PMTO on parent-reported
externalizing behavior problems as found in the present study was comparable to that
found in previous studies evaluating PMTO as an intervention for this type of child
psychopathology.
24
Effectiveness of PMTO | Chapter 2
INTRODUCTION
Ineffective parenting is a well-established risk factor for the development of
externalizing behavior problems in children (DeBaryshe, Patterson, & Capaldi, 1993;
Forgatch, Bullock, & Patterson, 2004; McCoy, Frick, Loney, & Ellis, 1999; Nix et al., 1999;
Patterson, Forgatch, Yoerger, & Stoolmiller, 1998). The role of parenting in the emergence
and maintenance of problematic child behavior is cogently explicated in Patterson’s
Social Interaction Learning (SIL) model (Forgatch et al., 2004). Briefly, the SIL model
assumes that contextual factors, such as socio-economic disadvantage and parental
psychopathology, have a negative impact on parenting quality. Further, ineffective
rearing practices have a direct detrimental influence on the behavior of the child,
thereby hindering its healthy social-emotional development. More precisely, persistent
coercive parenting – which is characterized by hostility and holding power over
children via punitive or psychologically controlling means – can promote overt forms of
externalizing behavior problems, such as noncompliance, temper tantrums, and verbal
and physical aggression, which in turn are maintained by negative reinforcement of
the parents (Snyder & Patterson, 1995). When children become more negative in their
behavior, they are harder to discipline, which leads to parents using even more aversive
strategies (Patterson, 1986). In this way, families become entangled in a downward
spiral of negativity.
The antisocial behaviors acquired at home also tend to generalize to other social
settings, such as school and sporting clubs (DeBaryshe et al., 1993). Within the peer
group, the antisocial behavior can lead to rejection by normal, prosocial peers. In turn,
this can lead to associations with deviant peer groups (DeGarmo & Forgatch, 2005;
Patterson, 1986; Snyder, 2002) in which it pays off to show negative behaviors like lying,
stealing, and vandalism (Forgatch, Patterson, DeGarmo, & Beldavs, 2009). However,
parents also make a contribution to such deviant behavior by poor monitoring of
the whereabouts and behaviors of their children outside the home environment. It
enables youngsters to wander away from home and to engage in, for example, drug
use and criminality (Patterson, Forgatch, & DeGarmo, 2010). These antisocial behaviors
in childhood may take the form of an Oppositional Defiant Disorder (ODD) or Conduct
Disorder (CD; American Psychiatric Association, 2013), which have been shown to be
possible precursors of Antisocial Personality Disorder in adulthood (Loeber & Burke,
2011).
The fact that externalizing behavior problems in children can have significant
negative long-term consequences, underlines the importance of early intervention
programs. Many of these programs focus on the improvement of parenting practices
and there is indeed evidence showing that the enhancement of positive and more
effective parenting is an important mechanism that promotes children’s prosocial
25
2
Chapter 2 | Effectiveness of PMTO
behavior (DeGarmo & Forgatch, 2005, 2007; Forgatch & DeGarmo, 1999; Martinez &
Forgatch, 2001; Ogden & Amlund-Hagen, 2008). A good example of an intervention
that is based on the key principles of the SIL model is Parent Management Training Oregon model (PMTO). The program is especially developed for the parents of children
between 4 and 12 years of age showing the severe behavior problems associated with
ODD or CD and aims to teach parents how to reduce coercive parenting practices
and to replace these with five effective parenting practices: encouragement (i.e.,
stimulation of prosocial behaviors in the child by using scaffolding techniques and
positive reinforcement), effective discipline (i.e., consistent use of mild sanctions like
giving a time out), monitoring (i.e., knowing the child’s friends and keeping track of its
activities), problem solving (i.e., responding effectively to rule-breaking behaviors and
settling arguments with the child), and positive involvement (i.e., giving love and warm
attention and engaging in fun activities with the child; Forgatch & DeGarmo, 1999;
Forgatch, Patterson, & DeGarmo, 2005b).
Initial studies conducted in the Unites States (US) have demonstrated that PMTO is
an effective intervention for reducing externalizing child behavior problems (e.g., Bank,
Marlowe, Reid, & Patterson, 1991; Patterson, Chamberlain, & Reid, 1982). For instance,
in the study by Forgatch and DeGarmo (1999), 238 recently divorced mothers were
randomly assigned to PMTO or a no intervention control condition. After 12 months, it
was found that in the PMTO condition the effective parenting practices had significantly
improved compared to the control condition. At a long term follow-up, 9 years after the
PMTO intervention, there was still a significant difference between the boys in the PMTO
condition and the control group with the former showing lower levels of delinquency,
criminal activities, and convictions (Forgatch et al., 2009). Furthermore, PMTO has also
been shown to be effective in newly formed families consisting of biological mothers
and stepfathers: again, parenting practices improved and behavior problems of the child
decreased, as compared with newly formed families who did not receive an intervention
(Forgatch, DeGarmo, & Beldavs, 2005a). Finally, in foster families, researchers found a
success rate of permanent placements of 90% for PMTO versus 64% for Care As Usual
(CAU) at an assessment which took place at 24 months after the interventions. PMTO
was also significantly associated with reductions of stress for both the children and the
foster parents (Fisher, Burraston, & Pears, 2005).
The first randomized controlled trial (RCT) on the effectiveness of PMTO conducted
outside of the US was completed in Norway, in which 112 clinically referred boys and
girls aged between 4 and 12 years were randomly assigned to either PMTO or CAU
(Ogden & Amlund-Hagen, 2008). Results indicated that PMTO was superior to CAU on
the post-treatment outcome measures relating to effective discipline, obedience of the
child, child-initiated negative behaviors and externalizing behavior problems. The effect
26
Effectiveness of PMTO | Chapter 2
of PMTO appeared to be moderated by the age of the child: that is, the intervention
proved to be more effective in children below 8 years of age than among older children
(Ogden & Amlund-Hagen, 2008). Further, at a one-year follow-up, the differences
between PMTO and CAU on child behavior problems and parental discipline were no
longer significant (Amlund-Hagen, Ogden, & Bjørnebekk, 2011). A highly similar RCT
was conducted in Iceland by Sigmarsdóttir, Thorlacius, Guŏmundsdóttir and DeGarmo
(2014), who also allocated clinically referred children with behavior problems aged 5 to
12 years (N = 102) to either PMTO or CAU. PMTO was found to be more effective than CAU
in improving general child adjustment post-treatment, although the only significant
effect was documented for social skills. Surprisingly, this study did not obtain support
for the idea that PMTO would have a positive effect on parenting skills (Sigmarsdóttir,
DeGarmo, Forgatch, & Guðmundsdóttir, 2013).
The present study evaluated the effectiveness of PMTO in The Netherlands. Onehundred-and-forty-six families of clinically referred children with externalizing behavior
problems aged between 4 and 11 years were assigned to either PMTO or CAU. Effects of
PMTO and CAU were examined by means of measures of child externalizing behavior,
parenting skills, and parental stress and psychopathology, which were administered
at baseline, and three follow-up measurements after six, 12, and 18 months. Parents’
treatment satisfaction was also evaluated after 6, 12, and 18 months. In addition, effect
size and clinically significant change in children’s externalizing behavior problems
was examined and compared across both treatment conditions, and several possible
moderators of the effects produced by PMTO were explored (i.e., child, parent, and
family variables). The following hypotheses were tested: (1) PMTO will result in greater
improvements of children’s behavior problems, parenting skills, and parental stress
and psychopathology than CAU; (2) PMTO will be associated with higher treatment
satisfaction of parents and fewer dropouts as compared to CAU; and (3) PMTO will
show a larger effect size and greater proportion of clinically significant change than
CAU. With regard to moderator effects, predictions were less obvious, although it can
be hypothesized that PMTO is more effective in families displaying the characteristics
that are the target of this intervention (i.e., poor parenting skills) or that facilitate the
application of the newly acquired skills in daily life (e.g., higher educational level of
parent).
27
2
Chapter 2 | Effectiveness of PMTO
METHOD
Participants
Participants were 146 children and their parents, who were recruited at five child
service agencies across The Netherlands. Of these children, 104 (71.2%) were boys and
42 (28.8%) were girls. At baseline, the age of the children ranged between 4 and 11 years,
with a mean age of 7.13 years (SD = 1.75). Based on the Diagnostic Interview Schedule
for Children (DISC; Shaffer, Fisher, Lucas, Dulcan, & Schwab-Stone, 2000), 75.4% of the
children met the DSM diagnostic criteria for ADHD, 67.3% for ODD, and 11.6% for CD.
The mean age of the main caregiver was 37.39 years (SD = 8.09). The vast majority of
the main caregivers was female (90.5%), had the Dutch nationality (89.7%) and was
employed (76.0%). One-hundred-and-thirteen of the 146 main caregivers (77.4%) were
living with a partner, which in the majority of the cases (n = 106) was the other biological
parent of the child. Five of the main caregivers were the adoptive parent of the child.
Sixty-eight percent of the main caregivers had a college or university degree, 30% had
completed high school, while the remaining 2% had only finished elementary school.
To be included in the study, the child had to reveal a T-score of 60 or higher on
the externalizing subscale of the Child Behavior Checklist (CBCL; Achenbach, 1991;
Dutch version: Verhulst, van der Ende, & Koot, 1996) and the child had to be residing
at home with at least one biological or adoptive parent. Exclusion criteria were: severe
intellectual disability or psychopathology of the parent(s) as this that would interfere
with participation in treatment, sexual abuse in the family, and a child IQ lower than 70.
Procedure
Families were included in the period between June 2009 and January 2014. As soon
as families were referred to the child service agency, it was checked whether they met
the inclusion criteria for the study. Families who met the criteria received information
about the study and its procedure and were invited to participate. When parents agreed,
they were asked to give their written consent. The study protocol was approved by the
Medical Ethics Committee of Maastricht University Medical Centre.
Allocation to the treatment conditions (PMTO and CAU) was random at three of the
five child service agencies. At the other two agencies no randomization took place. After
the study had started, one child service agency decided it would no longer offer CAU
and thus only recruited families for the PMTO condition. The fifth agency was specifically
included in the study to compensate for the missing CAU families, but unfortunately
this agency was less successful in recruiting participants for the study. Eventually, this
resulted in unequal sample sizes for the two treatment conditions, with 94 families
receiving PMTO and 61 families receiving CAU. In the PMTO condition, 17 families (18%)
dropped out, of which two families never started. In the CAU condition, 10 families
28
Effectiveness of PMTO | Chapter 2
(16%) dropped out of which seven families never started. No data could be collected
for the families that never started. These families were discarded from the data analysis,
leaving a final sample of 91 PMTO families and 55 CAU families.
Enrollment (N = 155)
Allocated to PMTO (n = 94)
Allocated to CAU (n = 61)
Received PMTO (n = 91)
Received CAU (n = 55)
Received PMTO (n = 91)
Baseline (T0)Received CAU (n = 55)







IQ screening
DSM-IV diagnosis
Child behavior problems
Parenting skills
Parental stress and psychopathology
Parental stress and psychopathology
After 6 months (T1)




Child behavior problems
Parenting skills
Parental stress and psychopathology
Parent satisfaction
After 12 months (T2)




Child behavior problems
Parenting skills
Parental stress and psychopathology
Parent satisfaction
After 18 months (T3)




Child behavior problems
Parenting skills
Parental stress and psychopathology
Parent satisfaction
Figure 1 | Flowchart of the present study.
PMTO = Parent Management Training – Oregon model; CAU = Care As Usual.
29
2
Chapter 2 | Effectiveness of PMTO
Figure 1 gives an overview of the assessments that were carried out during the
course of this study. As can be seen, assessments were performed at four time points:
at baseline (T0), and at six-months (T1), 12-months (T2), and 18-months (T3) followup. IQ tests, interviews and video observations were mostly conducted at the child
service agency (with a few exceptions at the families’ home), while questionnaires
were administered to parents and teachers through a web-based system that could
be approached by a computer in the home or agency environment. The assessments
were conducted by trained research assistants who were not involved in the treatment
of the families. The parent questionnaires and interviews were completed by the main
caregiver, who was the parent who spent most time with the child. If present in the
child’s family, the second caregiver was also assessed during the video observations.
Participating families received a small financial compensation in the form of gift
vouchers for the three follow-up assessments (i.e., €10 at T1; €20 at T2; and €30 at T3).
The five child service agencies were independent therapy clinics for children.
Therapists involved in the present study were all licensed clinicians working at the
participating agencies. The researchers were not involved in the treatment of either
PMTO or CAU. Therapists providing PMTO within the context of this study were not
allowed to give CAU to families included in the study. In addition, CAU therapists were
not allowed to be trained or to be in training as a PMTO therapist.
Assessment
Descriptive characteristics
IQ measurement. To have an indication of the intellectual abilities of the children, a
standardized IQ test was administered. When children were younger than 6 at baseline,
the complete Wechsler Preschool and Primary Scale of Intelligence, 3rd edition (WPPSIIII; Wechsler, 2002) was used for this purpose. For older children, the short form of
the Wechsler Intelligence Scale for Children, 3rd edition (WISC-III; Wechsler, 1991) was
employed, which consists of two verbal (i.e., similarities and vocabulary) and two
performance (i.e., picture arrangement and block design) subtests. The Dutch versions
of the WPPSI-III and the WISC-III are both reliable and well-validated instruments
that were current at the time that the study was conducted (Hamilton & Burns, 2003;
Kaufman, Kaufman, Balgopal, & McLean, 1996; Minshew, Turner, & Goldstein, 2005).
DSM-IV diagnosis. The Diagnostic Interview Schedule for Children (DISC-IV; Shaffer et
al., 2000) is a highly standardized interview schedule to identify the presence of DSM-IV
diagnoses in children. For the present study, the Disruptive Disorder module (module
E) was used, which assesses three disorders: Attention Deficit Hyperactivity Disorder
(ADHD), ODD, and CD. The interview was administered and scored by trained clinicians.
30
Effectiveness of PMTO | Chapter 2
Treatment outcome assessment
Child behavior problems. The Child Behavior Checklist (CBCL) and Teacher Report Form
(TRF) are widely used rating scales for assessing behavioral and emotional problems of
children aged 6 to 18 years (Achenbach, 1991; Dutch version: Verhulst et al., 1996). Each
scale consists of 120 items that are scored on a 3-point Likert scale (0 = not true; 1 =
somewhat or sometimes true; 2 = very or often true). Items can be allocated to narrowband or DSM-based scales which represent specific types of problems (e.g., somatic
complaints, social problems, affective problems, and conduct problems), or to the more
general broad-band scales of internalizing (emotional) and externalizing (behavioral)
problems, which can also be combined to a total score. For children aged 4 or 5 years, the
1.5–5 year version of the CBCL was used. The CBCL and TRF are widely used instruments
that have been demonstrated to possess good reliability and validity (Achenbach,
1991; Verhulst et al., 1996). In the present study, the internalizing, externalizing, and
total problems scores were used, which in the current study at baseline had Cronbach’s
alphas of respectively .83, .84, and .90 (CBCL) and .84, .93, and .96 (TRF).
The Parent Daily Report (PDR) is a reliable 34-item checklist covering the range of
mild (e.g., complaining) to severe (e.g., stealing) problem behaviors (Patterson et al.,
1982). The checklist is first administered face-to-face with the main caregiver to assess
which specific problem behaviors of the child have occurred during the past six months.
Next, these items are administered again via telephone on three consecutive days to
examine whether these behaviors have occurred during the past 24 hours. The number
of symptoms endorsed on the three consecutive days is summed to obtain the total
PDR score. The internal consistency coefficient of the PDR at baseline was .80.
Parenting practices. The Dutch translation of the Caregiver Wish List (CWL; Hodges,
2005; Hodges, de Ruiter, & Thijssen, 2009) is an interview-based instrument consisting
of 53 items questioning the parent about his/her parenting skills. The interviewer
reads the questions to the parent, who has to indicate the most applicable response
option using a 5-point Likert scale. Items are allocated to six domains of parenting
skills: providing direction and following up (4 items), encouraging good behavior (5
items), discouraging undesirable behavior (24 items), monitoring activities (13 items),
connecting positively with child (3 items), and problem solving orientation (4 items).
Each domain score should be regarded as a dimension with weak parenting skills on
one end and strong parenting skills on the other. In the current study, the reliability
of the CWL was not particularly strong: Cronbach’s alphas of various subscales were
.22 for providing direction and following up, .44 for encouraging good behavior, .80
for discouraging undesirable behavior, .50 for monitoring activities, .55 for connecting
positively with child, .37 for problem solving orientation, and .76 for the CWL total score.
31
2
Chapter 2 | Effectiveness of PMTO
Structured Interaction Tasks (SITs) were used to observe the parenting behaviors
during a series of seven structured tasks that had to be performed by parents and
their child: planning a fun activity for the weekend (3 min.), problem solving, in which
the family members discuss a topic chosen by the parent where they regularly have
arguments about (5 min.), drawing a picture of their own house (7 min.), a snack-break
to observe the family members when they do not have an assignment (5 min.), problem
solving for a topic chosen by the child (5 min.), teaching tasks, which consist of two
homework-like assignments of which one is slightly difficult for the child’s age and/or
intellectual ability to evoke frustration (9 min.), and monitoring, in which parents had
to interview their child about a moment when there was no supervision by an adult (5
min.). The SITs are designed to elicit parenting practices. The second problem solving
task and the teaching tasks are performed with the main caregiver. When present, the
drawing of a house is performed with only the second caregiver.
The tasks were videotaped and later coded by three trained, independent raters
(psychology Master’s students) using an adapted version of the Coder Impressions
developed by researchers at Oregon Social Learning Center (Forgatch, Knutson, &
Mayne, 1992). Briefly, the videotaped SITs were employed to score a number of items
referring to domains of effective parenting behaviors of positive involvement (12
items), encouragement (20 items), problem solving (27 items), discipline (25 items),
and monitoring (5 items) as well as coercion (16 items), child compliance (8 items), and
interpersonal atmosphere (24 items). For each SIT domain, item scores are accumulated,
with higher scores reflecting better parenting skills. Cronbach’s alphas were poor for
monitoring (mothers .28; fathers .06) and coercion (mothers .62; fathers .48). The other
parenting dimensions had good internal consistency (range .74 - .91).
All videotapes were coded by three trained and calibrated psychology graduate
students blind to treatment condition and assessment point (i.e., T0, T1, T2, or T3). Two
independent raters coded a random selection of 103 SITs (29.4% of the coded SITs).
Interrater reliability was examined by means of a two-way mixed, consistency, averagemeasures Intraclass Correlation Coefficient (ICC; McGraw & Wong, 1996). The ICC’s for
the parenting dimensions were fair to excellent, with the exception of ICC’s for discipline
and compliance, which were poor.
Parental stress and psychopathology. The Nijmeegse Ouderlijke Stress Index (NOSI; de
Brock, Vermulst, Gerris, & Abidin, 1992) is an adaptation of the Parenting Stress Index
(Abidin, 1983) and measures stress experienced by parents in the relationship with their
child. The NOSI comprises 123 items that have to be rated on a 5-point Likert scale (1
= strongly disagree, 5 = strongly agree). The items refer to parent (e.g., competence,
social isolation, health, relationship with spouse) as well as child (e.g., hyperactivity,
demandingness, mood) characteristics. Ratings on all items can be summed to create
32
Effectiveness of PMTO | Chapter 2
a total stress score, with higher scores reflecting higher levels of perceived stress by
the parent. The NOSI has adequate reliability and validity (de Brock et al., 1992). In
the present study, Cronbach’s alphas were .93 for the child domain, .94 for the parent
domain, and .96 for the total score.
Psychological symptoms of the parents were measured by the Dutch version of
the Symptom Checklist-90 Revised (SCL-90-R; Arrindell & Ettema, 1986). This version
of the questionnaire is based on the SCL- 90-R of Derogatis (1977). The 90 items are
rated using a 5-point scale (1 = no problem to 5 = very serious) to indicate the extent
to which the parent has experienced the listed symptom during the previous week.
In contrast to the original version of the SCL-90-R, the Dutch version comprises eight
instead of nine subscales: Anxiety, Agoraphobia (in original version: Phobic Anxiety),
Somatic Symptoms, Depression, Inadequacy of Thinking and Acting (in original version:
Obsessive-Compulsive), Distrust and Interpersonal Sensitivity, Hostility, and Sleeping
Problems. In the Dutch version of the SCL-90-R, the subscales Interpersonal sensitivity
and Paranoid ideation (and three items from the original Psychoticism subscale) are
combined into Distrust and Interpersonal Sensitivity due to insufficient discrimination
between these dimensions (Arrindell & Ettema, 1986). Higher scores on the SCL-90-R
indicate more serious psychopathology. In the present study, only the total score was
used, which had an internal consistency coefficient of .97 at baseline.
Parent satisfaction. The short form of the Working Alliance Inventory (WAI-S; Tracey
& Kokotovic, 1989) was used to assess the quality of the parent-therapist alliance. The
WAI-S comprises 12 items that can be allocated to three subscales of four items each:
(a) agreement between parent and therapist on the goals of the therapy; (b) agreement
that the tasks of the therapy will address the parent’s problems, and (c) the quality of
the bond between the parent and the therapist. Normally, the items of the WAI-S are
rated using a 7-point Likert scale. However, in the present study, a 5-point Likert scale
was used for practical reasons. Ten items are positively worded and two items (items 4
and 10) are negatively worded. The scores on the negatively stated items are recoded,
so that all scores can be summed to obtain a total score. Higher total scores indicate a
better parent-therapist working alliance. In the present study, Cronbach’s alpha for the
total score was .71.
Interventions
PMTO. PMTO is a therapy consisting of weekly sessions in which the therapist works
with the parent(s) of one family. The children are not present during these sessions. The
PMTO therapy aims to replace parents’ coercive parenting practices by the five effective
parenting practices as defined by the SIL model. Role play is an important mechanism in
the PMTO sessions to teach and extensively practice these effective parenting skills. The
33
2
Chapter 2 | Effectiveness of PMTO
therapist uses the role play to demonstrate good and bad examples and to determine
which parenting skills need extra attention (Forgatch et al., 2004). As soon as the parent
has sufficiently practiced the parenting skill, the therapist and parent choose a specific
situation for the next week during which the parent will try to apply the newly acquired
skill at home. In between sessions, the therapist calls the parent for support and to answer
questions. Usually, the therapy takes place at the agency, but occasionally sessions are
given at the parents’ home as well. Treatment duration depends on the family’s needs
and progress throughout the therapy, but typically takes between 15 and 25 weekly
sessions. Parents in the present study received PMTO from 25 certified therapists. All
therapists had completed the full PMTO training program of approximately 24 months.
During this training period, therapists had to treat at least three families with PMTO
before they were allowed to take part in the official PMTO certification procedure. This
procedure involved treating another family with PMTO. On the basis of videotaped
sessions of this therapy, it was determined whether or not the therapists received their
license to carry out PMTO in clinical practice. Following the completion of the training
program, therapists were regularly monitored on their treatment fidelity, leading to
annual recertification of their license. The association between treatment integrity and
treatment outcome is addressed in a separate study (Thijssen, Albrecht, Muris, & De
Ruiter, 2015). Of the children for which medication use was documented, 14.8 % used
additional ADHD medication.
CAU. CAU-treatments were treatments that were available at the child service
agencies for children with externalizing behavior problems and included family
therapy (n = 31), psychiatric intensive home care (n = 10), parent therapy (n = 9), or
other treatments (n = 6). In 9 CAU-families (17.6%), children received ADHD medication
in combination with one of the mentioned therapies. Two families in the CAU condition
received more than one treatment, which explains the higher number of CAU therapies
than CAU families.
Missing data
The percentage of missing values in the dataset ranged from 0 % for the demographic
variables to approximately 50 % for scores on the TRF. For many variables the missing
values could be considered as bonafide because no score can be observed if the variable
is not applicable (e.g., no observations for the second caregiver when only the main
caregiver was present).
We applied an intention-to-treat design by conducting multiple imputation (Rubin,
1987) to handle the missing data through a chain of conditional regression models (fully
conditional specification; Van Buuren, Boshuizen, & Knook, 1999). We used predictive
mean matching (PMM; Little, 1988; Rubin, 1986) for the scale variables, a custom
version of PMM for scale variables that contain bonafide missings, logistic regression
34
Effectiveness of PMTO | Chapter 2
for dichotomous variables and polytomous regression for ordered categorical data.
All computations were carried out with Mice (Van Buuren & Groothuis-Oudshoorn,
2011) in R (R Core Team, 2014), with 150 iterations for the algorithm to converge and
25 multiply imputed datasets, using available and custom imputation routines in Mice.
The outcomes over the 25 datasets were combined into a single inference using Rubin’s
rules (Rubin, 1987, p. 76).
Analytic strategy
Overall effectiveness was examined for all outcome measures by using repeated
measures ANOVA, with time (outcome measures at T0 through T3) as the within subjects
factor and treatment condition (PMTO vs. CAU) as the between subjects factor. Because
previous research found that child age and gender were significantly associated with
treatment outcome (Ogden & Amlund-Hagen, 2008; Sigmarsdóttir et al., 2013), we
checked whether these variables had any influence on the outcome measures in our
study. When this appeared to be the case, we controlled for the pertinent variable by
performing an ANCOVA.
Furthermore, clinically significant change in externalizing behavior problems was
examined using the Jacobson-Truax Reliable Change Index (RCI), since this is the most
widely-used and recommended method (Bauer, Lambert, & Nielsen, 2004; Jacobson,
Roberts, Berns, & McGlinchey, 1999). This method consists of two steps. First, a cutoff
point needs to be established to determine whether the child has moved from
the dysfunctional to the functional range. The second step is to calculate the RCI to
determine if the child’s change from pretreatment to follow-up is not the result of
measurement error. When these two criteria are combined (cutoff and RCI), the children
can be classified as Recovered (i.e., passed both criteria), Improved (i.e., passed RCI
criterion but not the cutoff ), Unchanged (i.e., passed neither criterion), or Deteriorated
(i.e., passed RCI criterion but worsened; Jacobson et al., 1999).
RESULTS
Baseline comparisons
Families in the PMTO and CAU condition did not differ significantly on any of the
demographic variables. It should be noted, however, that there was a trend towards
significance for ADHD, with fewer children having this diagnosis in the PMTO condition
[χ2(3) = 7.33, p = .06]. Then, we checked whether there were differences in outcome
measures between the PMTO and CAU group at the baseline assessment. Only one
significant difference was found: children from the CAU group displayed higher levels of
behavioral problems on the PDR than children in the PMTO group [t(119) = 2.28, p = .05].
35
2
Chapter 2 | Effectiveness of PMTO
Treatment attendance
The number of sessions families received was better documented for PMTO than
for CAU. Reports on treatment attendance were available for 61 PMTO families and 18
CAU families, and these showed that PMTO families received more treatment sessions
than CAU families. Families in the PMTO condition received on average 23.85 (SD = 9.86)
treatment sessions, while families in the CAU condition received a mean of 20.50 (SD =
10.67) sessions. This difference was not significant [t(77) = 1.24, p = .22], and therefore
we did not have to control for treatment attendance in our effect analyses.
Effects of PMTO versus CAU
Child behavior problems
Mean scores (and standard deviations) of children in the two treatment conditions
on various CBCL and TRF scales and the PDR are shown in Table 1. First, we compared
the effectiveness of PMTO and CAU on externalizing behavior problems because this
was the main outcome variable. For CBCL externalizing and the PDR, a significant main
effect of time was found [F(3, 68.87) = 14.75, p < .001 and F(1.99, 315.02) = 13.17, p
< .001, respectively]: post-hoc comparisons with Bonferroni correction showed that in
both conditions, symptom levels significantly decreased from T0 to T1 after which they
remained fairly stable at T2 and T3. No interaction effects of treatment condition and
time were found, indicating that there were no significant differences in effectiveness
between PMTO and CAU on these outcome measures [F(3, 112.49) < 1 and F(1.99,
897.57) = 2.37, p = .09, respectively]. For TRF externalizing, neither a main effect of time
[F(3, 33.87) = 1.79, p = .17] nor an interaction effect of treatment conditions and time
could be documented [F(2.66, 182.89) < 1].
Second, treatment effects on parent and teacher rated internalizing and total
problems were analyzed. The pattern of results resembled that found for the externalizing
behavior problems. That is, for CBCL internalizing as well as total behavior problems,
a significant main effect of time was found [F(2.62, 223) = 14.14, p < .001 and F(2.82,
74.87) = 15.10, p < .001, respectively]. Pairwise comparisons using Bonferroni correction
again only revealed a significant decrease in behavior problems between T0 and T1.
No interaction effects of treatment condition and time were found [F(2.62, 348.94) < 1
and F(2.82, 126.29) = 1.33, p = .27, respectively]. For TRF internalizing and total behavior
problems, neither significant main effects nor interaction effects were found.
36
PDR
53.75 (16.54)
58.61 (17.42)
14.14 (11.79) b
60.43 (9.40)
14.92 (15.87) b
58.81 (19.08)
55.31 (20.06)
12.77 (14.51) b
62.05 (15.48)
58.88 (14.54)
58.01 (10.51)
21.92 (12.13)
63.32 (10.79)
62.32 (11.59)
T2
M (SD)
T3
M (SD)
16.00 (16.77)
61.75 (10.35)
60.01 (12.80)
56.76 (12.21)
15.53 (13.52)
59.76 (16.01)
58.52 (19.83)
53.19 (18.41)
15.52 (19.21)
62.87 (15.31)
60.23 (15.93)
56.58 (18.21)
65.80 (11.61) b 64.42 (11.83) b 64.32 (11.00) b
65.80 (10.90) b 65.35 (11.18) b 66.57 (10.46) b
57.62 (11.69) b 59.81 (14.22) b 57.91 (13.21) b
T1
M (SD)
CAU
13.17*
2.12
1.79
1.76
15.10*
14.75*
14.14*
Main
effect
F
2.37
.32
.53
.49
1.33
.61
.48
Time X
Treatment
F
Note. T-scores are presented for the internalizing, externalizing and total scales of the Child Behavior Checklist (CBCL) and the Teacher Report Form (TRF). Means
with different subscripts indicate significant difference at p < .05 (Bonferroni corrected). PDR = Parent Daily Report.
* p < .05
62.13 (10.00)
17.64 (14.57) a
TRF Total
60.38 (12.66)
56.56 (12.83)
56.95 (10.74)
62.92 (10.27)
TRF Int
TRF Ext
61.06 (13.10) b 60.97 (13.77) b
69.32 (6.18) a
68.60 (6.71) a
CBCL Total
71.23 (6.53) a
70.90 (6.32) a
CBCL Ext
63.07 (9.59) b
62.66 (9.38) a
65.32 (10.70) b 62.99 (13.60) b 63.90 (12.93) b
T0
M (SD)
63.62 (9.01) a
T3
M (SD)
57.86 (13.05) b
T2
M (SD)
58.39 (11.93) b 58.71 (17.19) b
T1
M (SD)
CBCL Int
T0
M (SD)
PMTO
Table 1 | Means, standard deviations, and treatment effects for child behavior problems
Effectiveness of PMTO | Chapter 2
2
37
Chapter 2 | Effectiveness of PMTO
Parenting practices
Parenting practices were assessed using self-report (CWL) and structured
observations (SIT). Since only the CWL subscale ‘discouraging undesirable behavior’ and
the total score on this self-report measure displayed acceptable internal consistencies,
only these scores were used in subsequent analyses. For discouraging undesirable
behavior, a significant main effect of time was found. Parents in both conditions reported
a significant increase in their employment of discouragement of undesirable behavior
during the first six months [F(2.79, 53.21) = 7.21, p < .001]. For the CWL total score, also
a significant main effect of time emerged [F(2.77, 79.07) = 11.05, p < .001] whereas the
interaction of treatment condition and time did not attain significance [F(2.77, 978.85) =
2.15, p = .10]. Thus, in general (i.e., irrespective of treatment condition, parents reported
to use more effective parenting skills over time during the first 12 months.
Using the SIT data in relation to the main caregiver, no significant main effects of time
or interaction effects of treatment condition and time were found. For the second parent,
a significant Time x Condition interaction was found for interpersonal atmosphere [F(3,
6942.45) = 2.74, p = .04]. Pairwise comparisons using Bonferroni correction showed
that this interaction was apparent between T2 and T3: The second parent in the PMTO
condition showed an increase in positive interpersonal atmosphere between T2 and T3,
while the second parent in the CAU condition showed a decrease. No significant effect
was found for child compliance [F(3, 18) = 1.47, p = .26], which was probably due to the
low base rate of child problem behavior during the SITs. The mean number of coded
problem behaviors during the seven tasks at baseline (T0) was .90 (SD = 1.54), indicating
that on average children showed problem behavior in fewer than one of the seven SITs.
Parental stress and psychopathology
The mean scores (and standard deviations) of the NOSI and SCL-90-R are presented
in Table 2. To assess treatment effects on parenting stress, the NOSI was used. On the
parent domain, child domain, and the total score of the NOSI, a significant main effect of
time was found [F(2.42, 45.73) = 9.60, p < .001, F(2.81, 64.14) = 12.82, p < .001, and F(2.55,
79.5) = 21.37, p < .001, respectively]. Post-hoc comparisons with Bonferroni correction
showed that in both treatment conditions, parenting stress significantly decreased
from T0 to T1 and then remained stable at T2 and T3. No interaction effects of treatment
condition and time were found, which indicates there were no significant differences in
effectiveness between PMTO and CAU on parenting stress.
For psychopathological complaints, as measured by the SCL-90-R, there was a
significant main effect of time [F(2.26, 145.71) = 9.71, p < .001]. Pairwise comparisons,
using Bonferroni correction, showed that for both treatment conditions the level of
psychopathology significantly decreased from T0 to T1 after which no further change
was observed. Again, no significant interaction between treatment condition and time
was found [F(2.26, 259.43) < 1].
38
111.42 (35.22) b
125.86 (41.96) a 109.66 (31.85) b
117.87 (56.27) b
184.33 (75.17) b
113.04 (36.25) b
9.71*
185.75 (70.20) b 12.82*
9.60*
.63
1.72
.83
1.03
Main
Time X
effect Treatment
F
F
Note. Means with different subscripts indicate significant difference at p < .05 (Bonferroni corrected). NOSI = Nijmeegse Ouderlijke Stress Index (Parenting Stress
Index); SCL-90-R = Symptom Checklist-90 Revised.
* p < .05
SCL-90-R 129.81 (46.29) a 115.73 (34.89) b 118.82 (64.35) b
207.41 (50.92) a 183.19 (65.87) b
127.24 (53.43) b
T3
M (SD)
358.46 (82.68) a 300.50 (104.72) b 287.22 (107.35) b 283.89 (132.06) b 353.32 (95.63) a 310.31 (112.58) b 310.08 (113.89) b 313.08 (118.80) b 21.37*
168.65 (66.32) b
135.35 (60.44) b
T2
M (SD)
NOSI
total
stress
T1
M (SD)
148.21 (49.84) a 128.56 (46.95) b
T0
M (SD)
207.94 (45.69) a 176.85 (64.05) b 173.67 (91.32) b
115.97 (51.19) b
T3
M (SD)
NOSI
child
domain
T2
M (SD)
148.93 (45.42) a 123.12 (47.43) b 125.74 (54.22) b
T1
M (SD)
CAU
NOSI
parent
domain
T0
M (SD)
PMTO
Table 2 | Means, standard deviations, and treatment effects for parental stress and psychopathology
Effectiveness of PMTO | Chapter 2
2
39
Chapter 2 | Effectiveness of PMTO
Parent satisfaction
To examine whether there was a difference in treatment satisfaction between PMTO
and CAU, the scores of the WAI-S at T1 were analyzed using an independent samples
t-test. Results showed that the difference between PMTO (M = 43.95, SD = 4.67) and CAU
(M = 42.94, SD = 4.75) was not significant [t(142) = 1.05, p = .16].
Effect size and clinical significance
Cohen’s effect size d was calculated for the main outcome variable of the present
study, i.e., parent-rated child externalizing behavior problems, in order to compare our
findings to those of previous European studies.1 As can be seen in Table 3, the effect
size of PMTO in the present study was comparable to the effect sizes for PMTO in the
Norwegian studies and even somewhat higher than the effect size for PMTO as found
in the investigation conducted in Iceland. With the exception of the effect size found for
aggressive behavior for the control condition in the Norwegian follow-up study (Cohen’s
d = .63), the effect size for CAU in the present study (Cohen’s d = .55) was generally
higher than that obtained for the treatment control conditions in the Norwegian and
Icelandic studies (Cohen’s d’s between .22 and .43).
Our analyses indicate that both PMTO and CAU produced statistically significant
decreases of externalizing problem behavior. However, to examine whether the
improvement in child behavior was also clinically significant at the individual level, the
RCI was calculated for both the PMTO and CAU condition using the CBCL externalizing
behavior problems scores at T0 and T3. Based on the Jacobson-Truax method (Jacobson
et al., 1999), children could be classified as recovered, improved, unchanged, or
deteriorated. The percentages of children in each category for PMTO and CAU are
presented in Table 4. In total, 45.8% of children in the PMTO condition improved
against 42.8% of the children in the CAU condition. In the PMTO group, 16.9% of the
children recovered compared with 9.7% in the CAU group. The distribution over the
four categories did not differ significantly between PMTO and CAU [χ2(3) = 1.60, p = .66].
1
40
We limited our comparison to European studies because these studies were similar to our study in terms
of design, sample, and culture.
The Netherlands
Present study
Iceland
Norway
CBCL ext
CBCL ext
Outcome measure
Post treatment
CBCL ext
1 year after post treatment CBCL agg
CBCL del
Post treatment
18 months after baseline
Follow-up assessment
.47
.85
.70
.73
.73
Cohen’s d PMTO
Note. CBCL = Child Behavior Checklist, ext = externalizing, agg = aggressive, del = delinquent
a
Based on intention-to-treat (ITT) analyses. bBased on raw scores not presented in the original paper, but requested from the authors.
Sigmarsdóttir et al. (2014)b
a
Amlund-Hagen et al. (2011)
Ogden & Amlund-Hagen (2008) Norway
Country
Study
Table 3 | Cohen’s d in four European PMTO effectiveness studies for parent reported externalizing behavior problems
.39
.63
.22
.43
.55
Cohen’s d control group
Effectiveness of PMTO | Chapter 2
2
41
Chapter 2 | Effectiveness of PMTO
Table 4 | Percentages of reliable change based on parent rated externalizing behavior problems
PMTO
CAU
Recovered
16.9
9.7
Improved
45.8
42.8
Unchanged
25.5
33.6
Deteriorated
11.4
14.1
Moderators
To examine whether PMTO works better for certain families than for others, we
tested whether factors could be identified that moderate the effect of PMTO on the
main outcome variable (i.e., CBCL externalizing). Children who were classified as
recovered and improved based on the RCI (n = 58) were compared with children who
were classified as unchanged or deteriorated (n = 34). We examined child variables
(i.e., severity of problem behavior at baseline, IQ), parent variables (i.e., age, ethnicity,
education level, level of parenting skills and parenting stress at baseline, job status),
and family variables (i.e., single parent household, number of siblings). Only one
significant result was found for the CBCL externalizing subscale at T0 [t(333) = 2.41, p
< .001]. Children who showed reliable improvement in parent rated child externalizing
behavior problems showed significantly more severe externalizing behavior problems
at baseline (M = 72.19) as compared to children who did not show reliable improvement
(M = 68.69).
DISCUSSION
The present study compared the effectiveness of PMTO and CAU in Dutch children
who had been referred to child care organizations because of externalizing behavior
problems. It was hypothesized that children in the PMTO condition would show a
greater reduction in externalizing behavior problems relative to children in the CAU
condition. Furthermore, parents who received PMTO were expected to show greater
improvements in effective parenting skills in comparison to parents who received CAU.
Finally, we predicted that parents in the PMTO condition would show less parenting
stress and psychological complaints than parents in the CAU condition.
In contrast with our expectations, the results revealed no statistically significant
differences in effectiveness between PMTO and CAU on the primary treatment outcome
measures of parent-reported externalizing behaviors. That is, children in both conditions
showed a significant decrease in CBCL externalizing and PDR scores within the first 6
months of treatment, after which symptom levels remained fairly stable. For parentrated internalizing and total behavior problems, a similar pattern was found: in both
42
Effectiveness of PMTO | Chapter 2
treatment conditions significant decreases were found during the first six months, but
no evidence was obtained that children in the PMTO condition fared better than those
who received CAU. The fact that internalizing symptoms were also reduced following
interventions which essentially target externalizing problems, suggests that either
non-specific treatment factors were at work or that both interventions were capable
of tackling a process underlying both types of problems. No effects were found for
teacher-reported behavior problems. One explanation for this unexpected result might
be that children’s behavior problems are less apparent at school and that, therefore,
change was less noticeable. Indeed, the data indicated that teachers in general reported
less problem behavior as compared to parents. Alternatively, it is also possible that
the positive treatment effects did not generalize to the school setting and that the
interventions are only effective in the context where they have been implemented (i.e.,
at home).
The finding that PMTO did not result in a greater decrease of externalizing behavior
problems than CAU, is in contrast with the results of previous studies showing a
superiority of PMTO over control interventions (e.g., DeGarmo & Forgatch, 2005;
Forgatch & DeGarmo, 1999; Forgatch et al., 2005a). However, it is important to note that
most of the earlier studies that have been conducted in the United States compared
PMTO to a waiting list control condition. The families included in the control condition
of our study also received a proper treatment, which turned out to be rather effective in
reducing children’s externalizing problems. Our findings seem to be more in line with
the results of two PMTO effectiveness studies conducted in Norway and Iceland, which
also included a control group that received an alternative treatment (Ogden & AmlundHagen, 2008; Sigmarsdóttir et al., 2014). The Norwegian study demonstrated that PMTO
initially resulted in a larger decrease in problem behaviors than CAU, but also found
that this difference was no longer significant at one-year follow-up (Amlund-Hagen et
al., 2011). In the Icelandic study, PMTO produced a better treatment effect than CAU
on children’s social skills, but not on behavior problems (Sigmarsdóttir et al., 2014). It
is noteworthy that the effect size of CAU in our study was generally larger than that
obtained in the other studies, which indicates that the general treatment offerings for
children with externalizing problems in The Netherlands appears to be of good quality.
This probably is a result of the fact that many psychologists in this country are trained
to apply cognitive-behavioral techniques, which seem to be an important ingredient of
effective interventions for children with externalizing problems (Greene, Ablon, Goring,
Fazio, & Morse, 2004). In addition, PMTO is not the only treatment for externalizing
behavior problems in The Netherlands that was not more effective than CAU (e.g., Triple
P; Kleefman, Jansen, Stewart, & Reijneveld, 2014; Spijkers, Jansen, & Reijneveld, 2013).
43
2
Chapter 2 | Effectiveness of PMTO
Contrary to our expectations, no significant differences between PMTO and CAU
were found with regard to the application of effective parenting skills. Only three
significant findings on parenting skills emerged. The first one was that parents in
both conditions reported a significant increase in self-reported discouragement of
undesirable behaviors over time. This suggests that parents in general became more
responsive to the misbehaviors of their child. Second, parents reported an increase in
their overall use of self-reported effective parenting practices over time. Third, when
analyzing the behavioral observation data on parenting behavior, a difference between
PMTO and CAU was found for interpersonal atmosphere of the second caregiver. The
second caregiver who had received PMTO demonstrated a more positive interpersonal
atmosphere over time as compared to the second caregiver who had received CAU.
PMTO, as derived from the SIL model, assumes that the reduction of problematic
child behavior is mediated by improvements in parenting skills. In particular effective
discipline is thought to be an important target mechanism involved in the elimination
of child externalizing problems (Ogden & Amlund-Hagen, 2008; Patterson, 2005). Note,
however, that this could only be demonstrated with the self-report measure in our
study, and this may be due to several reasons. First of all, the observational tasks we
used did not elicit particularly high levels of negative behaviors in the child, so parents
hardly had to discipline their child during these assessments. Even at baseline, when
children were expected to show clear signs of externalizing behavior, the frequency
of such problems was less than one out of the seven observation tasks. A second
explanation concerns the (un)reliability of the observations. It should be noted that not
all parenting scales had satisfactory inter-rater reliability (e.g., discipline). Further, one
could argue that the SITs were too well-structured for the oppositional-defiant behavior
of the child and the accompanying parenting responses to emerge, which of course
questions the ecological validity of our observation measure. Still, it eludes us why
our children ‘behaved so well’ during the tasks, because we used tasks very similar to
the ones used in the original studies (e.g., Forgatch & DeGarmo, 1999; Forgatch et al.,
2005a). One difference is that our SITs were typically administered in a plain room with
few distractors, while in the original studies toys and other distractors were available
and present in the room. Similar points of critique can be raised regarding the selfreport measure of parenting skills. The internal consistency of five out of six subscales
of the CWL was unsatisfactory, and there are data that seriously question the validity of
this measure (Thijssen, Muris, & de Ruiter, 2015). Nevertheless, the two reliable scales of
the CWL (discouraging undesirable behavior and CWL total score) did show a positive
treatment effect.
In both conditions, significant reductions of parenting stress and parental
psychopathology within the first six months were found, with no significant differences
44
Effectiveness of PMTO | Chapter 2
observed between PMTO and CAU. These results indicate that parents generally felt
better as a result of both types of treatment. Apparently, the improvements in their
child’s behavior make parents feel less stressed during daily interactions with their child,
which may well translate into an overall improved sense of well-being, although the
direction of this effect may also be reversed: receiving treatment may boost parental
self-efficacy and well-being, which in turn has a positive impact on children’s behavior
(e.g., DeGarmo, Patterson, & Forgatch, 2004; Sigmarsdóttir et al., 2013).
Not all children profited equally from the PMTO and CAU interventions. A detailed
analysis (combining reliable change and clinical cut-off ) indicated that 17% of the
children within the PMTO group recovered and 46% showed reliable improvement
in externalizing behavior. In comparison, in the CAU condition 10% of the children
recovered and 43% reliably improved. To determine if some children benefited
more from PMTO than others, several possible moderators were examined. Only one
moderator effect was found: children who improved or recovered had significantly
higher parent-rated externalizing behavior problems at baseline as compared to
children who did not change or worsened. Thus, especially children with serious
externalizing behavior problems appeared to benefit more from PMTO. This result is
probably due to the fact that there was simply more room for improvement for these
children. Possibly, more moderator effects would have been found when using only the
recovered and deteriorated children in the comparison. However, in the present study,
these subgroups were too small to conduct such analyses.
A number of limitations of the present study should be mentioned. First, although
the study was originally designed as a RCT, due to practical constraints, we had to
continue as a quasi-experimental investigation about halfway through the study.
This also resulted in an unequal number of families in the PMTO and CAU conditions.
Second, we did not have information about the actual number of treatment sessions
that families in both conditions received, and therefore we were not able to control for
treatment exposure. Third, as described above, the assessment of parenting practices
appeared to be quite problematic, and this appeared true for both the self-report
measure (CWL) and the observations (SITs). With regard to the observational index,
an additional shortcoming was that coders not always remained blind to treatment
condition and time-of-assessment (i.e., T0, T1, T2, T3), because of (unwanted) comments
about the treatment made by parents or the assessor during the interaction tasks.
In spite of these limitations, we can conclude that a PMTO intervention produced
positive effects in a clinically referred sample of children with externalizing problems in the
Netherlands. More precisely, this treatment was effective in reducing children’s problem
behaviors (even showing a quite large effect size), increasing the use of self-reported
effective parenting practices, and reducing parenting stress and psychopathological
45
2
Chapter 2 | Effectiveness of PMTO
symptoms of the parents, albeit no more effective than CAU. For both conditions, the
improvements were most evident during the first six months of the study and remained
stable until 18 months after baseline. Although many effects of the present study were
in favor of PMTO and comparable to the effects of PMTO in other European countries,
CAU in our study appeared to perform better than the control conditions in most other
studies. It is remarkable to note that many of the CAU interventions performed within
the Dutch youth care system also include the therapeutic ingredients, such as the
use of ‘time out’ for disciplining and rewarding desired behaviors, that are considered
important in PMTO. In a future study, the cost-effectiveness of PMTO will be compared
to CAU. Annual youth service costs have been rising steadily over the past decade in
The Netherlands, and a cost-benefit analysis will provide policy makers and insurance
companies with quality information to guide decision-making, in the interest of young
children, families and society at large.
46
Treatment fidelity as
a determinant of the
effectiveness of Parent
Management Training Oregon model in The
Netherlands
3
Thijssen, J., Albrecht, G., Muris, P., & de Ruiter, C. (2015). Treatment fidelity as a
determinant of the effectiveness of Parent Management Training – Oregon model in
The Netherlands. Submitted for publication
Chapter 3 | Treatment fidelity in PMTO
ABSTRACT
The present study examined treatment fidelity as a possible determinant of
the effectiveness of Parent Management Training – Oregon model (PMTO) in The
Netherlands. Clinically referred children (N = 86) aged 4 to 11 years and their parents
received PMTO and were assessed at four time points: at baseline, and after 6, 12, and 18
months. Difference scores between baseline and follow-up assessments of externalizing
behavior problems, parenting practices, and parental psychopathology and parents’
overall ratings of treatment satisfaction, were correlated with treatment fidelity scores
measured prior to the intervention. Furthermore, differences in therapists’ fidelity scores
between treatment completers and drop-outs were examined. Results showed that
higher fidelity scores of PMTO therapists were associated with larger improvements in
externalizing behavior, parenting practices, and parental psychopathology, especially
after 18 months. In addition, parents who completed the treatment had a significantly
more adherent therapist than families who dropped out. However, the correlations
between treatment fidelity and treatment satisfaction was non-significant. These
findings indicate that therapists’ high adherence to the PMTO treatment principles
decreases the chance of treatment drop-out and positively affects the long-term
effectiveness of PMTO.
48
Treatment fidelity in PMTO | Chapter 3
INTRODUCTION
Several longitudinal studies have demonstrated that early childhood behavioral
problems such as disobedience, fighting, lying, and stealing are predictive of antisocial
conduct and delinquency in later life (Coté, Vaillancourt, LeBlanc, Nagin, & Tremblay,
2006; Simonoff et al., 2004). Moreover, the earlier these externalizing problems begin,
the higher the risk of serious problems during adulthood (Moffitt, 1993; Patterson,
DeBaryshe, & Ramsey, 1989; Reef, Diamantopoulou, van Meurs, Verhulst, & van der Ende,
2011). The fact that externalizing behavior problems in children may have adverse longterm consequences underlines the importance of early intervention programs. Many of
these programs focus on the improvement of parenting practices and there is indeed
evidence showing that the enhancement of positive and more effective parenting
strategies has positive effects on children’s development of prosocial behavior
(DeGarmo & Forgatch, 2005, 2007; Forgatch & DeGarmo, 1999; Martinez & Forgatch,
2001; Ogden & Amlund-Hagen, 2008).
Parent Management Training - Oregon model (PMTO) is a program especially
developed for the parents of children between 4 and 12 years of age with severe
externalizing behavior problems such as Oppositional Defiant Disorder or Conduct
Disorder. PMTO is based on the Social Interaction Learning (SIL) model, which
assumes that contextual factors, such as socio-economic disadvantage and parental
psychopathology, have a negative impact on child outcomes by undermining parenting
quality (Snyder & Patterson, 1995). Therefore, PMTO focuses on teaching parents how
to reduce negative, counterproductive parenting practices and to replace these with
positive, more effective parenting practices, such as encouragement (i.e., stimulation
of prosocial behaviors in the child, for example by using positive reinforcement),
effective discipline (i.e., consistent use of mild sanctions), monitoring (i.e., keeping track
of the child’s activities), problem solving (i.e., responding adequately to rule-breaking
behaviors and settling arguments with the child), and positive involvement (i.e., giving
love and warm attention to the child; Forgatch & DeGarmo, 1999; Forgatch, Patterson,
& DeGarmo, 2005).
In previous studies, PMTO has been shown to produce positive effects on child
externalizing behavior outcomes and parenting practices, and this appears not only
true in the United States where the program was developed (e.g., Forgatch & DeGarmo,
1999; Forgatch, Patterson, DeGarmo, & Beldavs, 2009), but also in other countries such
as Norway, Iceland, and The Netherlands (Ogden & Amlund-Hagen, 2008; Sigmarsdóttir,
Thorlacius, Guðmundsdóttir, & DeGarmo, 2014; Thijssen, Vink, Muris, & de Ruiter, 2015).
A logical next step in the process of evaluating PMTO would be to gain more insight
into the determinants of the effectiveness of this intervention. In the present study, we
investigated to what extent therapists’ adherence to PMTO principles (i.e., treatment
49
3
Chapter 3 | Treatment fidelity in PMTO
fidelity) is important for the effectiveness of PMTO. Increased knowledge of possible
determinants of PMTO’s effectiveness could be helpful in the further improvement of
the treatment method.
Prior research has provided tentative evidence indicating that therapists’ close
adherence to the treatment protocol is indeed associated with greater improvements in
parenting skills and more clear-cut reductions of externalizing behavior problems, and
this appeared not only true for PMTO (Forgatch & DeGarmo, 2011; Forgatch et al., 2005;
Hukkelberg & Ogden, 2013) but also for other interventions aimed at externalizing
behavior problems in children (e.g., Hogue et al., 2008; Huey, Henggeler, Brondino,
& Pickrel, 2000). Within PMTO, treatment fidelity is measured by using the Fidelity of
Implementation (FIMP) rating system (Knutson, Forgatch, Rains, & Sigmarsdóttir,
2009) which evaluates the therapist on five dimensions: PMTO knowledge, Structure,
Teaching, Process skills, and Overall development. The FIMP is a standard part of the
PMTO training program to evaluate the progress of PMTO therapists and it is also used
for certification and recertification.
The present study relies on the data collected for a study evaluating the
effectiveness of PMTO in The Netherlands (Thijssen et al., 2015c). In this study, only
certified PMTO therapists were involved, which implies that all therapists had at least
a sufficient FIMP score. However, therapists may still show considerable variation in
the extent to which they adhere to the method, with some of them being really strict
at following the guidelines of the intervention and others carrying out the protocol in
a looser fashion. The aim of the present study is to examine whether treatment fidelity
scores as obtained before treatment (i.e., at certification) are positively associated with
treatment outcome in terms of externalizing behavior problems, parenting practices,
parental psychopathology, and treatment satisfaction. Furthermore, we will examine
whether treatment fidelity was associated with treatment completion. We expected
that higher treatment fidelity scores of the therapists would be associated with larger
improvements on outcome measures and higher rates of treatment completion. The
current study’s procedure was different from previous studies on treatment fidelity in
PMTO (Forgatch & DeGarmo, 2011; Forgatch et al., 2005; Hukkelberg & Ogden, 2013) in
a number of ways: the present study (a) included multiple outcome measures of PMTO
instead of only parenting practices or externalizing behavior; (b) examined the FIMP
dimensions separately instead of using a mean or a single construct; and (c) examined
the association between fidelity scores and treatment outcome at different assessment
points. This new approach could provide new insights into the association between
PMTO treatment fidelity and treatment outcomes.
50
Treatment fidelity in PMTO | Chapter 3
METHOD
Participants
FIMP scores were available for 86 of the 91 families receiving PMTO. Children and their
parents were recruited through five child service agencies across The Netherlands. The
mean age of these children (62 were boys and 24 were girls) was 7.16 years (SD = 1.81).
Mean age of the main caregivers was 38.51 years (SD = 1.27); 25% of them were single
parents and 76% had Dutch nationality. Approximately 66% had a college or university
degree, 28% had completed high school, and 3% had only finished elementary school.
Families were recruited as part of a study evaluating the effectiveness of PMTO in
The Netherlands, in which this intervention was compared to care as usual. For obvious
reasons, only families who received PMTO were used in the present study. To be
included, the child had to reveal a T-score of 60 or higher on the externalizing subscale
of the Child Behavior Checklist (CBCL; Achenbach, 1991; Dutch version: Verhulst, van
der Ende, & Koot, 1996) and the child had to be residing at home with at least one
biological or adoptive parent. Exclusion criteria were severe intellectual disability or
psychopathology of the parent(s) (including substance use disorders) that would
interfere with participation in treatment, sexual abuse in the family, an IQ of the child
lower than 70.
Parents in the present study received PMTO from 25 certified therapists whose
treatment fidelity scores were determined prior to the start of the treatment. All
therapists had completed the full PMTO training program of approximately 24 months.
During this training period, therapists had to treat at least three families with PMTO
before they were allowed to take part in the official PMTO certification procedure. This
procedure involved treating another family with PMTO. On the basis of videotaped
sessions of this therapy, it was determined whether or not the therapists received their
license to carry out PMTO in clinical practice. Following the completion of the training
program, therapists were regularly monitored on their treatment fidelity, leading to
annual recertification of their license.
Procedure
Families were included in the period between June 2009 and January 2014. As
soon as families were referred to the child service agency, it was checked whether they
met the criteria for being included in the study. Families who met the criteria received
information about the study and its procedure and were invited to participate. When
parents agreed, they were asked to give their written consent. The study protocol was
approved by the Medical Ethics Committee of Maastricht University Medical Centre.
51
3
Chapter 3 | Treatment fidelity in PMTO
Families were assessed at four time points: a baseline assessment (T0) and three
follow-up measurements, after 6 (T1), 12 (T2), and 18 months (T3). Trained research
assistants, who were not the family’s therapist, administered the interview for measuring
parenting practices. A web-based system was used for the self-report assessments,
which enabled parents to complete questionnaires on a computer at home. In all cases,
the main caregiver (i.e., the parent who spent most time with the child) filled out the
scales. All participating parents received a small financial compensation in gift vouchers
for the three follow-up measurements (i.e., €10 at T1; €20 at T2; and €30 at T3).
In PMTO, the therapist works with the parent(s) of one family. The children are
not present during these sessions. Role play is an important mechanism in the PMTO
sessions to teach and practice the effective parenting practices. In between sessions, the
therapist calls the parent to ask about homework, to provide advice and support, and
to answer questions. Treatment duration depends on the family’s needs and progress
throughout the therapy, but typically lasts between 15 and 25 weekly sessions.
Assessments
Treatment fidelity
Videotapes of four standard therapy sessions, obtained during the certification
procedure, were assessed by means of the FIMP rating system (Knutson et al., 2009).
This rating system measures the degree of therapists’ adherence to the PMTO treatment
model along five dimensions: PMTO knowledge, Structure, Teaching, Process skills, and
Overall development, which will be described briefly hereafter. Each dimension is scored
on a 9-point scale, in which 1-3 can be considered as ‘unacceptable’ (‘needs work’),
4-6 as ‘acceptable’, and 7-9 as ‘good’. The FIMP ratings were performed by a group of
certified FIMP coders (all PMTO educators), both from the US and the Netherlands. All of
the FIMP raters have passed the annual reliability tests conducted by the Oregon Social
Learning Center. Each session of the same PMTO therapist was rated by a different FIMP
coder. In order to ‘pass’, an average rating of 6.0 or higher (as a mean score including all
5 dimensions) was needed.
The FIMP dimensions are defined as follows. PMTO knowledge, which reflects the
therapist’s understanding of PMTO and SIL principles. Structure refers to balancing
several activities during the session, such as following an agenda, maintaining an
orderly flow, leading without dominating, responsiveness to family issues, and using
sensitive timing and pacing. Teaching comprises proficiency in strategies and tools
that promote parents’ mastery and use of PMTO practices. The use of verbal teaching
(e.g., giving information, making suggestions, providing rationale) and active teaching
(engaging the family in the learning process by brainstorming, prompting, role play,
asking solution-eliciting questions) should be balanced. Process skills provide support
52
Treatment fidelity in PMTO | Chapter 3
to create a safe and supportive learning context (e.g., maintaining appropriate balance,
encouraging skill development, joining the family’s storyline). And finally, Overall
development reflects the accomplishment of session goals and includes, for example, the
likelihood that the parents will use the procedures, the parent’s apparent satisfaction,
and the likelihood they will continue with the intervention.
Externalizing behavior problems
The Child Behavior Checklist (CBCL) is a widely used rating scales for assessing
behavioral and emotional problems of children aged 6 to 18 years (Achenbach, 1991;
Dutch version: Verhulst et al., 1996). Each scale consists of 120 items scored on a 3-point
Likert scale (0 = not true; 1 = somewhat or sometimes true; 2 = very or often true). For
the present study, only the externalizing scale was used. The CBCL has good reliability
and validity (Achenbach, 1991; Verhulst et al., 1996).
The Parent Daily Report (PDR) is a reliable 34-item checklist of problem behaviors
(Patterson, Chamberlain, & Reid, 1982) and is administered by a research assistant other
than the therapist. At first, the checklist is administered face-to-face with the main
caregiver to assess whether specific problem behaviors of the child have occurred
during the past six months. Next, the items to which the parent responds affirmatively
are again administered via telephone on three consecutive days to examine whether
these behaviors have occurred during the past 24 hours. The items involve both severe
(e.g., arson, stealing) and less serious problem behavior (e.g., being rude, complaining).
The scores of the three consecutive days are summed to obtain the total PDR score.
Parenting practices
The Dutch translation of the Caregiver Wish List (CWL; Hodges, 2005; Hodges, de Ruiter,
& Thijssen, 2009) is an interview-based instrument consisting of 53 items questioning
the parent about his/her parenting skills. The interviewer reads the questions to the
parent, who has to indicate the most applicable response option using a 5-point Likert
scale. Items are allocated to six domains of parenting skills: providing direction and
following up (4 items), encouraging good behavior (5 items), discouraging undesirable
behavior (24 items), monitoring activities (13 items), connecting positively with the
child (3 items), and problem solving orientation (4 items). Each domain score should be
regarded as a dimension with weak parenting skills on one end and strong parenting
skills on the other. In the current study, only the discouraging undesirable behavior
domain and the CWL total score were used, since these were the only scales showing
adequate internal consistency (.80 and .76, respectively).
53
3
Chapter 3 | Treatment fidelity in PMTO
Parental stress and psychopathology
The Nijmeegse Ouderlijke Stress Index (NOSI; de Brock, Vermulst, Gerris, & Abidin,
1992) is an adaptation of the Parenting Stress Index (Abidin, 1983) and measures stress
experienced by parents in the relationship with their child. The NOSI comprises 123
items that have to be rated on a 5-point Likert scale (1 = strongly disagree, 5 = strongly
agree). Ratings on all items can be summed to create a total stress score, with higher
scores reflecting higher levels of perceived stress by the parent. The NOSI has adequate
reliability and validity (de Brock et al., 1992).
Psychopathological symptoms of the parents were measured by the Dutch version
of the Symptom Checklist-90 Revised (Arrindell & Ettema, 2003). The 90 items are rated
using a 5-point scale (1 = no problem to 5 = very serious) to indicate the extent to which
the parent has experienced the listed symptom during the previous week. In contrast to
the original version of the SCL-90-R, the Dutch version comprises eight instead of nine
subscales (e.g., Anxiety, Somatic Symptoms, Depression). For the present study, the SCL90-R total score was used, for which higher scores on the SCL-90-R indicate more serious
psychopathology.
Parent satisfaction
The short form of the Working Alliance Inventory (WAI-S; Tracey & Kokotovic, 1989)
was used to assess the quality of the parent-therapist alliance. The WAI-S comprises 12
items that can be allocated to three subscales of four items each: (a) agreement between
parent and therapist on the goals of the therapy; (b) the degree to which the tasks of
the therapy will actually address the parent’s problems; and (c) the quality of the bond
between the parent and the therapist. Normally, the items of the WAI-S are rated using a
7-point Likert scale. However, in the present study, a 5-point scale was used for practical
reasons. Ten items are positively worded and two items are negatively worded. The scores
on the negatively stated items are reversed, so that all scores can be summed to obtain
a total score, with higher scores reflecting a better parent-therapist working alliance.
Treatment completion
The PMTO treatment was considered completed when the core of PMTO parenting
skills was dealt with in a satisfactory way (good directions – encouragement – setting
limits). Of the 86 PMTO families in the present study, 64 families completed the PMTO
treatment.
Missing data
The proportion of missing values in this data set ranged from 0% in the demographic
variables to approximately 50% on the NOSI. We applied an intention-to-treat procedure
54
Treatment fidelity in PMTO | Chapter 3
by conducting multiple imputation (Rubin, 1987) to handle the missing data through
a chain of conditional regression models (fully conditional specification; Van Buuren,
Boshuizen, & Knook, 1999). We used predictive mean matching (PMM; Little, 1988; Rubin,
1986) for the scale variables and polytomous regression for ordered categorical data. All
computations were carried out with Mice (Van Buuren & Groothuis-Oudshoorn, 2011) in
R (R Core Team, 2014), with 150 iterations for the algorithm to converge and 25 multiply
imputed datasets, using available and custom imputation routines. The outcomes over
the 25 datasets were combined into a single inference using Rubin’s (1987) rules.
Analytic strategy
To examine the association between the FIMP scores and treatment outcome,
Pearson correlation coefficients were calculated between the five FIMP dimensions and
outcome scores for the questionnaires measuring externalizing behavior problems,
parenting skills, parenting stress, psychopathological symptoms of the parents, and
treatment satisfaction. Only those measures were used for which the effectiveness
study revealed significant effects of PMTO (see Thijssen et al., 2015c). Treatment effects
were operationalized in terms of difference scores and calculated for each assessment
instrument: scores at baseline (T0) were subtracted from scores at 6 (T1), 12 (T2), and
18 (T3) months. Negative difference scores indicated improvement for externalizing
behavior problems, parenting stress, and parental psychopathology. For parenting
practices, improvement was indicated by positive difference scores.
RESULTS
First, we analyzed whether treatment fidelity was related to treatment outcome in
terms of externalizing behavior problems. We calculated Pearson correlation coefficients
between the five FIMP dimensions and overall FIMP score and difference scores
obtained with the CBCL and PDR at the various time points. As can be seen in Table 1, a
number of significant associations emerged for the CBCL. For the T3-T0 difference score,
significant negative correlations were found for three of the five FIMP dimensions: PMTO
knowledge, Teaching, and Overall development. This finding indicates that the higher
the score on these FIMP dimensions, the larger the decrease in CBCL externalizing
behavior from baseline to 18 months follow-up. Furthermore, a significant negative
correlation was found between PMTO knowledge and the CBCL T1-T0 difference score.
For the T2-T0 difference score, none of the correlations attained statistical significance
and this was also the case for correlations involving PDR difference scores.
55
3
Chapter 3 | Treatment fidelity in PMTO
Table 1 | Pearson correlation coefficients between FIMP scores and treatment outcome as indexed by CBCL and
PDR difference scores
CBCL
PDR
T1-T0
T2-T0
T3-T0
T1-T0
T2-T0
T3-T0
PMTO knowledge
-.22*
-.09
-.33**
.06
.16
.08
Structure
-.17
-.00
-.19
-.07
.06
.01
Teaching
-.12
-.03
-.23*
-.05
.04
-.01
Process skills
-.16
-.04
-.15
-.06
.05
-.01
Overall development
-.17
-.07
-.27*
.02
.12
.04
Note. FIMP = Fidelity of IMPlementation; CBCL = Child Behavior Checklist; PDR = Parent Daily Report
*p < .05 (1-tailed)
**p < .01 (1-tailed)
Second, we examined the association between treatment fidelity and changes in
parenting practices. Pearson correlation coefficients were calculated between the FIMP
and the difference scores on CWL Discouraging undesirable behavior domain and
the CWL total score. Table 2 shows that only one correlation reached significance for
Discouraging undesirable behavior: that is, FIMP Overall development was positively
associated with discouraging undesirable behavior from T0 to T3, indicating that the
higher the score on Overall development, the more improvement parents reported
on Discouraging undesirable behavior. For the CWL total score, for both the T1-T0
and T3-T0 difference scores, significant positive correlations were found for the FIMP
dimensions referring to Structure, Teaching, and Overall development. These results
show that the higher a PMTO therapist’s scores on these FIMP dimensions, the greater
the improvement in parenting skills. Again, no significant correlations were found for
the T2-T0 difference score.
Table 2 | Pearson correlation coefficients between FIMP scores and treatment outcome as indexed by CWL
discouraging undesirable behavior domain and CWL total difference scores
Discouraging undesirable behavior
T1-T0
T2-T0
T3-T0
PMTO knowledge
-.07
-.10
Structure
.05
.05
Teaching
.04
Process skills
.01
Overall development
.07
T1-T0
T2-T0
.06
.11
-.06
.09
.14
.27*
.06
.21*
.08
.18
.21*
.10
.24*
.02
.10
.16
.03
.15
.09
.21*
.24*
.12
.26*
Note. FIMP = Fidelity of IMPlementation; CWL = Caregiver Wish List
*p < .05 (1-tailed)
56
CWL total score
T3-T0
Treatment fidelity in PMTO | Chapter 3
Third, we examined the association between treatment fidelity and parental
psychopathology. Difference scores of the NOSI and SCL-90-R were correlated with the
FIMP dimensions. As shown in Table 3, for the NOSI, a significant negative correlation
was found between the T1-T0 difference score and FIMP Process skills. Furthermore,
all FIMP dimensions were significantly negatively associated with the NOSI T3-T0
difference score. For the SCL-90-R, a significant correlation was found between the T3T0 difference score and FIMP PMTO knowledge. Thus, a higher score on this dimension
is related to a larger decrease in psychopathological symptoms reported by parents. No
significant correlations were found for the T1-T0 and T2-T0 difference scores in parental
psychopathology.
Table 3 | Pearson correlation coefficients between FIMP scores and treatment outcome as indexed by NOSI and
SCL-90-R difference scores
NOSI
SCL-90-R
T1-T0
T2-T0
T3-T0
T1-T0
T2-T0
T3-T0
PMTO knowledge
-.17
-.11
-.30*
-.13
-.10
-.23*
Structure
-.23
-.11
-.30*
-.11
-.13
-.21
Teaching
-.18
-.12
-.28*
-.07
-.10
-.15
Process skills
-.26*
-.13
-.30**
-.17
-.12
-.18
Overall development
-.21
-.16
-.31**
-.08
-.13
-.20
Note. FIMP = Fidelity of IMPlementation; NOSI = Parenting Stress Index; SCL-90-R = Symptom Check List-90Revised
*p < .05 (1-tailed); **p < .01 (1-tailed)
Fourth, we tested whether FIMP scores were related to treatment satisfaction of the
parents. Scores on the WAI-S at T1, T2, and T3 were used in the analyses. No significant
correlations were found between the FIMP dimensions and the WAI-S scores, implying
that FIMP scores were unrelated to treatment satisfaction of the parents.
Finally, we examined whether FIMP scores were related to treatment completion.
Independent samples t-tests were performed to compare completers and drop-outs on
the five FIMP dimensions. Significant differences between treatment completers and
drop-outs were documented for all FIMP dimensions (see Table 4). Therapists’ of families
who completed the core of the PMTO treatment had mean FIMP scores in the ‘good
work’ range on all dimensions, whereas therapists’ of families who dropped out had
scores in the ‘acceptable’ range. This indicates that parents were more likely to complete
the PMTO treatment when their therapists adhered more closely to the PMTO principles.
57
3
Chapter 3 | Treatment fidelity in PMTO
Table 4 | Mean FIMP scores of treatment completers and drop-outs
PMTO knowledge
completers
(n = 64)
drop-outs
(n = 22)
t
p
7.10
6.66
2.70
.01
Structure
7.23
6.55
4.19
.00
Teaching
7.03
6.31
4.20
.00
Process skills
7.23
6.70
3.67
.00
Overall
development
7.20
6.65
4.09
.00
Note. The maximum score is 9. FIMP = Fidelity of IMPlementation
DISCUSSION
The aim of the present study was to gain further insight into the role of treatment
fidelity as a determinant of the effectiveness of PMTO in The Netherlands. We examined
whether treatment fidelity scores obtained for PMTO certification purposes prior to the
intervention would be associated with larger treatment effects on various outcome
variables, including child externalizing behavior problems, parenting practices, parental
psychopathology, parental treatment, and treatment completion. This is the first study
that explores the link between treatment fidelity and multiple outcome measures of
PMTO. Previous studies on this topic mainly focused on the association between FIMP
scores and parenting practices or externalizing behavior problems as the outcome
variable (Forgatch & DeGarmo, 2011; Forgatch et al., 2005; Hukkelberg & Ogden, 2013).
In line with earlier investigations, the results of the present study provide indications
that treatment fidelity is indeed related to greater improvements in child externalizing
behavior. The higher therapists’ level of PMTO knowledge the larger the change in
externalizing behavior on the CBCL between baseline and 6 months. Furthermore,
higher levels of PMTO knowledge, Teaching, and Overall development were related
to larger improvements in externalizing behavior at the 18-months follow-up. This
points out that higher treatment fidelity scores of the therapist on these dimensions
were accompanied by a larger decrease in externalizing behavior on the CBCL, and that
this was especially the case at the 18-months follow-up. It is noteworthy that the 18
months follow-up point lies well beyond the PMTO treatment period, which usually
lasts between 6 and 9 months. Thus, our findings point to the relevance of treatment
fidelity of PMTO’s longer term effectiveness. Remarkably, no significant associations
were found between FIMP scores and PDR difference scores. However, the treatment
effect based on the PDR (d = .32) is lower compared to the treatment effect based on
the CBCL (d = .71). This could explain why no associations were found between the FIMP
and PDR scores.
58
Treatment fidelity in PMTO | Chapter 3
With regard to parenting practices, it was found that higher scores on the FIMP
dimension Overall development were related to larger increases in the use of
discouraging undesirable behavior from baseline to 18-months follow-up. Higher
CWL total scores were associated with higher levels of the FIMP dimensions Structure,
Teaching, and Overall development from baseline to 6 and 18 months. Our findings
imply that higher treatment fidelity is associated with larger improvements in
parenting practices and with an increase in discouragement of undesirable behavior
in particular. Treatment fidelity was also related to decreases in parenting stress and
parental psychopathology. Parents experienced significantly less parenting stress after
six months when therapists had higher levels of Process skills. After 18 months, higher
scores on all FIMP dimensions were associated with decreased levels of parenting
stress. Psychopathological symptoms were significantly lower after 18 months when
therapists had higher levels of the FIMP dimension PMTO knowledge. Thus, parents
report less parenting-related stress and psychopathological symptoms, especially at
18-months follow-up, when their therapists more strongly adhere to the PMTO program.
Meanwhile, treatment fidelity was not associated with greater parental satisfaction, in
other words, parents were not more satisfied with their therapist and the intervention
when their therapist adhered more strictly to PMTO guidelines. Treatment satisfaction
appears independent of how well the therapist delivers the PMTO treatment.
Most significant associations were found from baseline to 18-months followup, whereas fewer significant associations could be documented from baseline to 6
months and from baseline to 12 months. This suggests that high treatment fidelity is
particularly important for the long-term effects of PMTO. In the short term, it seems to
be less important how PMTO is delivered by the therapist. However, for the endurance
of the effects of PMTO, it seems to be essential that the therapist strongly adheres to
the PMTO method. Note that in the present study all therapists had been certified and
therefore all of them had been judged as sufficiently adherent. To achieve optimal longterm treatment effects, it seems to be necessary for PMTO therapists to perform better
than merely sufficient.
The FIMP dimensions Structure and Overall development were most frequently
related to treatment outcome measures and, thus, seem to be the most important
dimensions of treatment fidelity. Hence, the therapist should be especially capable
of balancing several activities during the session, to increase the likelihood that the
parents apply the PMTO parenting strategies, and enabling them to manage unique or
difficult situations. PMTO knowledge was mainly related to a decrease in externalizing
behavior problems, suggesting that the therapist’s understanding of PMTO and the SIL
principles are important for improvement in externalizing behavior problems, while
creating balance between several activities during the session seems to be important
59
3
Chapter 3 | Treatment fidelity in PMTO
for improvements in parenting practices. The FIMP Process skills dimension was only
related to a decrease in parenting stress. Therefore, creating a safe and supportive
learning context seems to be important in reducing stress of parents, which might
enable them to benefit more from the treatment.
Finally, we examined whether FIMP scores differed between therapists of families
who completed the treatment and families who dropped out. Parents who completed
PMTO more often had a therapist who adhered more strongly to the PMTO principles
than parents who dropped out. This implies that PMTO therapists scoring high on
treatment fidelity are better able to keep parents in treatment than therapists showing
lower treatment fidelity.
Previous studies into PMTO treatment fidelity showed that competent adherence
to the PMTO principles predicted the degree of change in parenting (i.e., Forgatch
& DeGarmo, 2011; Forgatch et al., 2005; Hukkelberg & Ogden, 2013). These previous
studies used a different analytical approach, which makes direct comparison with
our results difficult. However, two studies reported descriptive statistics, including
correlations. In the study of Hukkelberg and Ogden (2013), none of the correlations
between the FIMP total score and child problem behavior outcomes were significant.
Forgatch and DeGarmo (2011) only found a significant positive correlation between the
mean FIMP total score and father’s post-treatment parenting (r = .27). FIMP scores were
not significantly related to change in mothers’ and fathers’ parenting. The differences in
correlations between the present study and previous studies could be explained by the
different approach of the studies. The present study examined more aspects of treatment
fidelity than previous studies evaluating treatment fidelity in PMTO. None of the
previous studies examined the separate FIMP dimensions. Our study showed that some
dimensions are more strongly related to therapy outcome variables than others, which
could be useful information for improving the treatment method. Further, in contrast to
the previous studies on treatment fidelity in PMTO, the present study investigated the
relations between fidelity and effectiveness at different assessment points and showed
that treatment fidelity was especially related to longer-term outcomes.
Some limitations of the present study should be mentioned. First, the present
study was correlational in nature. Obviously, this type of research indicates that there
is a relationship between two variables, but cannot prove that one variable causes a
change in the other. Second, therapists’ FIMP scores obtained during certification were
used; these scores pertained to other families than the families treated in the present
study. It is possible that some therapists deteriorated in terms of fidelity while others
improved since the time of their certification. As a result, PMTO therapists might have
performed differently during certification than during treatment of the PMTO families
in the present study. However, correlations between the FIMP scores and outcome
60
Treatment fidelity in PMTO | Chapter 3
measures found in our study are not lower than those reported in previous studies in
which FIMP scores were obtained from the same families as the outcome variables (see
Forgatch & DeGarmo, 2011; Hukkelberg & Ogden, 2013). Furthermore, Hukkelberg and
Ogden (2013) found a high stability in PMTO therapists’ fidelity scores over a period of
approximately 9 months (correlations ≥ .30). Therefore, FIMP scores obtained prior to
treatment seem to be appropriate as measure of subsequent treatment fidelity.
In conclusion, the current study found support for the notion that treatment fidelity
in PMTO is related to treatment outcome. The higher the therapist’s fidelity scores, the
larger the improvements in externalizing behavior, parenting practices, parenting stress,
and parental psychopathological symptoms. This was not translated into higher levels
of treatment satisfaction; parents were not more satisfied with their therapist and the
intervention when the therapist was more adherent to the protocol. The finding that
treatment fidelity was most consistently associated with outcome at the 18-months
follow-up assessment indicates that for the long-term effects of PMTO, it is important
that PMTO is delivered as intended by the developers and according to the treatment
principles in more than merely an acceptable fashion. Furthermore, it was found that
parents were more likely to complete the PMTO treatment when their therapist strongly
adheres to the PMTO principles. Based on these results, it is advisable to raise the bar
for certification, because performing acceptably seems to be insufficient for attaining
positive long-term treatment effects with PMTO.
61
3
Initial validation of the
Dutch translation of
the Caregiver Wish List,
an interview-based
scale for measuring
parenting practices
4
Thijssen, J., Broers, N. J., Muris, P., & de Ruiter, C. (2015). Initial validation of the Dutch
translation of the Caregiver Wish List, an interview-based scale for measuring
parenting practices. Submitted for publication
Chapter 4 | Initial validiation of the Caregiver Wish List
ABSTRACT
Research has demonstrated that adequate parenting is an important determinant
of a healthy social-emotional development in children. There is a great need for valid
assessment tools for measuring the quality of parenting, particularly in clinical settings.
The Caregiver Wish List (CWL) is a new 53-item interview-based scale for assessing
parenting practices. We examined the factor structure of the CWL in a sample of 348
parents of children aged between 4 and 11 years, of which 220 were drawn from the
general population and 128 from a clinical setting. Exploratory factor analysis revealed
five factors, which did not fully correspond with the hypothesized, original factor
structure. Nonetheless, the extracted factors were meaningful and could be labeled as:
Adequate discipline, Controlled responding, Focus on positive behavior, Consistency,
and Monitoring. The factors demonstrated adequate internal consistency. The factor
structures in the community and clinical samples were by and large comparable,
which supports the generalizability of the factor structure. Furthermore, the factors
differentiated between the community and clinical sample, with better parenting
skills observed in the community sample. Finally, all factors were significantly and
negatively related to child psychopathology, with stronger correlations demonstrated
for externalizing than for internalizing problems. Directions for future research with the
CWL are discussed.
64
Initial validiation of the Caregiver Wish List | Chapter 4
INTRODUCTION
Ineffective parenting is the most widely studied and empirically established risk
factor for externalizing behavior problems in children (DeBaryshe, Patterson, & Capaldi,
1993; Forgatch, Bullock, & Patterson, 2004; McCoy, Frick, Loney, & Ellis, 1999; Nix et al.,
1999; Oliver, Guerin, & Coffman, 2009; Patterson, Forgatch, Yoerger, & Stoolmiller, 1998;
Sharma & Sandhu, 2006; Williams et al., 2009). In particular, harsh and inconsistent
discipline, poor monitoring and supervision, and low positive involvement have all been
demonstrated to be significantly associated with young people’s oppositional-defiant
and conduct problems (Bierman & Smoot, 1991; Cunningham & Boyle, 2002; Ehrensaft
et al., 2003; Frick et al., 1992; Nicholson, Fox, & Johnson, 2005). Many interventions for
childhood behavior problems focus on increasing effective parenting and decreasing the
use of ineffective rearing practices. Research generally supports the theoretical notion
that improving parenting practices is the most important mechanism contributing to a
decrease in children’s behavior problems (DeGarmo & Forgatch, 2005, 2007; Forgatch
& DeGarmo, 1999; Kazdin, 2007; Martinez & Forgatch, 2001; Ogden & Amlund-Hagen,
2008). Improvements in the behavior problems of the children also tend to reduce stress
and depression in parents, which in turn will increase the likelihood that the effective
parenting practices will be maintained, thereby promoting the long-term effects of
therapy (DeGarmo, Patterson, & Forgatch, 2004).
Many parent training programs are based on the Social Interaction Learning (SIL)
model (Forgatch et al., 2004). The SIL model assumes that the psychosocial development
of the child is directly influenced by the parents and their parenting strategies.
Contextual factors, however, can have a negative impact on the parenting quality, and
thus, indirectly on the child. For example, when parents are going through a divorce, this
puts pressure on the parenting quality (Conger, Patterson, & Ge, 1995; Hetherington,
Bridges, & Insabella, 1998). According to the SIL model, there are five core effective
parenting practices: skill encouragement, discipline, monitoring, problem solving, and
positive involvement. Skill encouragement refers to the enhancement of a prosocial
development by using scaffolding techniques (e.g., breaking behavior into small steps,
prompting appropriate behavior) and the provision of positive reinforcement (e.g.,
giving compliments when the child behaves well; Forgatch & DeGarmo, 1999). Discipline
decreases problem behaviors by the appropriate and consistent use of mild sanctions,
like giving a time out or taking away privileges (Patterson, 1986). Monitoring involves
knowing the child’s friends and keeping track of the child’s activities to protect him/her
against the negative influence of deviant peers (Snyder, 2002). Problem solving helps
families to negotiate during arguments, to clearly determine rules and consequences
when the rules are violated (Forgatch, Patterson, & DeGarmo, 2005). And finally, positive
involvement concerns the many ways in which parents give their child loving attention
65
4
Chapter 4 | Initial validiation of the Caregiver Wish List
and engagement in joint fun activities (Forgatch & DeGarmo, 1999; Forgatch et al., 2005).
Many studies rely on observations of parent-child interactions as a method to assess
parenting skills (Martinez & Forgatch, 2001; Ogden & Amlund-Hagen, 2008; Patterson,
DeGarmo, & Forgatch, 2004). For instance, parents and child can be observed during a
structured playroom session or in a natural situation at home or in school. Self-report
measures can be used as an alternative assessment method which complements
the observation measures of parenting. In fact, there are indications that a multimethod assessment of parenting possesses incremental validity over using only one
measurement method (Harvey, Danforth, Ulaszek, & Eberhardt, 2001; Meyer et al., 2001).
There are many self-report measures that can be used for the assessment of
parenting, although there is a wide variety in terms of definition and theoretical
framework. The Caregiver Wish List (CWL; Hodges, 2002) is a structured interview-based
instrument that enables parents and other caregivers to reflect on their own parenting
practices and talk about their experiences with parenting. Interestingly, the CWL was
designed to measure the core parenting practices of the SIL model and seems to be
particularly useful in clinical practice for a number of reasons. First, the CWL may help
clinicians to gather information on the strengths and weaknesses in parenting practices
as perceived by parents themselves. Furthermore, the CWL can be expected to enhance
the therapeutic alliance and the positive engagement between parents and the clinician.
Finally, the CWL may assist in identifying and clarifying the goals of treatment. The goal
of the CWL is to empower parents by encouraging them to see themselves as the main
agent of change for their child (Hodges, 2005). The CWL contains six a priori domains
of parenting skills: providing direction and following up, encouraging good behavior,
discouraging undesirable behavior, monitoring activities, connecting positively with
the child, and problem solving orientation.
In the Netherlands, the CWL is being used as an intake assessment for Parent
Management Training - Oregon model (PMTO), a parent training program that teaches
parents effective parenting skills according to the SIL model with the aim of reducing
externalizing behavior problems in children aged 4 to 12 years. This makes sense as the
CWL directly measures the parenting skills that are the main focus of change in PMTO.
A Dutch translation of the CWL was also employed in a large-scale study evaluating
the effectiveness of PMTO in The Netherlands (Thijssen, Vink, Muris, & de Ruiter, 2015).
However, in that study, it was found that the reliability of the CWL was rather poor, with
only one of the six original parenting domains showing adequate internal consistency.
In fact, to our knowledge, no study can be found that examined the psychometric
properties of the CWL. The aim of the present study was a first attempt to fill this gap.
The CWL was administered in a large sample of parents of children from the community
as well as a clinical setting visited by parents of whom children displayed disruptive
66
Initial validiation of the Caregiver Wish List | Chapter 4
behavior problems. In this way, it became possible to explore the factor structure of the
instrument as well as its reliability (internal consistency and correlations between factors)
and validity (relations between parenting scores and children’s psychopathological
symptoms).
METHOD
Participants
The total sample consisted of 348 parents (92.1% mothers) of 4- to 11-year-old
children (mean age = 8.07 years, SD = 1.61). Of these children, 185 (53.2%) were boys
and 163 (46.8%) were girls. The mean age of the main caregiver was 39.06 years (SD
= 5.15). The majority of the main caregivers was the biological parent of the child
(92.8%), had the Dutch nationality (89.1%), and was employed (72.4%). Of the 348 main
caregivers, 222 (63.8%) were living with a partner. Seventy-three percent of the main
caregivers had a college or university degree, 17.9% had completed high school, while
1.1% had only finished elementary school. In Table 1, the demographic characteristics
are reported for the community and clinical sample separately. In comparison to the
clinical sample, children and parents in the community sample were significantly older
[t(346) = 7.94, p < .01 and t(327) = 4.95, p < .01, respectively]. Moreover, the community
sample contained more girls [χ2(1) = 16.00, p < .01] and more two-parent households
[χ2(1) = 8.57, p < .01] than the clinical sample.
Table 1 | Demographic data and significant differences for the community and clinical sample
Community
(n = 220)
M (SD) or %
Clinical
(n = 128)
M (SD) or %
p
Child age
8.58 (1.27)
7.18 (1.75)
.00
Parent age
40.11 (4.42)
37.00 (5.83)
.00
Child is a boy
45.0
67.2
.00
Main caregiver is female
93.6
89.1
.21
Biological parent
96.8
98.2
.64
Two parent household
89.1
77.0
.00
Dutch nationality
92.7
94.6
.74
Employment
77.7
76.4
.79
67
4
Chapter 4 | Initial validiation of the Caregiver Wish List
Procedure
The parents of the community sample were recruited in elementary schools
where information letters explaining the nature of the study and consent forms were
distributed, together with a brief survey to obtain background information on the
parents (e.g., ethnicity, educational level, employment status). When parents agreed to
participate, they signed the consent form, filled out the demographic characteristics
survey, and returned these materials to the researchers via the child’s teacher at school.
The parent who spent most time with the child (main caregiver) was contacted to make
an appointment for the CWL interview. The CWL interview was administered at the
child’s school and lasted for about 30 minutes. As a reward, parents received a €10 gift
voucher. This study was approved by the Ethical Committee of the Faculty of Psychology
and Neuroscience at Maastricht University.
The parents of the clinical sample were recruited as part of a study on the
effectiveness of PMTO in children with disruptive behavior problems. As soon as families
were referred to the child service agency, it was checked whether they met the inclusion
criteria for the study (for details, see Thijssen et al., 2015c). Families who met the criteria
received information about the study and its procedure and were invited to participate.
When parents agreed, they were asked to give their written consent. The CWL interview
was completed by the main caregiver and mostly took place at the child service agency.
For the present study, only the CWL data that were obtained prior to the treatment were
employed.
Instruments
As noted in the introduction, the Caregiver Wish List (CWL; Hodges, 2005; Hodges,
de Ruiter, & Thijssen, 2009) is an interview-based instrument consisting of 53 items
questioning the parent about his/her parenting skills. The interviewer reads the
questions to the parent, who has to indicate the most applicable response option using
a 5-point Likert scale. The response options are specific to the question, although they
mostly refer to the frequency of behaviors. The parent also has the option to respond
with ‘not applicable’ when the described situation does not apply to their family.
Twenty-nine items are positively formulated, whereas 24 items are negatively phrased.
The latter items are reversed so that higher CWL scores reflect higher levels of effective
parenting. On the basis of conceptual work by Hodges (2002), items can be allocated to
six a priori domains of parenting skills: providing direction and following up (4 items),
encouraging good behavior (5 items), discouraging undesirable behavior (24 items),
monitoring activities (13 items), connecting positively with child (3 items), and problem
solving orientation (4 items). Each domain score can be regarded as a dimension with
weak parenting skills on one end and strong parenting skills on the other.
68
Initial validiation of the Caregiver Wish List | Chapter 4
The Child Behavior Checklist (CBCL) is a widely used rating scales for assessing
behavioral and emotional problems of children aged 6 to 18 years (Achenbach, 1991;
Dutch version: Verhulst, van der Ende, & Koot, 1996). Each scale consists of 120 items
scored on a 3-point Likert scale (0 = not true; 1 = somewhat or sometimes true; 2 = very
or often true). For the present study, only the internalizing and externalizing scales were
used. The CBCL has good reliability and validity (Achenbach, 1991; Verhulst et al., 1996).
Data analyses
Since no research on the factor structure of the CWL has been published, a
confirmatory factor analysis was considered as less appropriate. Instead, we performed
an exploratory factor analysis (EFA), using the pre-imposed conceptual structure as a
guide in the process of exploration. Exploratory factor analyses were conducted with
Mplus (version 5.21), using robust maximum likelihood estimation (MLR) to correct for
the non-normality in the distribution of the items. An oblique rotation method (geomin)
was applied as it seemed plausible that the extracted factors of positive parenting
would be correlated. Unlike principal components analysis, EFA does not permit the
quantification of the proportion of common variance explained. For an alternative
indication of the quality of our factor solution, we decided to examine a few model
fit indices that are usually reported in the context of confirmatory factor analysis: the
Tucker Lewis Index (TLI), the Root Mean Square Error of Approximation (RMSEA) and the
Standardized Root Means Square Residual (SRMR). These fit indices give complementary
information. Although they should be considered with caution when conducting EFA,
they provide a useful impression on the quality of the factor solution. Fit values that
would indicate an accurate description of the original correlation matrix would be a TLI
value larger than 0.95, a RMSEA value smaller than 0.05, and an SRMR value smaller than
0.08 (Hu & Bentler, 1999).
Even though the separate samples were relatively small in comparison to the number
of items to be processed in the factor analysis, we decided to initially explore the factor
structure for each of the samples separately. Using smaller samples in the factor analysis
will increase the instability of the solution, but prevents possible confounding because
of using very heterogeneous populations. Only in case of sufficient similarity in the
factor solutions for the separate samples, it seemed justified to interpret the analysis for
both samples combined.
To examine whether the CWL factors were significantly able to discriminate between
the community and clinical sample, independent samples t-tests were performed. The
internal consistency of the extracted factors was examined using Cronbach’s alpha.
Convergent validity was assessed by means of correlational tests between the CWL
factors and CBCL. An adapted version of Steiger’s (1980) formula was used to conduct
69
4
Chapter 4 | Initial validiation of the Caregiver Wish List
tests for comparing correlation coefficients (Lee & Preacher, 2013), in order to examine
whether there were significant differences in the correlations between CWL factors and
internalizing and externalizing problems.
RESULTS
Preliminary considerations
Before conducting the factor analysis, some preliminary decisions had to be made.
First, the issue of ‘not applicable’ responses in the CWL had to be resolved. As noted
earlier, parents had the option to select this response in case a given situation did not
occur in daily life. We decided to treat these ‘not applicable’ responses as missing values.
As we used robust maximum likelihood estimation for our explorative factor analysis,
all cases with missing values on some of the items could be retained in the analysis.
Meanwhile, items that generated a relatively high (> 10%) percentage of ‘not applicable’
responses were removed from the analysis. The threshold value of 10% was chosen
because until this percentage is reached, most of the traditional methods for dealing
with missing values produce fairly similar results (e.g., see Barzi & Woodward, 2004).
Based on this criterion, three items were removed.
Second, we screened all item distributions in order to identify items that were
excessively skewed. Items for which a score of 4 or 5 was given in 95% of the cases in
one sample and at least 90% of the cases in the other, were excluded from the analyses,
because it can be assumed that these items had little discriminatory power. There were
six items showing such extreme distributions in both samples, which were therefore
excluded. Further, we noticed a number of items that had conspicuously low mean
scores, indicating that the pertinent behaviors were hardly practiced by the parents.
One would expect that the scores on parenting skills in the community sample would
at least be ‘neutral’ (i.e., somewhere around the rating of 3 or higher). To be conservative,
we excluded all items with a mean value lower than 2.50 in the community sample.
Obviously, a relatively low item mean in the community sample does not necessarily
invalidate the item as a diagnostic marker when the corresponding item mean in the
clinical sample is still significantly lower. However, the items that had a mean value
lower than 2.50 in the community sample all had higher means in the clinical sample,
with the exception of one item. This item had a lower mean in the clinical sample, but
did not differ significantly from the mean in the community sample. We therefore
decided to exclude four items, because they seemed to represent behaviors that were
hardly practiced by parents in The Netherlands. In total, 13 items were removed before
conducting the EFA.
70
Initial validiation of the Caregiver Wish List | Chapter 4
Exploratory Factor Analysis
The Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy showed acceptable
values for both the community sample (KMO = 0.67) and the clinical sample (KMO
= 0.64). The initial attempt to conduct EFA for the community sample failed to reach
convergence in the estimation of parameter values. Inspection of the output provided,
showed conspicuous estimates of means and covariances involving four items.
Although failure to reach a convergent solution may have several causes, the most
likely explanation seemed related to low initial communality values for these four
items. We therefore decided to remove these items from the analysis, after which no
further problems in convergence occurred. The EFA in the community sample was
then rerun with the remaining 36 items and also performed for the clinical sample.
Based on the Kaiser criterion (eigenvalues > 1), 12 factors were identified for both the
community and clinical sample. Obviously, this was not a useful criterion. Inspection
of the scree plot suggested a five-, six-, or a seven-factor solution. Inspection of the fit
indices for both the community and clinical sample pointed in the direction of either a
five- or a six-factor solution for both samples. The sixth factor was difficult to interpret
and, therefore, a five-factor solution seemed most appropriate. This resulted in wellinterpretable factors with acceptable RMSEA and SRMR values, but rather low TLI values
(see Table 2). Since the pattern of factor loadings was quite comparable in both samples,
we decided to rerun the analysis for both samples combined. The resulting five-factor
solution produced good fit values for the RMSEA and the SRMR measures, while the TLI
improved but was still below the acceptable limit.
Table 2 | Fit indices of the five-factor solution for the community, clinical, total sample, and final factor solution
based on the included items of the total sample
Fit index
TLI
RMSEA
SRMR
Community sample
.50
.069
.052
Clinical sample
.54
.074
.061
Total sample
.78
.049
.042
Final factor solution total sample
.87
.050
.035
Note. TLI = Tucker Lewis Index; RMSEA = Root Mean Square Error of Approximation; SRMR = Standardized Root
Means Square Residual
71
4
Chapter 4 | Initial validiation of the Caregiver Wish List
Table 3 | Factor loading for items on the Caregiver Wish List for the community, clinical, and total sample
Items
Community Clinical
sample
sample
Total
sample
Factor 1: Adequate discipline
.434
.418
.487
.580
.805
.668
.696
.330
.597
.749
.444
.630
.408
.632
.528
When you tell your child what to do, how often are you in an
unhappy, mad or frustrated mood?
25 When your child misbehaves, how often do you get angry?
.450
.396
.422
.740
.775
.849
26 When your child misbehaves, how often do you yell?
.678
.746
.708
29 When you correct your child/give a consequence — and your child
reacts badly (yelling back, refusing) — how often do you get angry
and let it show?
.375
.581
.557
.457
.231
.514
.490
.700
.444
.459
.405
.513
.529
.410
.437
.406
.698
.518
.070
.479
.374
.499
.673
.643
.550
.787
.689
.394
.431
.354
10 When your child misbehaves, how often do you choose to give no
consequences because you think that your child is too stressed?
11 When your child misbehaves, how often do you choose to give no
consequences because you think your child may have a temper
tantrum?
13 When your child misbehaves, how often do you choose to give no
consequences because you think your child may feel less loved?
14 When your child misbehaves, how often do you choose to give no
consequences because you don’t want to lose your temper?
15 When your child misbehaves, how often do you choose to give
no consequences because it would cause a problem for you (for
example, you would have to stay home)?
Factor 2: Controlled responding
3
Factor 3: Focus on positive behavior
2
Which do you do more—telling your child what NOT to do (STOP
doing a bad behavior) versus telling your child what TO DO (start
doing a good behavior)?
5 When your child behaves well, how often do you praise or
compliment your child?
17 What do you do most—praise your child for good behavior or
correct your child for bad behavior?
49 How often are you able to express positive feelings toward your
child?
Factor 4: Consistency
16 When your child misbehaves, how often do you choose to give NO
consequences because it is hard for you to be strict?
18 When your child misbehaves, how often do you do nothing,
because nothing seems to work?
20 When your child misbehaves, how often do you say you will give
consequences, but don’t follow through?
30 When you correct your child/give a consequence — and your child
reacts badly (yelling back, refusing) — how often do you give your
child what he/she wants?
31 When you correct your child/give a consequence — and your child
reacts badly (yelling back, refusing) — how often do you give up
what you wanted (for example, a cleaned-up room)?
72
Initial validiation of the Caregiver Wish List | Chapter 4
Table 3 | Factor loading for items on the Caregiver Wish List for the community, clinical, and total sample
Items
Community Clinical
sample
sample
Total
sample
Factor 4: Consistency
32 When you correct your child/give a consequence — and your
child reacts badly (yelling back, refusing) — how often do you give
additional consequences, but NOT follow through?
.382
.458
.457
.778
.597
.668
.765
.905
.907
.601
.634
.583
.351
.411
.347
Factor 5: Monitoring
41 How much do you know about your child’s playmates, friends, or
kids he/she hangs out with?
42 How much do you know about the families of these kids — like
where they live, their phone number, and what their parents do?
44 When your child goes out, how much do you know about who he/
she will be with?
45 When your child goes out, how much do you know about what he/
she is doing and where he/she is going?
4
Note. Boldfaced factor loadings are significant at p < .05
In the five-factor solution for the total sample, 13 items had non-significant factor
loadings or factor loadings less than .40 on all factors and were, therefore, not included
in the final solution. The final factor loadings for each included item in the community,
clinical, and total sample are shown in Table 3. Factor 1 consisted of five items and was
named Adequate discipline because all items reflect disciplining behaviors of parents after
misbehavior of their child. All five items were from the original ‘discouraging undesirable
behavior’domain. Factor 2 encompassed four items that referred to Controlled responding
when the child misbehaves (e.g., calm, neutral tone of voice). Again, all items belonged
to the original ‘discouraging undesirable behavior’ scale, except for item 3 which came
from the original scale ‘providing direction and following up’. Factor 3 contained four
items and was labelled Focus on positive behavior. These items were all from different
original domains, but all appeared to tap parents’ responsiveness to positive, prosocial
behavior of the child. Factor 4 was composed of six items reflecting parents following
through after disciplining their child. Therefore, this factor was labeled Consistency.
All items were from the original ‘discouraging undesirable behavior’ domain. The final
and fifth factor was named Monitoring and consisted of four items which were all part
of the original ‘monitoring activities’ scale. Items 17 and 45 had substantial secondary
loadings, but based on the highest loadings they were assigned to Factor 3 and Factor
5, respectively.
73
Chapter 4 | Initial validiation of the Caregiver Wish List
Reliability
The internal consistency coefficients of the five factors were calculated for the
community, clinical, and total sample. As can be seen in Table 4, all Cronbach’s alphas
were in the .61 to .82 range. When adopting an alpha of .70 as cut-off point, it can be
concluded that in the clinical and total sample four of the five factors displayed good
internal consistency. The CWL total score had a good internal consistency in both the
community and clinical sample, as well as in the total sample. In the community sample,
Cronbach’s alphas were in general somewhat lower: only the factors Adequate discipline
and Monitoring had adequate Cronbach’s alphas. Note, however, that all CWL factors
consisted of a small set of items, and that alpha values above .60 for a limited item set
could still be regarded as acceptable.
Table 4 | Internal consistencies (Cronbach’s α) of the extracted factors of the community, clinical, and total sample
Community sample
Clinical sample
Total sample
Adequate discipline
.70
.77
.76
Controlled responding
.63
.71
.71
Focus on positive behavior
.64
.65
.66
Consistency
.61
.72
.73
Monitoring
.75
.73
.76
Total score
.71
.78
.82
Pearson product moment correlations among the five factors were calculated for the
two samples separately and for the total sample (Table 5). Overall, the correlations were
quite low. Two correlations were significant in all three samples, namely the correlation
between Controlled responding and Focus on positive behavior and between Adequate
discipline and Consistency. These correlations were positive, which indicates that parents
who were well able to control their emotions during limit setting were also good in
focusing on positive behavior of the child, and vice versa. Furthermore, parents who did
not allow other factors to influence their limit setting were also more consistent, and
vice versa. It was expected that especially the three factors derived from the original
‘discouraging undesirable behavior’ domain (i.e., Adequate discipline, Controlled
responding, and Consistency) would be correlated, but this was only true for the total
sample.
74
Initial validiation of the Caregiver Wish List | Chapter 4
Table 5 | Pearson correlations among the five extracted factors for the community, clinical, and total sample
1
2
3
4
5
Community sample
1. Adequate discipline
1.00
2. Controlled responding
.03
3. Focus on positive behavior
.03
.26**
1.00
4. Consistency
.41**
.13
.11
1.00
5. Monitoring
.14*
-.06
.17*
.01
1.00
4
1.00
Clinical sample
1. Adequate discipline
1.00
2. Controlled responding
.09
1.00
3. Focus on positive behavior
-.09
.47**
1.00
4. Consistency
.39**
.27**
.14
1.00
5. Monitoring
.05
.09
.04
.14
1.00
Total sample
1. Adequate discipline
1.00
2. Controlled responding
.16**
1.00
3. Focus on positive behavior
.09
.44**
1.00
4. Consistency
.47**
.32**
.26**
5. Monitoring
.15
.10
.18**
1.00
.17**
1.00
* p < .05; ** p < .01
External validity
As stated before, the pattern of factor ladings was roughly similar across the two
samples, motivating us to re-analyze the data on the basis of the total sample, thus
reducing instability in the parameter estimates. Table 6 shows the mean scores and
standard deviations on the factors for both the community and clinical sample. The
difference between both samples was significant for all factors and the CWL total score
(with 23 items included in the factors). On all factors, parents from the community
sample had a significantly higher mean than parents from the clinical sample, indicating
that the parenting skills of parents in the community sample were better than those of
parents in the clinical sample.
75
Chapter 4 | Initial validiation of the Caregiver Wish List
Table 6 | Means (standard deviations) of the community and clinical sample on the extracted factors and their
significant differences
Community sample
M (SD)
Clinical sample
M (SD)
t
p
Adequate discipline
22.77 (2.98)
20.33 (4.79)
5.19
.00
Controlled responding
13.63 (2.90)
11.09 (2.85)
7.89
.00
Focus on positive behavior
15.78 (2.56)
13.62 (2.57)
7.58
.00
Consistency
27.23 (2.80)
23.83 (4.95)
7.11
.00
Monitoring
17.22 (2.07)
16.01 (3.15)
3.90
.00
Total score
96.62 (7.38)
84.85 (10.86)
10.82
.00
Pearson product moment correlations between the five CWL factors and CBCL
internalizing and externalizing subscales were calculated. As can be seen in Table 7, all
CWL factors and the total score were significantly and negatively correlated with both
CBCL internalizing and externalizing scales. Relatively low, but still significant, negative
correlations were found for the CWL factor Monitoring. Compared to the correlations
for the CBCL internalizing scale, the correlations for externalizing were higher, even
significantly higher for three of the five CWL factors, namely Controlled responding,
Focus on positive behavior, and Consistency, as well as for the CWL total score (see Table
7). This result was as expected, since the assessed parenting skills are considered to be
important for improving externalizing behavior problems of the child.
Table 7 | Pearson correlations between the five extracted CWL factors and CBCL internalizing and externalizing
scores and the results of tests comparing the strength of these correlations
Adequate discipline
CBCL Internalizing
CBCL Externalizing
p
-.24
-.30
.13
Controlled responding
-.26
-.41
.00
Focus on positive behavior
-.29
-.40
.02
Consistency
-.23
-.34
.02
Monitoring
-.19
-.18
.43
Total score
-.38
-.52
.00
Note. All correlations were significant at p < .01. CWL = Caregiver Wish List; CBCL = Child Behavior Checklist
DISCUSSION
This study was a first attempt to investigate the psychometric qualities of the CWL,
an interview-based scale for measuring parenting behaviors as defined in the SIL model
(Forgatch et al., 2004). EFA was performed to examine the factor structure of the CWL
using the data of parents from both a community and a clinical sample of elementary
76
Initial validiation of the Caregiver Wish List | Chapter 4
school age children. Furthermore, the internal reliability of the extracted factors
was investigated and we examined whether the CWL factors showed theoretically
meaningful associations with behavior problems of the parents’ children and whether
they were able to differentiate between the community and clinical sample.
The EFA revealed five factors, which were named Adequate discipline, Controlled
responding, Focus on positive behavior, Consistency, and Monitoring. The five-factor
structure provided a reasonable description of the original correlation matrix, according
to two of the conventional fit indices. However, on one fit index, the fit was quite poor.
The factor Adequate discipline measures disciplining behavior of the parents when their
child misbehaves. Controlled responding assesses parents’ responding to misbehavior
of their child. For example, whether the parent stays calm and responds in a neutral
tone of voice. The factor Focus on positive behavior relates to parents’ ability to note and
reinforce desirable behaviors of their child. Consistency reflects the tendency of parents
to follow-through on their expressed consequences, despite aversive responses of their
child to the sanction. Finally, the Monitoring factor assesses parents’ knowledge of their
child’s whereabouts and friendships.
The extracted factors are only partially in line with the domains proposed by the
developer of the CWL. Hodges (2002) had in mind that the CWL would be composed of
six domains of parenting, whereas in the present study only five factors were found. Three
of these five factors, viz. Adequate discipline, Controlled responding, and Consistency,
were entirely composed of items from the original ‘discouraging undesirable behavior’
domain. This finding was hardly surprising as the original ‘discouraging undesirable
behavior’ domain contained about half of all original CWL items which included a
rather heterogeneous set of parenting behaviors all targeting unwanted behaviors of
the child. The factor Monitoring was in keeping with the original ‘monitoring activities’
domain, but contained only four of the original 13 items. The factor Focus on positive
behavior contained items from several CWL domains. The only domain that did not
emerge in the present factor structure was Problem solving. The four original Problem
solving items were discarded during the EFA procedure, indicating that problems
solving did not emerge as a separate and homogeneous parenting factor. Altogether,
the five factors found for the CWL in the present study correspond to four of the five
core parenting practices of the SIL model (i.e., effective discipline, encouragement,
positive involvement, monitoring).
Only 23 of the 53 original CWL items loaded on at least one of the five factors,
indicating that the original CWL can probably be reduced in length without losing
critical content. Despite the limited number of items included in the five factors, the
factors showed adequate internal consistencies, with Cronbach’s alphas ranging
between .61 and .77. The CWL total score, based on the items included in the factors,
77
4
Chapter 4 | Initial validiation of the Caregiver Wish List
had good internal consistency coefficients in the three samples (.71 - .82). Only two
correlations among CWL factors were consistently found in the community, clinical,
and total sample, namely those between Controlled responding and Focus on positive
behavior and between Adequate discipline and Consistency. The correlations between
the other factors were quite low. Especially the correlation between Adequate discipline
and Controlled responding was unexpectedly low, given the fact that these two factors
were derived from the same original CWL domain (i.e., discouraging undesirable
behavior).
The factor structure of the community and clinical samples was highly similar, and
thus the extracted CWL factor structure seemed to be applicable to parents in both
populations. Meanwhile, it was found that the two samples differed in terms of absolute
scores on the five extracted CWL factors and the total score in the expected direction:
parents in the community sample exhibited higher scores and hence better parenting
skills than parents in the clinical sample. Finally, all factors and the CWL total score were
significantly negatively related to internalizing and externalizing behavior of the child:
the better the parenting practices of the parents, the lower the behavior problems.
The factors Controlled responding, Focus on positive behavior, and Consistency were
significantly stronger associated with externalizing problems than with internalizing
symptoms. Since the CWL measures parenting practices that are assumed to be
especially relevant within the context of externalizing behavior problems, these results
can be taken as support for the external validity of this parenting measure.
A few limitations of our study have to be noted. First, the majority of the families
consisted of two biological Dutch parents with steady employment, so it is unknown to
what extent our findings can be generalized to, for example, single parents, stepfamilies
or parents from other ethnic and lower socioeconomic backgrounds. Second, the
sample size was quite small for factor analysis when considering the number of items
included in the original CWL. The criterion for factor analysis is that there should be at
least five cases per item (Hair, Black, Babin, & Anderson, 2009). The CWL contains 53
items, which means that at least 246 participants were needed to meet this criterion.
However, before the final factor analysis was run, 17 items needed to be excluded based
on various criteria, which means that only 36 items were used in the analysis. Thus, a
sample of 180 participants was sufficient. The community sample and the total sample
met this criterion, but the clinical sample (n = 128) did not. Third, there was a difference
in the frequency of endorsement of ‘not applicable’-responses between the community
and the clinical sample. The clinical sample showed more not applicable-responses in
comparison to the community sample. Two possible explanations could be given for this.
It could be that certain parenting situations are less common in a clinical sample than in
a community sample (for example, giving allowance) or it could be related to the person
78
Initial validiation of the Caregiver Wish List | Chapter 4
who interviewed the parents. Parents from the community sample were interviewed by
trained students who were specifically instructed to avoid not-applicable responses and
to stimulate parents to give the best suitable answer. Parents from the clinical sample
were interviewed by trained clinicians who received the same instructions. However,
these parents were recruited and interviewed over a longer period of time, which might
have caused drift in the application of these instructions.
Although the CWL was not specifically developed as an outcome measure for
research on treatment effectiveness, this instrument could be used to examine change
in the specific parenting practices of the SIL model. However, the factor structure found
in the present study should be replicated before more definitive conclusions about the
instrument’s structural validity can be drawn. Furthermore, it needs to be examined if
the extracted factors of the CWL are sensitive to change. In our study evaluating the
effectiveness of Parent Management Training – Oregon model, in which the CWL was
used as an outcome measure, only the original ‘discouraging undesirable behavior’
domain and the CWL total score had adequate internal consistencies and these scales
showed significant change in response to parent training (Thijssen et al., 2015c).
Additionally, future research should explore whether the parenting practices assessed
by the CWL predict the development or maintenance of externalizing behavior
problems in children. The association between CWL self-reported parenting practices
and parenting assessed on the basis of actual parent-child interactions studied in real
life or in the lab should also be examined.
In conclusion, findings from our study suggest that the CWL in its current form needs
to be amended to render it more suitable for research purposes. The a priori factor
structure could not be replicated and several of the extracted factors need additional
items to increase the CWL’s psychometric properties. Furthermore, although the CWL
was designed to assess parenting skills defined by Social Interaction Learning theory,
it failed to assess Problem solving as one of the main SIL skills. Notwithstanding, the
CWL’s interview format provides opportunities for rapport building between parent and
interviewer, and for reflection on parenting skills in terms of strengths and vulnerabilities.
79
4
Callousunemotional
t r a i t s
PART
II
Emotional memory for
central and peripheral
details in children
with callous-
unemotional traits
5
Thijssen, J., Otgaar, H., Meijer, E.H., Smeets, T., & de Ruiter, C. (2012). Emotional
memory for central and peripheral details in children with callous-unemotional traits.
Behavioral Sciences and the Law, 30, 506-515. doi: 10.1002/bsl.2021
Chapter 5 | Emotional memory in children with CU traits
ABSTRACT
A limited number of studies have shown that adults and adolescents with
psychopathic traits suffer from emotional memory impairment. The present study
examined whether this finding could be replicated in a sample of children between 8 and
12 years of age with callous-unemotional (CU) traits. Children with high CU traits (n=24)
were compared with children with low CU traits (n=18) with regard to performance on
a peripheral memory recognition test that examined memory for central and peripheral
components of neutral and negative pictures. Results showed that overall recognition
rates did not differ between the high- and low-CU groups. For negative pictures, both
groups demonstrated better recognition of the central component at the expense of
the peripheral component, while for neutral pictures, the peripheral component was
better recognized than the central component. This study is the first to demonstrate
that children with high CU traits do not suffer from an impaired emotional memory.
84
Emotional memory in children with CU traits | Chapter 5
INTRODUCTION
According to the DSM-IV-TR (American Psychiatric Association, 2000), conduct
disorder (CD) refers to children’s and adolescents’ recurrent and persistent rule-breaking
behavior in which the basic rights of others or major age-appropriate societal norms
or rules are violated. Children with CD, for example, tend to frequently steal, vandalize,
lie, and cause physical harm to other people or animals. A subset of children with CD
is characterized by so-called callous-unemotional (CU) personality traits, which closely
resemble the emotional detachment component of psychopathy in adult forensic
samples (Frick, 2006; Frick, Cornell, Barry, Bodin, & Dane, 2003). Although not all children
with CU traits become psychopaths, they are at greater risk for developing psychopathy
(Fontaine, McCrory, Boivin, Moffitt, & Viding, 2011). Children with CU traits are
characterized by low fear in combination with high impulsivity, are not truly concerned
with other people’s feelings, and typically do not feel bad or guilty when showing
rule-breaking behavior. Furthermore, children with CU traits have more difficulty in
identifying fearful and sad facial expressions (Blair, Colledge, Murray, & Mitchell, 2001;
Stevens, Charman, & Blair, 2001; Woodworth & Waschbusch, 2008) and show reduced
psychophysiological responding to distressing and threatening pictures, films and
words, which indicates reduced affective arousal (Anastassiou-Hadjicharalambous &
Warden, 2008; Blair, 1999; Kimonis, Frick, Fazekas, & Loney, 2006; Loney, Frick, Clements,
Ellis, & Kerlin, 2003). Collectively, these studies show that children with CU traits have a
specific deficit in processing negative emotional stimuli (Frick, 2006).
To date, however, less is known whether this deficit in processing negative stimuli
also impacts memory performance. Generally, emotional events are remembered
better than neutral ones (Buchanan, 2007; LaBar & Cabeza, 2006). Moreover, research
has demonstrated that high levels of arousal enhance memory for negative stimuli in
particular (Cahill & McGaugh, 1995; Steinmetz, Addis, & Kensinger, 2010). Specifically,
central details of an emotional event tends to be well remembered over time at the
expense of memory for peripheral details (i.e., less contextually relevant or spatially
peripheral to the attended event; see e.g. Burke, Heuer, & Reisberg, 1992; Christianson,
1992; Christianson, Loftus, Hoffman, & Loftus, 1991; Heuer & Reisberg, 1990; Levine &
Edelstein, 2009; Otani, Libkuman, Widner, & Graves, 2007).
However, in contrast to normally functioning healthy individuals, people with
psychopathic traits do not seem to exhibit enhanced memory for (negative) emotional
material (Christianson et al., 1996; Dolan & Fullam, 2005, 2010). For example,
Christianson and colleagues (1996) showed that relative to non-psychopathic offenders,
psychopathic offenders did not display enhanced memory for central emotionally
negative details over peripheral details. In a similar vein, Dolan and Fullam (2005) found
that psychopathic offenders were in fact worse at recalling emotional slides than healthy
85
5
Chapter 5 | Emotional memory in children with CU traits
controls (for a replication study among conduct disordered adolescents, see Dolan &
Fullam, 2010). To the best of our knowledge, however, there are no studies investigating
whether children with CU traits also display diminished memory for emotional stimuli. If
one would indeed find impairments in emotional memory in children with CU traits, this
would provide valuable insights into why children with CU traits learn less from negative
emotional experiences, such as being punished for misbehavior (see Dadds & Salmon,
2003). Research has shown that children with CU traits are less responsive to effective
parenting, because they respond with insufficient anxiety when confronted with their
misbehavior (Hipwell et al., 2007; Oxford, Cavell, & Hughes, 2003; Viding, Fontaine,
Oliver, & Plomin, 2009; Vitacco, Neumann, Ramos, & Roberts, 2003; Wootton, Frick,
Shelton, & Silverthorn, 1997). The assumption is that they are less likely to remember
the experience of being punished, which makes them less likely to inhibit their behavior
in future situations in order to avoid the punishment.
Thus, the primary aim of the present study was to examine whether memory for
central and peripheral components of neutral and emotionally negative pictures in
children with high CU traits is impaired relative to children with low CU traits. Based on
previous research, it was expected that memory for central details would be enhanced
at the expense of memory for peripheral details overall (i.e., independent of CU traits or
picture valence). Furthermore, for neutral pictures we expected no differences between
the high and low CU groups with respect to the memory enhancing effect for central
details relative to peripheral ones. However, for negative pictures it was expected that
this central versus peripheral memory ratio would be larger in the low CU group relative
to the high CU group.
METHOD
Participants
The current sample of children was recruited through elementary schools.
Information and consent forms that explained the nature of the study, together with
the Antisocial Process Screening Device (APSD) to assess psychopathic traits in children,
were distributed at elementary schools. Parents were asked to sign the consent form,
fill out the APSD and to return them to the child’s school where they were collected.
Children were allowed to participate only when they assented to the procedure and
parental consent was obtained. This study was approved by the standing Ethical
Committee of the Faculty of Psychology and Neuroscience, Maastricht University.
In total, parents of 77 children between 8 and 12 years of age (mean age = 9.9) filled
out the APSD and signed the consent form. To obtain extreme groups, we selected only
the children with raw scores of 6 or higher on the CU subscale of the APSD as the high
86
Emotional memory in children with CU traits | Chapter 5
CU group (n = 24; 13 male) and the children with CU-subscale scores of 2 or lower as
the low CU group (n = 18; 11 male). These two groups did not differ with respect to
mean age (high CU group: 10.3; low CU group: 9.8; t (40) = -1.15, p = .26) and gender
distribution (χ2 (2) = .68, p = .65). In Table 1, the mean scores of both groups on the APSD
subscales are presented.
Table 1 | Mean scores, standard deviations, and significant differences between groups on the APSD subscales
Low CU group
High CU group
Narcissism
1.72 (1.49)
3.25 (3.22)*
Impulsivity
2.89 (2.17)
4.29 (2.35)
Callous-unemotional
1.33 (.84)
6.83 (1.20)*
APSD total
6.17 (3.92)
17.88 (6.26)*
5
* p < .05
Materials
Antisocial Process Screening Device
The Antisocial Process Screening Device (APSD; Frick & Hare, 2001) is a 20-item
questionnaire aimed to assess psychopathic traits in children and adolescents and is
based on the Psychopathy Checklist-Revised (PCL-R; Hare, 1991). The APSD has to be
completed by the child’s parents or teachers. It consists of three dimensions: callousunemotional traits (6 items), impulsivity (5 items) and narcissism (7 items). All items are
answered with 0 (not at all true), 1 (sometimes true), or 2 (definitely true). Examined in
both community and clinical samples, the internal consistency of the three subscales
ranged from .65 to .85 (Frick, Bodin, & Barry, 2000). The APSD has been found to have
good convergent and construct validity (Vitacco, Rogers, & Neumann, 2003). The Dutch
translation that was used in the present study has also been well-validated (Bijttebier &
Decoene, 2009).
Peripheral Memory Test
To test the memory for central and peripheral components of neutral and negative
pictures, we developed a peripheral memory test (PMT). Stimuli consisted of pictures of
a single object (central component) surrounded by a grey frame in which a red symbol
was presented in each corner (peripheral component). The original stimuli consisted of
3 pictures with a width by height of 259 by 416 pixels, 12 pictures of 416 by 259 pixels,
4 pictures of 529 x 340 pixels, and 1 picture of 416 by 416 pixels. These stimuli were
presented at 50% of their original size on a 15-inch screen with a resolution of 1440 by
900 pixels.
87
Chapter 5 | Emotional memory in children with CU traits
A total of 20 pictures served as the central component; 10 neutral (pictures of a boot,
lighthouse, locomotive, green traffic light, tree, bus, deer, fence, electric power pylon,
and lamp), and 10 negative (aggressive dog, premature baby, hand with stitches, shark,
weapon, black eye, crashed car, bloody knife, scary mask, and needle in an arm). Ten
different symbols were used in the frame, with each symbol occurring once in a neutral
and once in a negative picture, with varying orientation (see Figure 1 for an example).
Figure 1 | Example of a negative picture with symbol used in the PMT.
The PMT consisted of 3 practice trials and 20 test trials (10 with a neutral and 10
with a negative picture). Children were instructed to take a good look at the whole
picture, including the frame, and were asked to try to remember everything. They
were also warned that the picture would be presented only for a short moment. As
can be seen in Figure 2, a fixation cross was presented first for 500 ms. Next, the target
picture plus frame with symbols was presented for 1 sec followed by a black screen for
10 sec. Afterwards, the original target picture was presented together with 3 similar
88
Emotional memory in children with CU traits | Chapter 5
distractor pictures. The child had to identify the correct picture by entering the number
corresponding to the picture (1, 2, 3, 4). Next, the same procedure followed for the
symbol in which the correct orientation of the symbol (left, right, up, down) had to be
chosen. Within each trial, the order of the forced choice recognition of the central and
peripheral components was random, implying that sometimes the central components
were presented first and at other times the peripheral components were presented first.
The intertrial interval was 2 sec. The order of the test trials was also random. The PMT
was run using the Inquisit 3 program.
5
⬀
㔀 洀猀
㄀ 洀猀
㄀ 洀猀
Figure 2 | Timeline of a PMT trial.
Pilot data
To check if the negative pictures were actually experienced as negative and the
neutral ones as neutral, a group of children (n = 21) of 7 to 12 years of age rated the
valence of the pictures on five-point Likert scales using smileys (1= very negative, 2=
negative, 3= neutral, 4= positive, 5= very positive). The mean of the neutral pictures and
negative pictures was 3.43 (SD = .35) and 1.89 (SD= .30), respectively (t (20) = 20.68, p
<. 01, Cohen’s d = 4.76). Furthermore, we checked if either the neutral or the negative
target pictures were easier to recognize than the other. Children had to rate the similarity
of the target picture between the distractor pictures on a five-point Likert scale with
higher scores indicating a higher degree of similarity between the pictures. There was
a significant difference in similarity between the neutral and negative pictures with a
higher mean for the neutral pictures (t (20) = 3.19, p < .01, Cohen’s d = 0.36). However,
89
Chapter 5 | Emotional memory in children with CU traits
when piloting the PMT, children did not show longer reaction times for the recognition
of neutral pictures. The mean reaction time was 5.89 seconds (SD = 1.94) for the neutral
pictures and 6.09 seconds (SD = 2.00) for the negative pictures. This difference was not
significant (t (9) = -.44, p = .67).
Design and Procedure
The present experiment was a 2 (Valence: neutral vs. negative) × 2 (Detail: central vs.
peripheral) × 2 (Group: low CU vs. high CU) split plot design, with the last factor as the
between subjects factor.
The selected children were seated comfortably in front of a computer in a quiet room
at the child’s school. The PMT was always administered in the afternoon to control for
potential time of day effects on, for example, fatigue, attention and concentration. The
experimenter explained the test by reading the instructions presented on the computer
together with the child. After the three practice trials, the experimenter asked the child
whether the procedure of the test was clear before starting with the test trials. The PMT
took between 10 and 15 minutes. Afterwards, children were given a small present in
return for their participation.
RESULTS
Figure 3 shows the mean proportion of correctly recognized central and peripheral
components for the neutral and negative pictures for the two groups. To examine these
differences in correct recognition, a repeated measures Analysis of Variance (ANOVA)
was conducted with Group (high CU group vs. low CU group) as the between subjects
factor and Valence (neutral vs. negative) and Detail (central vs. peripheral) as within
subjects factors. The Group x Valence x Detail interaction was not statistically significant
(F (1, 40) = .01, p = .94). However, a significant interaction between Valence and Detail
was found (F (1, 40) = 34.87, p < .01, ηp2 = .47). To identify the differences between levels of
valence and detail, simple effects analyses were performed. For the neutral pictures, the
peripheral component was significantly better recognized than the central component
(t (41) = -3.03, p < .01, Cohen’s d = -.68). However, for the negative pictures, the central
component was significantly better recognized than the peripheral component (t (41)
= 4.15, p < .01, Cohen’s d = .93). Furthermore, there was a significant difference between
the central components for the neutral and negative pictures. The central component
was better recognized for negative pictures than for neutral pictures (t (41) = -5.32, p <
.01, Cohen’s d = -1.03). The peripheral component was significantly better recognized
for the neutral pictures than for the negative pictures (t (41) = 3.19, p < .01, Cohen’s d =
.64). None of the other interactions, nor main effects reached significance.
90
Emotional memory in children with CU traits | Chapter 5
Differences in latency were also examined, using a repeated measures ANOVA with
Group (high CU group vs. low CU group) as the between subjects factor and Valence
(neutral vs. negative) and Detail (central vs. peripheral) as within subjects factors. Again,
the Group x Valence x Detail interaction was not significant (F (1, 40) = 2.34, p = .13).
Only a significant main effect was found for detail (F (1, 40) = 34.35, p < .01, ηp2 = .49)
with longer reaction times for the central components (M = 5783 ms) compared to
the peripheral components (M = 4398 ms). No other main effects or interactions were
significant.
5
1
0,9
mean proportion correct
0,8
0,7
0,6
low CU group
0,5
high CU group
0,4
0,3
0,2
0,1
0
central
peripheral
neutral
central
peripheral
negative
Figure 3 | Mean proportion
correctly recognized central and peripheral components for the neutral and
Figure3.Meanproportioncorrectlyrecognizedcentralandperipheralcomponentsfortheneutral
negative pictures.
andnegativepictures.
DISCUSSION
Previous studies found that adults and adolescents with psychopathic traits have
a memory impairment for emotional stimuli (Christianson et al., 1996; Dolan & Fullam,
2005, 2010). The present study examined whether this finding could be replicated in a
non-clinical sample of children high and low on CU traits. Interestingly, the results of
this study showed no difference in emotional memory between the high and low CU
groups.
Our PMT was successful in replicating the peripheral memory effect found in
previous studies (e.g., Burke et al., 1992; Christianson, 1992; Christianson et al., 1991;
Heuer & Reisberg, 1990; Levine & Edelstein, 2009; Otani et al., 2007). For the neutral
pictures, the peripheral component was better recognized than the central component,
91
Chapter 5 | Emotional memory in children with CU traits
while the central component was better recognized than the peripheral component for
the negative pictures. Interestingly, the peripheral component was better recognized
for the neutral pictures than for the negative pictures. The ten symbols used as
peripheral components were the same for both the neutral and negative pictures. Only
the orientation of the symbol varied. Therefore, the difference in recognition of the
peripheral components could only be explained by the valence of the central picture.
Attention is being drawn to the negative picture, at the expense of the attention paid
to the symbol.
Generally, central details receive more attention than peripheral details, which
would lead to a better memory for the central details than for peripheral details (Riggs,
McQuiggan, Farb, Anderson, & Ryan, 2011). In contrast to our expectations, it was found
that the peripheral component was better recognized than the central component
of neutral pictures. In the present study, children were explicitly instructed to look
at both the picture (central component) and the frame with the symbols (peripheral
component). When both components receive the same amount of attention, it
seems logical that the peripheral component was better recognized than the central
component of neutral pictures since the central picture is more complex than a simple
symbol. For emotionally negative pictures, attention is drawn to the central component
despite the instruction to look at both components. Therefore, the central component
receives more attention than the peripheral component which leads to a better memory
for the central component at the expense of memory for the peripheral component
(Glass & Newman, 2009). Although attention is not the only condition contributing to
better memory for emotional details, it is a necessary one (Christianson et al., 1991;
Kramer, Buckhout, & Eugenio, 1990).
A possible explanation for not finding a difference in emotional memory between
the high and low CU groups could be that the impairment in emotional memory has not
developed yet in children of this age. Dolan and Fullam (2010) suggest that the deficit
in emotional memory in people with psychopathic traits is reasonably stable across
the lifespan. However, this does not seem to be the case for children with CU traits. In
a recent study from our lab, we also found that children with low and high CU traits
did not differ on true recall for both neutral and negative word lists (Thijssen, Otgaar,
Howe, & de Ruiter, 2013). Salekin, Debus, and Jackson (2008) noted that with regard
to performance tasks, the observed association between emotional processing deficits
and psychopathy in adults is not as robust in children. Therefore, the disorder might not
fully develop until adulthood and the impaired emotional processing in children with
high CU traits may thus have not affected memory for emotional material yet.
Previous studies focusing on central and peripheral memory have mainly used
pictures from the International Affective Picture System (IAPS; Lang, Bradley, & Cuthbert,
92
Emotional memory in children with CU traits | Chapter 5
1998). However, the problem with such studies is that it is difficult to ascertain what the
central and peripheral details exactly are. In our PMT, this distinction was clear and it was
easy to manipulate the peripheral details. A limitation of the PMT could be the central
pictures which were used. Since different pictures were used in the neutral and negative
condition, it cannot be ruled out that the negative pictures indeed might be easier to
recognize than the neutral pictures. In the pilot study, it was found that the distractor
pictures for the negative pictures were less similar than the distractor pictures for the
neutral pictures, which indicates that the negative pictures are easier to distinguish
than the neutral pictures.
Several possible explanations could be given for why the present study was not able
to replicate previous findings regarding emotional memory of adolescents and adults
with psychopathic traits (Christianson et al., 1996; Dolan & Fullam, 2005, 2010). First, this
might have to do with that the present study used a non-clinical sample. Specifically,
while the previous studies used clinical samples, the low and high CU groups in the
current study were selected from the general population. Thus, even though our groups
differed significantly in CU traits, the high CU group may nonetheless be difficult to
compare with a high CU group selected from a clinical population in which CU traits are
expected to be more extreme. Second, previous studies used a different memory task
in that they used a slide show together with a short narrative to measure memory for
central and peripheral details, while in the present study unrelated pictures were used.
Although speculative, it could be that in the previous studies, the memory test with
narratives evoked more relational processing (Hunt & Einstein, 1981), while our memory
test in the current study likely elicited more item-specific processing. Since persons with
psychopathic traits seem to have deficits in their relational processing of emotional
memories, one might argue that our memory test was not sensitive enough to detect
these relational processing deficits. Finally, our sample size was fairly small, which may
have led to insufficient power to detect small between group differences.
In sum, the present study did not find a difference in emotional memory between
children with high and low CU traits. However, we did show that both groups had a
better memory for the central component of negative pictures at the expense of the
peripheral component, while their memory for neutral pictures was better for the
peripheral component than for the central component. This finding suggests that the
PMT is a useful tool to examine the memory narrowing effect found in previous studies.
Moreover, based on the results of the present study, children with CU traits do not seem
to have a deficit in emotional memory. This suggests that impairments in emotional
memory may not be an explanation for why children with CU traits are less responsive
to effective parental correction.
93
5
Emotional true and
false memories in
children with callous -
unemotional traits
6
Thijssen, J., Otgaar, H., Howe, M.L., & de Ruiter, C. (2013). Emotional true and false
memories in children with callous-unemotional traits. Cognition & Emotion, 27,
761-768. doi: 10.1080/02699931.2012.744300
Chapter 6 | True and false memories in children with CU traits
ABSTRACT
Several studies have found that children with callous-unemotional (CU) traits have
a deficit in processing emotionally negative material. The present study examined
whether this deficit also affects emotional memory. Twenty-two children with low CU
traits and 24 children with high CU traits between 8 and 12 years of age were selected
from a community sample and presented with neutral and negative emotional words,
using the Deese-Roediger-McDermott paradigm. On true recall, there was no difference
between the groups. Both groups had higher true recall rates for the neutral word
lists than for the negative lists. However, on false recall, although there were no group
differences for neutral word lists, the high CU group recalled significantly fewer critical
lures on the negative word lists than the low CU group. Furthermore, the high CU group
had significantly less false recall on the negative word lists compared to the neutral
word lists, while the low CU group showed no difference in false recall between the
word lists. These results indicate that children with high CU traits have no deficiencies
in true memory performance, yet are less susceptible to developing false memories
concerning emotionally negative material.
96
True and false memories in children with CU traits | Chapter 6
INTRODUCTION
Some children with conduct disorder are characterized by so-called callousunemotional (CU) traits, which closely resemble the emotional detachment component
of psychopathy in adult forensic samples (Frick, 2006). Children with CU traits display
low fear in combination with high impulsivity, are not truly concerned with others’
feelings, and typically do not feel bad or guilty when showing rule-breaking behavior.
Like adults with psychopathic traits, children with CU traits have deficits in emotional
processing. Children with CU traits seem to have more difficulty in identifying fearful
and sad facial expressions compared to expressions of surprise, happy, disgust,
and anger (Blair, Colledge, Murray, & Mitchell, 2001) and may be less physiologically
responsive to distressing and threatening stimuli (Blair, 1999). This indicates that the
deficit in emotional processing in children with CU traits is specific to negative stimuli
(Frick, 2006).
There is evidence implying that this deficit can lead to an impaired memory for
negative emotional material, while emotional events generally are remembered better
than neutral ones (LaBar & Cabeza, 2006). For example, adults and adolescents with
psychopathic traits appeared to be worse at recalling negative emotional slides than
healthy controls (Dolan & Fullam, 2005, 2010). If these impairments in emotional
memory could also be found in children with CU traits, it could explain why these
children do not learn from negative emotional experiences. For example, research has
shown that children with CU traits are less responsive to parental limit setting, because
they respond with insufficient anxiety when confronted with their misbehavior (Oxford,
Cavell, & Hughes, 2003; Wootton, Frick, Shelton, & Silverthorn, 1997). Therefore, they are
less likely to remember the experience of being punished after their misbehavior, which
makes them less likely to inhibit their behavior in future situations in order to avoid
punishment (see Dadds & Salmon, 2003). To our knowledge, no studies have examined
emotional memory in children with CU traits.
Interestingly, it is likely that the deficiency in emotional processing found in
individuals with psychopathic traits will not only affect accurate remembering, but also
incorrect remembering (i.e., false memories). A robust paradigm to examine both true
and false memory is the Deese-Roediger-McDermott (DRM) paradigm (Deese, 1959;
Roediger & McDermott, 1995). In this paradigm, participants are presented with a list of
words which are all semantically related to a non-presented theme word that is called
the critical lure. A false memory occurs when the critical lure is incorrectly recalled or
recognized as being presented in the word list (see for an overview, Brainerd, Reyna, &
Ceci, 2008a).
The development of false memories elicited by the DRM paradigm could be explained
by the associative activation theory (AAT; Howe, Wimmer, Gagnon, & Plumpton, 2009)
97
6
Chapter 6 | True and false memories in children with CU traits
which is partly founded on activation monitoring theory (Roediger, Watson, McDermott,
& Gallo, 2001). According to the AAT, the processing of one word leads to the spreading
activation of corresponding nodes or concepts in our mental lexicon (i.e., knowledge
base). This process can also lead to the activation of the critical lure. AAT assumes that
the development of false memories is predominantly the result of increases in the
amount and strength of associative relations as well as the speed and automaticity with
which these associations are accessed and activated (Wimmer & Howe, 2009). Because
associated relatedness is higher among emotionally negative material than among
neutral material (Howe, Candel, Otgaar, Malone, & Wimmer, 2010; Howe et al., 2009; Talmi
& Moscovitch, 2004), AAT assumes that spreading activation is more automatic (faster)
when negative material is encountered than when neutral material is experienced.
This would lead to an increased risk of false memories for negative material. Indeed, a
number of studies have found higher false recognition for negative information than for
neutral information (e.g., Brainerd, Stein, Silveira, Rohenkohl, & Reyna, 2008b) although
some studies showed that false recall is also higher for neutral than negative emotional
word lists (e.g., Howe et al., 2010).
The aim of the present study was to examine whether emotional memory in children
with high CU traits differs from children with low CU traits. The DRM paradigm was
used to test true and false recall for neutral versus negative word lists. Since emotional
events are generally remembered better than neutral ones and AAT hypothesizes more
spreading activation for negative material, it was predicted that children with low CU
traits would have better true recall for negative rather than for neutral word lists. We did
not expect a difference between low and high CU children for neutral word lists. Children
with high CU traits are less likely to benefit from enhanced memory for negative material
than children with low CU traits, because of their deficit in emotional processing (Frick,
2006). Therefore, we hypothesized that the high CU group would recall fewer negative
words than the low CU group. This is also in accordance with the expectations of the AAT,
which implies that it is likely that children with high CU traits would process negative
information less automatically because they are trying to inhibit access to that material
more than children with low CU traits (Howe, Toth, & Cicchetti, 2011).
Based on AAT, one would expect that children with low CU traits would show
increased false memories for the negative word lists, because of the heightened
spreading activation. Again, no difference between the high and low CU groups was
expected for the neutral word lists. For the negative word lists, children with high
CU traits would be expected to have fewer false memories, because the automatic
activation spreads slower than in children with low CU traits.
98
True and false memories in children with CU traits | Chapter 6
METHOD
Participants
The current sample of children between 8 and 12 years of age was recruited through
elementary schools as part of a larger research project (see also Thijssen, Otgaar,
Meijer, Smeets, & de Ruiter, 2012). Children were allowed to participate only when they
assented to the procedure and parental consent was obtained. Information and consent
forms which explained the nature of the study, together with the Antisocial Process
Screening Device (APSD) to assess psychopathic traits in children, were distributed at
the elementary schools. Parents were asked to sign the consent form, fill out the APSD
and to return them to the child’s school where they were then collected. In total, data
from 111 children were obtained. This study was approved by the standing Ethical
Committee of the Faculty of Psychology and Neuroscience, Maastricht University.
To obtain extreme groups, we selected only the children with scores of 6 or higher
on the CU subscale of the APSD as the high CU group (n = 24; 13 male) and the children
with CU subscale scores of 2 or lower as the low CU group (n = 22; 13 male). These
scores were selected based on the means found for clinical and community samples in
previous studies. For clinical samples, means of around 6 on the CU-subscale are found
(e.g., Fite, Greening, Stoppelbein, & Fabiano, 2009). For community samples, means of
around 2 on the CU-subscale are found (e.g., Dadds, Fraser, Frost, & Hawes, 2005). The
two groups did not differ with respect to mean age (high CU group: 10.3; low CU group:
9.7; t (44) = -1.55, p = .13) and gender distribution (χ2 (2) = .97, p = .62).
Materials
Antisocial Process Screening Device
The Antisocial Process Screening Device (APSD; Frick & Hare, 2001) is a 20-item
questionnaire to assess traits of psychopathy in children (and adolescents). The
APSD has to be completed by the children’s parents or teachers. It consists of three
dimensions: callous-unemotional (6 items), impulsivity (5 items) and narcissism (7
items) with all items answered with 0 (not at all true), 1 (sometimes true), or 2 (definitely
true). Examined in both community and clinic samples, the internal consistency of the
three subscales ranged from .65 to .85 (Frick, Bodin, & Barry, 2000). The APSD has been
found to have good convergent and construct validity (Vitacco, Rogers, & Neumann,
2003). The Dutch translation that was used in the present study has also been validated
(Bijttebier & Decoene, 2009).
99
6
Chapter 6 | True and false memories in children with CU traits
Deese-Roediger-McDermott paradigm
Five neutral and five emotionally negative word lists were used. The critical lures
of the neutral lists were bread, window, sweet, smoke, and foot. The critical lures of the
negative word lists were murder, pain, punishment, death, and cry. Each list was 10 items
long. These word lists have been used in previous research (see Howe et al., 2010; Otgaar,
Peters, & Howe, 2012). The 10 word lists were orally presented through a computer at
a 3-second rate. Half of the children received the neutral word list first followed by the
negative lists, while the other half first received the negative word list. After a word list
had been presented, the child had to do a distractor task (circling the letter X in a string
of letters). Then the child had to recall as many words he/she could remember from the
word list. This procedure was repeated until all 10 word lists had been presented.
Design and Procedure
The present experiment was a 2 (valence: neutral vs. negative) × 2 (group: low CU
vs. high CU) split plot design, with the latter factor as the between subjects factor. The
selected children were tested individually in a quiet room at their school. They were
told they would hear a number of words through the laptop which they should try
to remember. The DRM took about 15 minutes and was always administered in the
morning to control for time of day effects on attention and concentration. Afterwards,
children were given a small present in return for their participation.
RESULTS
Separate repeated measures analyses of variance (ANOVAs) were conducted for
proportion true recall, recall of critical lures, recall of unrelated lures, and net accuracy
with VALENCE (neutral vs. negative) as within subjects factor and GROUP (high CU
group vs. low CU group) as between subjects factor. For both groups, Table 1 shows the
mean proportion of true recall, false recall, recall of unrelated lures, and net accuracy on
the neutral and negative word lists.
For true recall, no significant interaction between valence and group was found (F
(1, 44) = 1.56, p = .22). However, there was a significant main effect of valence (F (1, 44)
= 13.84, p < .05, ηp2 = .24). True recall was higher for the neutral word lists than for the
negative word lists (see Table 1).
For false recall, there was a significant interaction between valence and group (F (1,
44) = 5.79, p < .05, ηp2 = .12). Simple effects analyses revealed that there was no significant
difference in critical lures for the low CU group between the neutral and negative word
lists (F (1, 21) = 2.01, p = .17). However, for the high CU group, there was a significant
difference between the neutral and negative word lists with fewer critical lures for the
negative word lists than the neutral word lists (F (1, 23) = 4.31, p < .05, ηp2 = .16).
100
True and false memories in children with CU traits | Chapter 6
Table 1 | Mean proportions and standard deviations per group for true recall, false recall, recall of unrelated lures,
and net accuracy
Low CU group
High CU group
Neutral
Negative
Neutral
Negative
True recall
.51 (.11)
.47 (.11)
.56 (.11)
.49 (.10)
False recall
.26 (.22)
.35 (.25)
.30 (.21)
.20 (.19)
Unrelated lures
.07 (.07)
.10 (.09)
.08 (.07)
.08 (.05)
Net accuracy
.70 (.20)
.64 (.22)
.68 (.18)
.75 (.21)
Note. CU = callous-unemotional
Furthermore, independent samples t-tests showed no significant difference in critical
lures between the groups on the neutral word lists (t (44) = -.57, p = .57). However, the
difference between the groups on the negative word lists was significant with the high
CU group recalling significantly less critical lures than the low CU group (t (44) = 2.37, p
< .05). Figure 1 shows the proportion critical lures recalled for the neutral and negative
word lists per group.
0,45
0,4
Mean proportion false recall
0,35
0,3
0,25
Neutral
Negative
0,2
0,15
0,1
0,05
0
Low CU group
High CU group
Figure 1 | Mean proportion false recall on neutral and negative word lists in children with high and low CU
Figure1.Meanproportionfalserecallonneutralandnegativewordlistsinchildrenwithhighand
traits.
lowCUtraits.
the unrelated lures, which is the proportion of all recalled words that were
For
unrelated words (i.e., not in the word list and not the critical lure), no significant
interaction (F (1, 44) = 2.95, p = .09) or main effect was found (F (1, 44) = 1.51, p = .23).
101
6
Chapter 6 | True and false memories in children with CU traits
Furthermore, we examined any group differences in net accuracy (true recall/true recall
+ false recall; see Otgaar et al., 2012). Again, no main effect (F (1, 44) = .02, p = .88) or
significant interaction was found (F (1, 44) = 2.87, p = .10).
Since previous studies have found a difference in performance on the DRM paradigm
between younger and older children (e.g., Brainerd, Holliday, Reyna, Yang, & Toglia,
2010; Brainerd, Reyna, & Zember, 2011), an exploratory repeated measures ANOVA was
performed with VALENCE (neutral vs. negative) as within subjects factor and GROUP
(high CU group vs. low CU group) and AGE (8/9 years vs. 11/12 years) as between
subjects factors. For this analysis, there were 8 8/9-year old children with low CU traits, 6
8/9 year old children with high CU traits, 6 11/12 year old children with low CU traits, and
12 11/12 year old children with high CU traits. No significant interactions were found
for true recall and the recall of unrelated lures. However, there was a significant threeway interaction for false recall (F (1, 28) = 6.87, p < .05, ηp2 = .20). Simple effects analyses
showed only one significant difference between neutral and negative word lists, namely
for the 11/12 year old children with low CU traits. These children recalled significantly
more negative than neutral critical lures (F (1, 5) = 45, p < .05, ηp2 = .90).
DISCUSSION
Using the DRM paradigm, the present study examined whether there is a difference
in true and false memories for neutral and negative word lists between children with
high and low CU traits. Our findings can be summarized as follows. First, the two groups
did not perform differently on true recall. For both groups, true recall was higher for the
neutral word lists than for the negative word lists. This finding is not as predicted, but
consistent with previous studies (Howe et al., 2010; Otgaar et al., 2012). Interestingly, we
did not find that children with high CU traits differed in their true recall on the negative
word lists compared with children with low CU traits. Therefore, the present results do
not suggest that children with high CU traits have an impaired emotional memory. This
is in contrast to previous studies on adolescents and adults with psychopathic traits
(Dolan & Fullam, 2005, 2010), but consistent with findings from a recent study from our
lab. In this study, children high and low on CU traits performed similarly on a task where
they had to recognize central and peripheral components in neutral and emotionally
negative pictures (Thijssen et al., 2012).
For false recall, there was no difference between the high and low CU groups on
the neutral word lists. However, there was a difference between the groups on the
negative word lists: the high CU group recalled less critical lures than the low CU group.
Moreover, the low CU group showed no difference in false recall between the neutral
and negative word lists. This finding is not consistent with previous research in which
102
True and false memories in children with CU traits | Chapter 6
a difference between the word lists was found (e.g., Brainerd et al., 2008b; Howe et al.,
2010). This could be due to the small sample size, which may have led to insufficient
power to detect a significant difference. However, it should be mentioned that mixed
findings concerning the effect of valence on false memories have been reported in
previous studies (see Brainerd & Reyna, 2012).
The high CU group showed a significant difference in false recall between the word
lists. These children had fewer false memories on the negative word lists compared to
the neutral word lists. This result on false recall is in accordance with the predictions of
AAT. Children with high CU traits have difficulties in automatically processing negative
emotional material. Furthermore, as children with high CU traits are found to process
negative material differently than children with low CU traits (e.g., Blair, 1999; Blair et
al., 2001), it is equally likely that the associative networks of children with high CU traits
related to negative material are not as well-integrated and dense relative to children
low on CU traits. As a consequence, the flow of information in such a network in high CU
traits children will be less automatic and slower. Indeed, our study demonstrated lower
negative false memory rates in children with high CU traits compared to children with
low CU traits.
Since previous research has found an age effect in DRM-performance (e.g., Brainerd
et al., 2010; Brainerd et al., 2011), the low and high CU children in the present study were
divided by age. Results showed that 11/12-year old children with low CU traits recalled
significantly more critical lures for the negative word lists compared to the neutral word
lists. However, these results should be interpreted with caution, because the number of
children per group was very small.
The following limitations deserve some comment. First, we used a community
sample for our study. Children of both the low and high CU groups were selected from
the general population. In the present study, no differences on true recall were found
between the low and high CU group. It is possible that there would be a difference in
true recall when the high CU group is selected from the clinical population. However,
our findings concerning false recall could be even stronger for the high CU group when
selected from the clinical population. Second, only one source of information (i.e.,
the parent) was used to measure CU traits in the children. It would have been better
if additional measures of CU traits were used. Third, only five word lists per valence
were used. Especially in our relatively small sample, this could create little variability
in performance and limit the generalizability. To enhance the generalizability of the
findings, it would be advisable to repeat this study with a larger set of (clinical) children
including more word lists or other types of false memory measures (e.g., misinformation
paradigm; Loftus, 2005). Fourth, the present study did not control for arousal. There is
some evidence suggesting that false memories are related to arousal level (e.g., Corson &
103
6
Chapter 6 | True and false memories in children with CU traits
Verrier, 2007). However, in studies with children, valence tends to play a more important
role than arousal (i.e., most memory effects are driven by changes in valence not arousal)
when these factors are varied orthogonally using the DRM paradigm (e.g., see Brainerd
et al., 2010). Finally, one may wonder whether our findings can be translated to forensic
contexts. That is, can results obtained from semantically-related word lists provide us
with critical information about events that people experience in daily life? Although it
has been debated whether the DRM paradigm might be useful in the legal domain (e.g.,
Brainerd et al., 2011), research shows that the DRM paradigm is a robust tool to examine
the mechanism behind memory illusions. Even more important, studies show that the
DRM illusion is related to autobiographical memory (e.g., Gallo, 2010).
In sum, the present study found that children with high CU traits do not differ from
children with low CU traits concerning true recall, which implies that the deficit in
processing emotionally negative material does not affect correct emotional memory
in children with CU traits. However, for false recall, children with high CU traits recalled
fewer not presented words at the negative word lists than children with low CU traits.
This indicates that children with high CU traits are better at differentiating between
true and false memories concerning negative material than children with low CU traits.
Children with CU traits are more likely to encounter negative emotional situations
because of their aggressive/antisocial problem behavior. Our study shows that they are
less likely to spontaneously, falsely report elements related to these negative situations.
Conclusively, it seems that having high CU traits lowers the risk for inaccurate memories,
while leaving true emotional memory untouched.
104
General
discussion
7
General discussion | Chapter 7
Aims of the present dissertation
This dissertation examined the role of parenting strategies in children with
externalizing behavior problems. Research has demonstrated that effective parenting
practices are an important mechanism contributing to the development of prosocial
behavior in the child (DeGarmo & Forgatch, 2005, 2007; Forgatch & DeGarmo,
1999; Martinez & Forgatch, 2001; Ogden & Amlund-Hagen, 2008). Therefore, many
interventions for childhood behavior problems are parent-based interventions, focusing
on the promotion of effective parenting skills and a decrease of ineffective child rearing
behaviors. In part I of this thesis, the effectiveness of Parent Management Training –
Oregon model (PMTO) was evaluated at five child service agencies in The Netherlands.
The Dutch government requires that the effectiveness of interventions is also evaluated
in The Netherlands, before they will be entered into the Dutch youth institute (NJi)
database of effective interventions. In Chapter 2, the effects of PMTO were compared to
Care As Usual (CAU). Chapter 3 examined the association between treatment integrity of
the PMTO therapists and treatment outcome of PMTO. In Chapter 4, the factor structure
of one of the key measures of parenting practices used in the effectiveness study, the
Caregiver Wish List (CWL), was examined. The aim of Part II was to examine emotional
memory in children with callous-unemotional (CU) traits. These traits are characterized
by a persistent pattern of behavior that reflects a disregard for the feelings of others as
well as a general lack of empathy. Previous research has shown that children with CU
traits display highly persistent and serious antisocial behavior problems (Frick, Cornell,
Barry, Bodin, & Dane, 2003; Frick, Stickle, Dandreaux, Farrell, & Kimonis, 2005) and also
exhibit clear deficits in emotional processing, which could become manifest in their
emotional memory. The study presented in Chapter 5 examined whether a deficit in
emotional memory could be found in children high in CU traits. Furthermore, it was
explored whether deficits in emotional processing could also affect emotional false
memory (Chapter 6). The main findings of the studies described in this dissertation will
be discussed below. We will address methodological issues and implications for future
research and clinical practice.
The effectiveness of PMTO
Because longitudinal research has demonstrated a high degree of stability of
conduct problems in childhood and a possible aggravation into criminal and violent
behavior in adulthood (Moffitt, 1993; Nixon, 2002), early interventions seem to be of the
highest importance. In The Netherlands, there existed a need for effective treatment
programs for children 4-12 years with antisocial conduct problems. Therefore, the
Ministry of Health, Welfare and Sport decided in 2005 to fund the implementation of
Parent Management Training Oregon model (PMTO). PMTO is a theory-driven, evidence-
107
7
Chapter 7 | General discussion
based therapy for parents of children between 4 and 12 years of age who have serious
externalizing behavior problems. The intervention has proven to be effective in the
US, Norway, and Iceland (e.g., Forgatch & DeGarmo, 1999; Ogden & Amlund-Hagen,
2008; Sigmarsdóttir, Thorlacius, Guðmundsdóttir, & DeGarmo, 2014), but it cannot be
automatically assumed that it would also be effective in The Netherlands. According
to the Social Interaction Learning (SIL) model, parents have a major influence on the
development of their child (Forgatch, Bullock, & Patterson, 2004). By use of effective
parenting, parents stimulate a healthy, prosocial development in their children. However,
when parents use ineffective parenting strategies (i.e., coercion) they stimulate the
development of antisocial behavior. Therefore, parents are the main agents of change in
PMTO. The assumption is that by improving the parenting skills, the behavior of the child
will also improve. The study presented in Chapter 2 examined whether PMTO is more
effective in reducing children’s problem behaviors, by increasing the use of effective
parenting practices, and reducing parental psychopathology and stress, as compared
to CAU. In this study, 146 clinically referred children aged 4 to 11 years and their parents
were included of which 91 received PMTO and 55 CAU. Outcome data were collected via
parent- and teacher-rated questionnaires and by means of observations of structured
parent-child interactions. Families were assessed at four time points: pretreatment, and
after 6, 12, and 18 months. Results showed no statistically significant differences in the
effectiveness of the two interventions on the primary outcome measures relating to
parent-reported externalizing behaviors. That is, both treatment conditions showed
a significant decrease in parent-rated externalizing behavior problems over time.
Also on parenting stress and parental psychopathological symptoms, no significant
differences between PMTO and CAU were found, with parents experiencing significantly
less parenting stress and psychopathological symptoms over the course of both
interventions. For observed parenting skills, no significant effects were documented.
For self-report, the only parenting practice that significantly improved for both PMTO
and CAU was the use of discouragement of undesirable behavior (i.e., discipline). For
both conditions, improvements were most evident during the first six months of the
study and then remained stable until 18 months after baseline. These findings imply
that PMTO is effective in a clinically referred sample of children with externalizing
problems in the Netherlands, but not more effective than CAU.
When examining factors that may impact the effect of PMTO, we found that the
treatment effect was moderated by the level of externalizing behavior problems at
baseline: especially children with serious externalizing behavior problems appeared to
benefit more from PMTO. One could argue that children with higher levels of problem
behavior might be more difficult to treat. However, our results suggest the opposite,
which may have been due to the fact that there was simply more room for improvement
108
General discussion | Chapter 7
in these children. However, a recent review by Shelleby and Shaw (2014) showed that
this finding was demonstrated in more studies evaluating parenting interventions,
such as Incredible Years and Family Check-Up. These findings together demonstrate
that children with higher levels of antisocial behavior problems, which are thought
to be at greater risk for the persistence and exacerbation of these behaviors (Broidy
et al., 2003; Nagin & Tremblay, 1999), benefit from early intervention. Improvements
in behavior problems were only documented for parent ratings but not for teacher
ratings, which is a finding not uncommon in research on parenting interventions for
children with externalizing behavior problems (e.g., Kjøbli, Hukkelberg, & Ogden,
2013; Kjøbli & Ogden, 2012). One explanation for this result might be that children’s
behavior problems are less apparent in the more structured school setting and that,
therefore, change was less noticeable for the teachers. Indeed, our data indicated that
teachers overall reported lower levels of behavior problems compared to parents. This
is also in accordance with previous empirical findings showing a fairly low agreement
between parents and teachers with regard to these problem behaviors, with teachers
reporting lower behavior problem scores as compared to parents (Youngstrom, Loeber,
& Stouthamer-Loeber, 2000). Another explanation could be that parents and teachers
observe children’s problem behavior in different settings and roles. Teachers are more
likely to make norm-referenced assessments by comparing the behavior problems
of the child to behaviors of other children. Parents on the other hand are more likely
to make ipsative assessments, where the children are compared to themselves over
time (Hukkelberg & Ogden, 2013). Notwithstanding these considerations, the most
conservative interpretation of our results is that the positive treatment effects did not
generalize to the school setting and that these parent-based interventions may only be
effective in the context where they are implemented (i.e., at home).
This was the first study that showed that PMTO did not produce better effects
compared to a treatment control group. Although most of the earlier studies conducted
in the US compared PMTO to a waiting list control condition, families included in the
control condition of studies performed in Norway and Iceland also received treatment.
When comparing our effect sizes for PMTO and CAU with those obtained in the other
European studies, the effect size found for PMTO in our study was as high (Norwegian
study) or even higher (Icelandic study). However, the effect size of CAU in our study
seemed to be somewhat higher than the effect sizes of CAU in Norway and Iceland, which
indicates that the regular treatments offered to children with externalizing behavior
problems in The Netherlands appear to be of relatively good quality. This is likely the
result of the fact that many psychologists in our country are trained to apply cognitivebehavioral techniques, which are an important ingredient of effective interventions
for children with externalizing problems (Greene, Ablon, Goring, Fazio, & Morse, 2004),
109
7
Chapter 7 | General discussion
including PMTO. In contrast to other European countries such as Norway and Iceland,
implementation of evidence-based interventions is fairly common in The Netherlands.
Moreover, PMTO is not the only evidence-based treatment for externalizing behavior
problems that was not found to be more effective than CAU in The Netherlands.
For example, the Triple-P Positive Parenting Program developed by Sanders (2008)
produced similar effects to CAU as delivered in The Netherlands (Kleefman, Jansen,
Stewart, & Reijneveld, 2014; Spijkers, Jansen, & Reijneveld, 2013).
The study of Ogden and Amlund-Hagen (2008) showed that discipline is one of the
most significant parenting strategies contributing to improvements in child behavior.
The time-out procedure is an important part of PMTO treatment that helps parents to
effectively discipline their child. In the past five decades, numerous studies have shown
that the use of time-out is effective in reducing antisocial behavior problems, such as
non-compliance, yelling, and aggression (e.g., Donaldson & Vollmer, 2011; Everett et al.,
2007; Scarboro & Forehand, 1975). This procedure is not only used in PMTO, but also in
other evidence-based parent management training programs such as the Incredible
Years Program (Webster-Stratton & Reid, 2010) and the above mentioned Triple-P
program (Sanders, 2008). However, recently, there has been some debate about the
application of the time-out procedure. An article in the popular American magazine
TIME even claimed that the time-out procedure would be harmful for children (Siegel
& Bryson, 2014, September 23). This conclusion was based on research showing that
the experience of relational pain (e.g., rejection) activates the same brain areas as
the experience of physical pain (Eisenberger, Lieberman, & Williams, 2003). However,
a subsequent check of the original research paper (Siegel & Bryson, 2014, September
23) indicates that the conclusions drawn in the TIME article were far too liberal in that
relational pain is not fully equal to what happens during a time-out (Siegel & Bryson,
2014, October 21). More precisely, a time-out is not emotionally damaging when
applied properly. Within PMTO, parents are taught to give a time-out in a calm way using
a neutral tone of voice. Parents only give a time-out when the child has not complied
with a request that has been repeated two times. Furthermore, it is clearly stated to the
child why he/she receives a time-out. The procedure has been explained to the child
beforehand, and therefore the child knows what to expect and what to do to comply
with his/her parent’s request in order to avoid a time-out. Most importantly, when
the time-out is finished, the issue is dropped. Especially when given occasionally and
combined with a lot of praise and positive attention for appropriate, prosocial behavior,
the time-out procedure is a safe, predictable, and effective strategy that has little to do
with (interpersonal) rejection and isolation.
110
General discussion | Chapter 7
Determinants of the effectiveness of PMTO
The fact that PMTO was not more effective in reducing externalizing behavior
problems than CAU raises questions about possible factors that may have influenced the
effectiveness of PMTO in The Netherlands. For example, prior research found evidence
to suggest that higher PMTO treatment fidelity, which can be defined as the delivery
of the therapy according to the treatment principles, predicted larger improvements
in parenting skills and externalizing behavior problems (Forgatch & DeGarmo, 2011;
Forgatch, Patterson, & DeGarmo, 2005; Hukkelberg & Ogden, 2013). Within PMTO,
treatment fidelity is measured by means of the Fidelity of Implementation Rating System
(FIMP; Knutson, Forgatch, Rains, & Sigmarsdóttir, 2009). In Chapter 3, treatment fidelity
was examined as a possible determinant of the effectiveness of PMTO. Data of the 86
PMTO families from the effectiveness study presented in Chapter 2 were used. FIMP
scores of certified PMTO therapists were significantly and meaningfully associated with
treatment outcome measures. Results showed that treatment fidelity was related to
outcome measures of externalizing behavior problems, parenting practices, parenting
stress, and parental psychopathological symptoms. The higher the fidelity scores of
the therapists, the larger the improvements on these outcome measures. In addition,
significant differences in treatment fidelity scores were found between treatment
completers and dropouts: parents who completed the treatment had a more adherent
therapist than families who dropped out. However, no significant correlations were
found between treatment fidelity and treatment satisfaction.
Previous studies on treatment fidelity of PMTO only examined whether this factor was
predictive of outcome with regard to parenting practices or child behavior problems.
Our study showed that fidelity scores were also related to other outcome measures,
such as parental stress and psychopathological symptoms. Therefore, even for certified
PMTO therapists, all of whom had obtained at least a sufficient FIMP score, treatment
fidelity varied systematically with effectiveness. Furthermore, the present study showed
that most associations were significant for the outcome assessment at 18-months
follow-up. This indicates that while the PMTO treatment is ongoing, the effects of PMTO
are similar for families who have a therapist with sufficient fidelity and those who have a
therapist with excellent fidelity. However, for an enduring effect of PMTO, it seems to be
important that the therapist adheres very closely to the PMTO method.
The assessment of parenting practices
In our effectiveness study (Chapter 2), it appeared to be difficult to measure change
in parenting practices. Since multi-method assessment has been shown to possess
incremental validity over the reliance on only one measurement method (Harvey,
Danforth, Ulaszek, & Eberhardt, 2001; Meyer et al., 2001), both self-report and observation
111
7
Chapter 7 | General discussion
of structured child-parent interactions were used to assess parenting practices. Both
methods have their advantages and limitations. Self-report questionnaires provide
a more general picture of parents’ rearing behaviors and interactions with the child,
thereby giving an impression of parenting behaviors over longer time periods and
across diverse situations (Zaslow et al., 2006). However, it may be quite difficult for
parents to evaluate their own parenting behaviors for a time frame of three or six
months (Morsbach & Prinz, 2006). Furthermore, parents may give socially desirable
answers or simply lack insight into their own behaviors towards their child (Schwarz,
1999). Structured observations of parent-child interactions in a laboratory situation or at
home may be an improvement in this regard: this method of assessing parenting is more
objective and indeed has been found to be more predictive of child outcomes than selfreport questionnaires (Duijster et al., 2015; Zaslow et al., 2006). Although observational
methods are often seen as the gold standard in the assessment of parenting practices,
their use in clinical settings is limited. Structured observations are more complex, time
consuming and, therefore, more expensive as compared to self-report scales (Hawes &
Dadds, 2006). Moreover, the observed rearing behaviors are obviously only a sample of
the behaviors parents show in real life and some important parenting practices may be
quite difficult to assess, because they rarely occur during structured observation tasks
(Peterson, Tremblay, Ewigman, & Popkey, 2002; Shelton, Frick, & Wootton, 1996). The
latter also seemed to be the case in our study. The observational tasks we used did not
elicit particularly high levels of negative behaviors in the child, so parents hardly had
to engage in disciplining their child during the assessments. Even at baseline, when
children were expected to show clear signs of externalizing behavior problems, the
frequency of such behaviors was quite low. This may partly explain why we did not find
evidence for a treatment effect on parenting skills as assessed by means of observations.
Points of critique can also be raised regarding the self-report measure of parenting
skills, the Caregiver Wish List (CWL), that we used in our study. This relatively new measure
was employed because it was specifically developed to measure the core parenting
practices of the SIL model. In our effectiveness study, the internal consistency of the
CWL appeared to be unsatisfactory for five out of the six a priori domains of parenting.
Nevertheless, with the two scales of the CWL (discouraging undesirable behavior and
CWL total score) that did show sufficient reliability, positive treatment effects could
be documented. So far, no study on the psychometric properties of the CWL has been
conducted. Therefore, the aim of the study reported in Chapter 4 was to explore the
factor structure of the CWL as a first step in validating this instrument. Furthermore,
its reliability and validity were examined. A sample of 348 parents of children aged
between 4 and 11 years was used for this purpose, of which 220 were parents of children
from the general population and 128 were parents of children with externalizing
112
General discussion | Chapter 7
behavior problems that were clinically referred. The factor analysis conducted on the
data of the total sample revealed five factors, which were named Adequate discipline,
Controlled responding, Focus on positive behavior, Consistency, and Monitoring, and
deviated substantially from the original domains. Only 23 of the 53 original CWL items
were included in the five factors. Furthermore, three of the five factors contained items
from one original domain (i.e., discouraging undesirable behavior) and no items from
the problems solving domain were represented in any of the factors. The content of
the extracted factors for the community and clinical samples was quite comparable.
Therefore, the extracted CWL factor structure seems to be applicable to parents from
both the general and clinical populations.
The extracted factor structure of the CWL can be considered an improvement over
the original domains. Although the factors contained a lower number of items, four of
the five factors exhibited acceptable internal consistency coefficients (i.e., α > .70). In
contrast, of the six original domains, the internal consistency was acceptable for only
one domain. Therefore, in the PMTO effectiveness study, only this domain and the CWL
total score could be used to examine change in parenting and these measures indeed
showed significant change over time. Furthermore, parents from the community
sample displayed significantly higher scores on the five factors than parents from the
clinical sample, indicating that parents in the community sample reported a higher
quality of parenting skills than parents in the clinical sample. Results also showed that all
extracted factors were negatively related to child behavior problems, with significantly
higher associations for externalizing behaviors than for internalizing behaviors thereby
supporting the external validity of the CWL. The use of the reliable factors could provide
insight into change on various dimensions of parenting. For research purposes, however,
it would probably be better to rely on another, better validated self-report measure for
assessing parenting skills, such as the Alabama Parenting Questionnaire (APQ; Essau,
Sasagawa, & Frick, 2006), which has been shown to possess adequate psychometric
properties in various studies (e.g., Dadds, Maujean, & Fraser, 2003; Escribano, Aniorte, &
Orgilés, 2013; Shelton et al., 1996; Zlomke, Lamport, Bauman, Garland, & Talbot, 2014)
and which also measures parenting skills similar to the ones defined by the SIL model.
Nevertheless, for clinical purposes, the CWL could still be more suitable than the APQ,
because this interview-based instrument has been designed to help therapists gather
information on strengths and weaknesses in parenting practices as seen by parents
themselves, and is likely to enhance the therapeutic alliance and positive engagement
between parents and therapists (Hodges, 2005). Thus, the CWL may have higher clinical
utility than self-report paper and pencil tests such as the APQ, and further examination
of the psychometric properties of the CWL seems warranted. Probably, items could be
added to some of the factors to improve their internal consistency.
113
7
Chapter 7 | General discussion
Inflexible clinical practice vs. a Randomized Controlled trial as ideal
A randomized controlled trial (RCT) is considered to be the gold standard for
evaluating the effectiveness of interventions. The randomization procedure ensures
that there are no differences between the experimental and control group in known
and unknown factors that might influence the outcome. This minimizes confounding,
which is the bias that occurs when one treatment group exhibits certain (known or
unknown) features affecting the relationship between the intervention and outcome,
that the other group does not possess (Sibbald & Roland, 1998; Viera & Bangdiwala,
2007). Furthermore, expectancy effects and other heuristic biases are controlled
through randomization. Therefore, the observed differences between the experimental
and control group can be attributed to the effects of the treatment instead of other
factors (Eccles, Grimshaw, Campbell, & Ramsay, 2003). However, RCTs have a number
of practical and ethical limitations. First, they are expensive and time consuming.
Furthermore, assigning patients to a control treatment that is believed to be inferior
to the target intervention is often regarded as unethical (Sibbald & Roland, 1998).
Therapists may refuse to include patients who should be allocated at random (Fairhurst
& Dowrick, 1996). Patients may also have a strong preference for one of the treatments,
which limits recruitment and could bias the outcomes (Brewin & Bradley, 1989).
In our study of the effectiveness of PMTO, we encountered some of the problems
described above. Our study was originally designed as a RCT, but we had to continue
as a quasi-experimental study, which is of course a major limitation. Allocation to the
treatment conditions (i.e., PMTO and CAU) was random at three of the five child service
agencies. At the other two agencies no randomization took place, because after the
study had started, one child service agency decided to no longer offer CAU and thus
only recruited families for the PMTO condition. Another agency was then specifically
included in the study to compensate for the missing CAU families. Although the families
in the PMTO and CAU condition did not differ in terms of demographic characteristics
and almost all of the outcome measures at baseline, the possibility cannot be ruled out
that there were still unknown factors that may have affected the outcome.
Emotional memory in children with callous-unemotional traits
Some children with externalizing behavior problems are characterized by so-called
callous-unemotional (CU) traits, similar to the emotional detachment component of
psychopathy (Frick, 2006; Frick et al., 2003). Children with CU traits are more likely to show
persistent and more serious antisocial behavior during adolescence and adulthood and
they are at greater risk of developing psychopathy (Fontaine, McCrory, Boivin, Moffitt, &
Viding, 2011). Children with CU traits have low fear, high impulsivity, are unconcerned
with other people’s feelings, and do not exhibit feelings of guilt after misbehaving
114
General discussion | Chapter 7
(Frick, 2006). Previous research has shown that these children have a specific deficit in
the processing of negative emotional stimuli. For example, they have more difficulty in
identifying fearful and sad facial expressions and are less physiologically responsive to
distressing and threatening pictures, films and words (Anastassiou-Hadjicharalambous
& Warden, 2008; Blair, 1999; Blair, Colledge, Murray, & Mitchell, 2001; Kimonis, Frick,
Fazekas, & Loney, 2006; Loney, Frick, Clements, Ellis, & Kerlin, 2003; Stevens, Charman, &
Blair, 2001; Woodworth & Waschbusch, 2008), which might also affect their emotional
memory. Impairments in emotional memory could be an explanation for why children
with CU traits learn less from negative emotional experiences, such as being disciplined
for misbehavior (see Dadds & Salmon, 2003). Previous research has shown that children
with CU traits are unresponsive to discipline, which is an important parenting strategy
in PMTO (DeBaryshe, Patterson, & Capaldi, 1993; Nix et al., 1999; Ogden & AmlundHagen, 2008; Oliver, Guerin, & Coffman, 2009). The assumption is that they are less
likely to remember the experience of being disciplined, which makes them less likely
to inhibit their bad behavior in future situations. Studies in adults and adolescents
showed that, in contrast to normally functioning healthy individuals, individuals
with psychopathic traits do not seem to exhibit enhanced memory for (negative)
emotional events (Christianson et al., 1996; Dolan & Fullam, 2005, 2010). This implies
that psychopathic people experience adverse emotional situations as neutral, which
could explain why these people use antisocial behavior instrumentally to fulfill their
desires without feelings of guilt over the other person’s distress (Dolan & Fullam, 2010).
This finding has not yet been replicated in children. Therefore, emotional memory was
examined in children high and low on CU traits. We also examined whether the deficit in
processing emotional stimuli in children with high CU traits could have an effect on their
emotional false memory. If children with high CU traits are found to be at increased risk
of developing false memories for emotional information, this could provide information
on the involvement of memory in the learning and maintenance of antisocial behavior.
Emotional memory distortions can influence the child’s own behavior as well as the
interpretation of others’ behaviors.
Chapter 5 presents results of a self-developed peripheral memory test in which
memory performance for central and peripheral components of neutral and emotionally
negative pictures in children with high CU traits was compared with the performance
of children with low CU traits. The peripheral memory test is based on the weapon
focus effect, which refers to the phenomenon that attention is being drawn to the
weapon (central aspect) and consequently less attention is devoted to the other details
(peripheral aspects). Therefore, memory is generally poorer for the peripheral details
surrounding the weapon in comparison to memory for the weapon itself (Pickel, Narter,
Jameson, & Lenhardt, 2008). From a community sample of children between 8 and 12
115
7
Chapter 7 | General discussion
years of age, children with high (n=24) and low CU traits (n=18) were selected. We found
that overall recognition rates did not differ between the high- and low-CU groups.
For negative pictures, both groups demonstrated better recognition of the central
component at the expense of the peripheral component, while for neutral pictures, the
peripheral component was better recognized than the central component. Next, these
children also performed the Deese-Roediger-McDermott (DRM) paradigm (Deese,
1959; Roediger & McDermott, 1995) to examine both true and false memory formation.
In this paradigm, children are presented with a list of words that are all semantically
related to a non-presented theme word, which is called the critical lure. A false memory
occurs when the critical lure is incorrectly recalled as being presented in the word list
(see for an overview, Brainerd, Reyna, & Ceci, 2008). For the study described in Chapter
6, five neutral and five emotionally negative word lists were used. Analyses revealed
no significant difference between the high and low CU groups in their performance
on true recall. For both groups, true recall was higher for the neutral word lists than for
the negative word lists. For false recall, there was no difference between the high and
low CU groups on the neutral word lists. However, a difference between the groups was
found on the negative word lists: the high CU group recalled fewer critical lures than the
low CU group. Moreover, the low CU group showed no difference in false recall between
the neutral and negative word lists, while the high CU group children had fewer false
memories on the negative word lists compared to the neutral word lists.
Taken together, the results of these two studies suggest that children with high CU
traits do not differ from children with low CU traits in their emotional true memory.
Therefore, children with CU traits do not seem to have a deficit in emotional memory.
This implies that impairments in emotional memory may not be an explanation for
why children with CU traits are less responsive to effective parental correction. These
findings are in contrast with the results obtained previously in adolescents and adults
(Christianson et al., 1996; Dolan & Fullam, 2005, 2010), where psychopathic offenders
did not show enhanced memory for central emotionally negative details over peripheral
details while healthy controls did. Possibly, the impairment in emotional memory has
not developed yet in children of this age. Salekin, Debus, and Jackson (2008) noted
that with regard to performance tasks, the observed association between emotional
processing deficits and psychopathy in adults is not as robust in children. Therefore,
psychopathy might not fully develop until adulthood and the impaired emotional
processing in children with high CU traits may thus not have affected memory for
emotional material yet. However, we did find a difference between children with
high and low CU traits on emotionally negative false memories. The results indicate
that children with high CU traits are better at differentiating between true and false
memories concerning negative material than children with low CU traits. Conclusively,
116
General discussion | Chapter 7
it seems that having high CU traits lowers the risk of inaccurate memories, while leaving
true emotional memory untouched.
This finding is in contrast to the predictions of the hostile attribution bias and the
general aggression model for aggression. People with a hostile attribution bias interpret
intentions of others as hostile, whether they are or not (Dodge, 1980). According to
the general aggression model, aggression in people with this bias is the result of the
activation and application of aggression-related schemas and scripts stored in memory
(Anderson & Bushman, 2002). These aggression-related knowledge structures are more
extensive and elaborate, and operate automatically in aggressive-prone individuals,
causing them to choose to behave aggressively in response to another person’s
comment or behavior (Laney & Takarangi, 2013). Based on this cognitive model,
one would expect that aggressive-prone individuals are more susceptible to falsely
remember aggression-related material. Indeed, by use of a modified DRM-paradigm,
Takarangi and colleagues (2008) found that aggressive participants were more likely to
falsely recall unpresented aggressive words as compared to less aggressive participants.
In a similar vein, studies of Vannucci and colleagues (2012; 2014) showed that violent
false memories were positively associated with delinquent behaviors and bullying in
adolescents. Recently, Laney and Takarangi (2013) even demonstrated that individuals
with aggressive tendencies were more prone to falsely remember having committed
aggressive acts. All in all, these results suggest that aggressive people are more likely to
remember aggressive information they have never encountered (Vannucci et al., 2014).
Although aggressive behavior is only one component of antisocial behavior displayed
by children with CU traits, it could still be expected that these same results would have
been found in our study, since some of our word lists also contained aggressive-related
words (e.g., murder, pain). Obviously, this post-hoc explanation is speculative and
should be examined in future studies.
However, our findings are more in line with results found in a study conducted
by Mirandola and colleagues (2014). Their findings demonstrated that elaboration
of an experienced emotional event increases the risk of false memories. This can be
explained in light of the associative activation theory (AAT; Howe, Wimmer, Gagnon, &
Plumpton, 2009), which states that semantically related words, events, and experiences
are connected in our memory and share associative relations (Wimmer & Howe, 2009).
Because associated relatedness is higher among emotionally negative material than
among neutral material (Howe, Candel, Otgaar, Malone, & Wimmer, 2010; Howe et al.,
2009; Talmi & Moscovitch, 2004), spreading activation is more automatic for negative
than for neutral material, which would lead to an increased risk of false memories
for negative material. Thus, when children without CU traits encounter a negative
emotional situation, they may react with empathy or shame to the emotion-provoking
117
7
Chapter 7 | General discussion
event. Therefore, they are likely to reflect on this experience and to elaborate on it,
which puts them at risk for memory errors (Brainerd, Stein, Silveira, Rohenkohl, & Reyna,
2008; Christianson et al., 1996; Mirandola et al., 2014). Since children with CU traits seem
to experience emotionally negative information as neutral, they may be less likely to reevaluate a negative event, which protects them against erroneous memories.
The interpretation of the results of these studies should be considered in light of
their methodological limitations, which could also be an explanation for not replicating
previous findings regarding emotional memory in individuals with psychopathic
traits. First, the low and high CU groups in our studies were selected from the general
population, while previous studies relied on clinical samples. It is possible that there
would be a difference in true recall when the high CU group is selected from the clinical
population, since they can be expected to be more extreme in their CU traits. Second,
our studies used different memory tasks than the previous studies that used a slide
show together with a short narrative (Christianson et al., 1996; Dolan & Fullam, 2005,
2010). Possibly, the memory test with narratives relies more on relational information,
while our memory tests (PMT and DRM) likely depended on item-specific information
(Hunt & Einstein, 1981). Relational information refers to features that are similar between
items that have to be remembered and item-specific information refers to features that
differentiate one item from the other items that have to be memorized. False memories
are more easily induced when relying on relational information (Cirelli, Dickinson, &
Poirier, 2014). Since persons with psychopathic traits seem to have particular deficits
in their relational processing of emotional memories (Christianson et al., 1996), one
might argue that our memory tests were not sensitive enough to detect these relational
processing deficits. Finally, our sample size was fairly small, which may have led to
insufficient power to detect small between group differences.
Callous-unemotional traits in relation to parent management training
programs
Unfortunately, we were not able to assess CU traits in the children included in the
PMTO effectiveness study. By the time the idea of incorporating CU traits emerged, the
effectiveness study was already too far under way. Furthermore, we did not want to
increase the burden on the parents more by adding another questionnaire to the already
quite extensive battery of tests included in the study. Nevertheless, it would have been
interesting to examine whether the effects of PMTO (and CAU) are different for children
high and low on CU traits. As mentioned in the introduction of this dissertation, the
frequency and severity of externalizing behavior problems in children with CU traits
seem to be independent of the quality of parenting (Hipwell et al., 2007; Oxford, Cavell,
& Hughes, 2003; Viding, Fontaine, Oliver, & Plomin, 2009; Vitacco, Neumann, Ramos, &
118
General discussion | Chapter 7
Roberts, 2003; Wootton, Frick, Shelton, & Silverthorn, 1997). Surprisingly, little research
has been conducted on the association between CU traits and parent management
training outcome. The few trials that have been conducted show that children with
high CU traits have poor treatment outcomes as compared with children low on CU
traits (e.g., Hawes & Dadds, 2005; Hawes, Dadds, Brennan, Rhodes, & Cauchi, 2013;
Högström, Enebrink, & Ghaderi, 2013; Kimonis, Bagner, Linares, Blake, & Rodriguez,
2014). Importantly, parents of high and low CU children did not differ in improvement
of parenting skills during the treatment (Högström et al., 2013). However, this does
not mean that the level of behavior problems in children with CU traits is immutable.
Research has found that the level of externalizing behavior problems in CU children
decreased after parent management training, especially when these treatments were
based on social learning theory (Hawes, Price, & Dadds, 2014). Furthermore, it has been
found that parent management training for children with high CU traits is particularly
capable of improving CU traits when delivered early in childhood (i.e., preschool age;
Hawes et al., 2014). These findings clearly suggest that pre-treatment data on CU
traits are clinically informative regarding the prognostic status of children referred for
treatment for externalizing behavior problems (Hawes et al., 2014).
Some components of parent management training might be more effective in
children with CU traits than others. Since children with CU traits appear to have a reward
dominance response style and seem to be relatively insensitive to discipline, they are
expected to be more responsive to positive reinforcement than discipline. Therefore, to
achieve the most change in externalizing behavior problems in children with CU traits,
a focus on positive reinforcement and parental warmth may be particularly indicated in
parent management training (Hawes et al., 2014). In accordance with this notion, Hawes
and Dadds (2005) found that boys with high CU traits were less responsive to discipline
by means of a time-out and responded with less affect to this type of parenting than
boys without CU traits. Furthermore, a review by Frick and colleagues (2014) showed
that children with high and low CU traits responded equally well to the part of the parent
training that focused on positive reinforcement, while children with low levels of CU
traits demonstrated added improvement with the part of the intervention that focused
on teaching parents to use more effective discipline strategies. It should be noted,
however, that the focus should not be exclusively on warm and positive parenting.
Parents should also be equipped with non-coercive ways of discipline, because coercive
parenting has been associated with increases in CU traits (Waller et al., 2012).
119
7
Chapter 7 | General discussion
Concluding remarks and directions for future research
The findings of this dissertation suggest that PMTO is effective in The Netherlands
in improving child externalizing behavior problems, effective parenting practices,
parenting stress, and parental psychopathological symptoms, but not more effective
than CAU. Before drawing any conclusions on the future of PMTO in The Netherlands,
it should first be examined for whom and under which circumstances the treatment
is most effective. Therefore, a number of possible moderators and determinants of
the effectiveness of PMTO were investigated. Of several child, parent, and family
factors, only externalizing behavior problems at baseline appeared to be a moderator:
children with more severe externalizing behavior problems at baseline showed more
improvement than children with less severe problems. The current dissertation also
showed that the effectiveness of PMTO is significantly related to the treatment integrity
of the PMTO therapist. When PMTO therapists adhered closely to the program principles,
more positive long-term treatment effects were obtained. Moderate adherence to the
treatment model appears to be insufficient to achieve this. Furthermore, research by
others has shown that children with CU traits respond less well to effective parenting.
As a possible explanation for this, emotional memory was examined in children high
and low on CU traits, because the assumption is that they are less likely to remember
the experience of being disciplined, which makes them less likely to inhibit their bad
behavior in future situations. However, no differences in emotional (true) memory
were found between children high and low on CU traits, indicating that an impaired
emotional memory is not an explanation for why children with high CU traits are less
responsive to an intervention that focuses on effective parenting than children with low
CU traits.
Although PMTO was not more effective than CAU, it could be that PMTO is more
cost-effective than CAU, especially when long term societal effects are taken into
account. Evidence of cost-effectiveness is important, because it can assist in decisions
concerning which intervention to use, at what cost and with what benefit (Charles,
Bywater, & Edwards, 2011). Along with our study on the effectiveness of PMTO, data
concerning child and parental health, and the use of health care (benefits) as well as
time investment and other costs of the interventions (costs) were collected. These data
are currently being analyzed by researchers of the Medical Center of the University
of Groningen (UMCG). If these data show that PMTO is more cost-effective than CAU,
PMTO would be the preferable intervention for children with externalizing behavior
problems. When no differences between PMTO and CAU in cost-effectiveness are found,
PMTO could still be an improvement over CAU for some children with externalizing
behavior problems. In combination with previous studies, the results of the research
presented in this dissertation suggest that PMTO could be more beneficial for children
120
General discussion | Chapter 7
with more severe externalizing behavior problems. Further, especially when delivered
by strongly adherent therapists, PMTO could be more effective than CAU. The ultimate
test of PMTO’s effectiveness, compared to CAU, lies in its long-term ability to curb a
child’s development into an adolescent and adult with an antisocial lifestyle. We
strongly recommend follow-up research in our sample of treated children, to monitor
their developmental trajectory.
More research on the factors that predict, influence, or account for variations in the
treatment effectiveness of PMTO is needed, and would contribute to our knowledge
about how to adjust the treatment to the specific needs of the individual family in order to
improve the intervention. Furthermore, the presence of CU traits should be investigated
in PMTO effectiveness trials, because research has shown that the presence of CU traits
in children with externalizing behavior problems seems to have a negative impact
on treatment outcome and there are tentative indications that some components of
parent management training are more effective in these children than others. In order
to meet the specific needs of children with CU traits, accurate screening of CU traits
in children referred for parent management training is necessary. Considering the fact
that CU traits are more likely to improve when addressed early in childhood, screening
of these traits at a young age is recommended. Finally, the effectiveness of PMTO has
primarily been investigated in Caucasian samples and should also be examined in other
ethnic populations. This is relevant as PMTO has not been specifically developed for
families with other ethnic backgrounds and research has demonstrated that ethnic
minority families are less likely to be referred to or attend interventions (Lau, 2006).
Therefore, cultural adaptation of the PMTO manual is needed to improve engagement
and acceptability in order to achieve better retention rates of ethnic minority families.
The PMTO manual has been adapted to meet the cultural patterns, meanings, and
values of Mexican (Baumann, Rodríguez, Amador, Forgatch, & Parra-Cardona, 2014)
and of Somali and Pakistani (Bjørknes & Manger, 2013) families. The effectiveness of
the adapted version of PMTO has not yet been examined for the Mexican families, but
has been demonstrated to improve parenting skills and mother-reported child problem
behavior in Somali and Pakistani families in Norway (Bjørknes & Manger, 2013).
121
7
Summary
S
Summary |
Externalizing behavior problems in children can have significant negative long-term
consequences and constitute a major social and financial burden for the community.
Hence, it is important to intervene as soon as these behavior problems occur. Ineffective
parenting is one of the well-established risk factors for externalizing behavior problems
in children. Therefore, many interventions for childhood externalizing problems focus
on improving parenting skills. Parent Management Training –Oregon model (PMTO)
is an intervention that focuses on parents of children between 4 and 12 years of age
with severe externalizing behavior problems. In the current thesis, a series of studies is
described that examined the effectiveness of PMTO in The Netherlands as well as factors
that could have an influence on the effectiveness of parent management trainings in
general, and PMTO in particular. The thesis consists of two parts. Part I contains three
chapters on the effectiveness of PMTO in reducing externalizing problem behavior
in children in The Netherlands. Part II consists of two chapters concerning emotional
memory in children with callous-unemotional traits.
The first part concerns research that examined the effectiveness of PMTO as an
intervention for clinically referred children with externalizing behavior problems in
The Netherlands. In Chapter 2, PMTO was compared with Care As Usual (CAU), which
consisted of a mix of other viable interventions employed for treating children with
externalizing behavior problems. Families were recruited at five different youth care
agencies throughout The Netherlands. Of the 146 children and their parents included
in the study, 91 received PMTO while 55 were given CAU. Families were assessed at four
time points: at pretreatment, and after 6, 12, and 18 months. Results showed that PMTO
was effective in reducing children’s problem behaviors, increasing the use of effective
parenting practices, and alleviating parenting stress and psychopathological symptoms
of the parents. However, PMTO appeared to be no more effective than CAU. For both
treatments, the improvements were most evident during the first six months of the
study and remained stable until the 18-months follow-up assessment. Additionally, we
found that especially children with serious externalizing behavior problems at baseline
benefited from PMTO. Comparing the effect size of PMTO delivered in The Netherlands
with previous PMTO effectiveness studies in Norway and Iceland, we demonstrated that
PMTO had a similar, large effect size as shown in previous studies. From these findings it
can be concluded that PMTO is effective in a clinically referred sample of children with
externalizing problems in the Netherlands, although it seems to be no more effective
than CAU.
The finding that PMTO was not more effective in reducing externalizing behavior
problems than CAU raises questions about possible factors that may have influenced the
effectiveness of PMTO in The Netherlands. For example, prior research found evidence
to suggest that higher PMTO treatment fidelity might predict larger improvements
125
S
| Summary
in parenting skills and externalizing behavior problems. In Chapter 3, treatment
fidelity was examined as a possible determinant of the effectiveness of PMTO. Data
of 86 PMTO families from the effectiveness study presented in Chapter 2 were used.
Results showed that treatment fidelity scores of certified PMTO therapists as obtained
before treatment (i.e., at certification) were significantly associated with the outcome
measures. The higher the fidelity scores of the therapists, the larger the improvements
on externalizing behavior problems, parenting practices, parenting stress, and parental
psychopathological symptoms. Interestingly, the strongest associations with fidelity
were found for the outcome assessment at the 18-months follow-up. This indicates
that very close adherence to the PMTO method is especially important for the longer
term sustenance of the treatment effects of PMTO. In addition, we found that parents
who completed the treatment more often had a therapist who adhered more closely
to the PMTO protocol than families who dropped out. However, parents were not more
satisfied with their therapist and the intervention when their therapist adhered more
strictly to PMTO guidelines, revealing that treatment satisfaction is largely independent
of how well the therapist delivers the PMTO treatment.
In our study examining the effectiveness of PMTO (Chapter 2), we found that the
parenting domains measured by our self-report measure of parenting strategies, the
Caregiver Wish List (CWL), showed poor internal consistency. That is, reliability was
only found to be adequate for the CWL total score and for one of its domains. Thus,
we decided to examine the psychometric properties of the CWL in more detail in the
study presented in Chapter 4. More precisely, the factor structure of the instrument as
well as its reliability and validity were explored in a sample of 348 parents of children
aged between 4 and 11 years, of which 220 were parents of children from the general
population and 128 were parents of clinically referred children with externalizing
behavior problems. Factor analyses indicated that a five factor solution (obtained
after eliminating quite a number of unsatisfactory items) provided the best fit. The five
factors were labelled as Adequate discipline, Controlled responding, Focus on positive
behavior, Consistency, and Monitoring, and all showed adequate internal consistency
coefficients. The correlations among the factors were rather low, suggesting that they
represent fairly independent parenting domains. Furthermore, the factors differentiated
between the community and clinical sample, with significantly better parenting skills
observed for the community sample. Finally, all factors were significantly and negatively
related to child behavior problems, with stronger correlations found for externalizing
compared to internalizing problems, which provides support for the external validity of
this parenting measure.
126
Summary |
The second part of this dissertation describes two studies that focus on the
emotional memory of children with CU traits, that is, children who are particularly prone
to develop externalizing psychopathology. In Chapter 5, a self-developed memory
test was administered to a community sample of children between 8 and 12 years of
age with high (n = 24) and low (n = 18) CU traits. The peripheral memory test assesses
memory performance for central and peripheral components of emotionally neutral
and negative pictures. We found that overall recognition rates did not differ between
the high and low CU groups. For negative pictures, both groups demonstrated better
recognition of the central components at the expense of the peripheral components,
while for neutral pictures, the peripheral components were better recognized than the
central components. These results suggest that there are no differences in emotional
memory between children with high and low CU traits, implying that children with CU
traits do not have an impaired emotional memory.
Chapter 6 describes a study in which emotional false memory was examined by use
of the Deese-Roediger-McDermott paradigm. Children high and low on CU traits were
subjected to five neutral and five emotionally negative word lists. Analyses revealed
that true recall was higher for the neutral word lists than for the negative word lists in
both the high and low CU groups. For false recall, memory performance did not differ
between the high and low CU groups on the neutral word lists, but the high CU group
recalled fewer unpresented words than the low CU group on the negative word lists.
Moreover, the low CU group showed no difference in false recall between the neutral
and negative word lists, while the high CU group children had fewer false memories
on the negative word lists compared to the neutral word lists. The results indicate
that children with high CU traits are better at differentiating between true and false
memories concerning negative material than children with low CU traits. Therefore,
children with high CU traits seem to be at lower risk of inaccurate memories.
The final chapter of this dissertation (Chapter 7) contains a general discussion of
the main findings of the studies in the preceding chapters. Furthermore, this chapter
outlines methodological issues of the research presented in this dissertation and
implications for clinical practice that follow from these results. The main message of this
chapter, and of this dissertation as a whole, is that PMTO is effective in improving child
externalizing behavior problems, effective parenting practices, parenting stress, and
parental psychopathological symptoms, but not more effective than CAU. However, the
effectiveness of PMTO is related to the treatment integrity of the PMTO therapist.When PMTO
therapists adhered closely to the program principles, more positive long-term treatment
effects were obtained. Furthermore, this chapter concludes that an impaired emotional
memory is not an explanation for why children with high CU traits are less responsive
to an intervention that focuses on effective parenting than children with low CU traits.
127
S
Samenvatting
S
Samenvatting |
Kinderen met externaliserende gedragsproblemen worden gekenmerkt door
gedragingen als vechten, vernielen, ongehoorzaamheid, spijbelen, liegen en stelen.
Deze gedragsproblemen kunnen op de lange termijn leiden tot verdere ontsporing
in crimineel gedrag en vormen zo een grote sociale en financiële last voor de
samenleving. Daarom is het belangrijk om zo vroeg mogelijk in te grijpen wanneer
deze gedragsproblemen zich voordoen. Ineffectieve opvoeding is één van de meest
onderzochte en vastgestelde risicofactoren voor externaliserende gedragsproblemen
bij kinderen. Veel behandelingen richten zich dan ook op het verbeteren van
opvoedingsvaardigheden van ouders. Parent Management Training – Oregon
model (PMTO) is een training die ontwikkeld is voor ouders van kinderen tussen de
4 en 12 jaar met ernstige externaliserende gedragsproblemen. PMTO is gebaseerd
op het Social Interaction Learning (SIL) model. Het SIL model veronderstelt dat de
opvoedingsvaardigheden van ouders een direct effect hebben op de ontwikkeling
van het kind. Contextuele factoren, zoals een lage socio-economische status en
psychopathologie bij ouders, zouden een negatieve invloed hebben op de kwaliteit van
opvoeding en daarmee een indirect negatief effect op de ontwikkeling van het kind.
In dit proefschrift wordt een aantal studies beschreven naar de effectiviteit van PMTO
in Nederland en naar factoren die mogelijk de effectiviteit van PMTO beïnvloeden. Dit
proefschrift bestaat uit twee delen. Het eerste deel bevat drie hoofdstukken over de
effectiviteit van PMTO in Nederland. Het tweede deel bestaat uit twee hoofdstukken
over het emotioneel geheugen van kinderen met psychopathische trekken.
In Hoofdstuk 2 wordt PMTO vergeleken met Care As Usual (CAU). CAU bestaat
uit een mix van andere interventies die worden aangeboden aan kinderen met
externaliserende gedragsproblemen en hun ouders. Van de 146 gezinnen die aan de
studie deelnamen, kregen er 91 PMTO en 55 CAU. De uitkomstmetingen van deze
studie vonden op vier momenten plaats: vóór de start van de behandeling (voormeting)
en vervolgens na 6, 12 en 18 maanden. PMTO bleek effectief in het verminderen van
externaliserende gedragsproblemen bij kinderen, het verbeteren van de effectieve
opvoedingsvaardigheden en het verlichten van de opvoedingsstress en lichamelijke
en psychische klachten (bijv. angst, depressie, slaapproblemen) bij de ouders. PMTO
bleek echter niet significant effectiever dan CAU. Voor beide behandelingen waren de
verbeteringen het grootste tijdens de eerste zes maanden en deze bleven vervolgens
stabiel tot de laatste meting na 18 maanden. Verder vonden we dat vooral kinderen
met ernstige externaliserende gedragsproblemen, zoals gemeten bij de voormeting,
meer baat hadden bij de PMTO behandeling. Uit deze resultaten kan geconcludeerd
worden dat PMTO effectief is voor kinderen met externaliserende gedragsproblemen
in Nederland, maar PMTO blijkt niet effectiever dan het al bestaande behandelaanbod
voor deze kinderen (CAU).
131
S
| Samenvatting
Aangezien PMTO niet effectiever bleek te zijn dan CAU in het verminderen
van externaliserende gedragsproblemen, hebben we onderzocht welke factoren
mogelijk invloed hebben op de effectiviteit van PMTO. Eén van die factoren is de
behandelintegriteit van therapeuten. Behandelintegriteit is de mate waarin de
therapeut de behandelprincipes van een interventie toepast zoals deze bedoeld zijn.
Eerder onderzoek heeft aangetoond dat een grotere behandelintegriteit gepaard
gaat met sterkere verbeteringen in opvoedingsvaardigheden en externaliserende
gedragsproblemen. In Hoofdstuk 3 werd onderzocht of behandelintegriteit invloed
heeft op de effectiviteit van PMTO in Nederland. Hier werden gegevens van 86
PMTO gezinnen gebruikt die hadden deelgenomen aan de effectiviteitsstudie uit
Hoofdstuk 2. Alle PMTO therapeuten die de behandelingen gaven waren in het
kader van hun certificering, voorafgaand aan ons effectiviteitsonderzoek, gescoord
op behandelintegriteit. Deze behandelintegriteitsscores hingen samen met de
uitkomstmaten: hoe hoger de scores van therapeuten op behandelintegriteit, hoe
sterker de verbeteringen op externaliserende gedragsproblemen van de kinderen,
en opvoedingsvaardigheden, opvoedingsstress en lichamelijk en psychische klachten
van de ouders. Een opmerkelijk resultaat was dat de sterkste samenhang met
behandelintegriteit werd gevonden voor de laatste meting na 18 maanden. Dit geeft aan
dat het sterk vasthouden aan de behandelprincipes vooral belangrijk is voor het lange
termijn behoud van het behandeleffect van PMTO. Verder vonden we dat ouders die de
behandeling afgemaakt hebben een therapeut met hogere behandelintegriteit hadden
dan ouders die voortijdig waren gestopt. Ouders bleken echter niet meer tevreden over
hun therapeut wanneer de therapeut zich strakker aan de PMTO richtlijnen hield. Dit
betekent dat de behandeltevredenheid van ouders los staat van hoe goed de therapeut
de PMTO behandeling geeft.
In de studie waarin de effectiviteit van PMTO werd onderzocht (Hoofdstuk 2) maakten
we onder andere gebruik van de Caregiver Wish List (CWL), een gestandaardiseerd
interview met ouders voor het meten van de kwaliteit van hun opvoedingsvaardigheden,
zoals door henzelf beoordeeld. De CWL meet zes verschillende opvoedingsdomeinen,
zoals toezicht houden en positieve betrokkenheid bij het kind, maar deze bleken een
lage betrouwbaarheid (interne consistentie) te hebben. Dit betekent dat de vragen in
de verschillende opvoedingsdomeinen niet hetzelfde opvoedingsaspect meten. In onze
effectiviteitsstudie bleek de betrouwbaarheid van de CWL alleen voor de CWL totaal
score en voor één van de zes opvoedingsdomeinen adequaat te zijn. Daarom besloten
we om de psychometrische kenmerken van de CWL meer gedetailleerd te onderzoeken
in de studie beschreven in Hoofdstuk 4. De factor structuur, betrouwbaarheid en
validiteit van het instrument werden onderzocht in een steekproef van 348 ouders
van kinderen tussen de 4 en 11 jaar. Hiervan kwamen 220 ouders uit de algemene
132
Samenvatting |
bevolking (zij werden via basisscholen geworven) en 128 ouders van kinderen die
waren doorverwezen vanwege externaliserende gedragsproblemen (klinische
populatie). Factoranalyses gaven aan dat een vijf-factoren-oplossing het beste bij de
CWL-gegevens paste. De vijf factoren kregen de labels Adequate discipline, Beheerst
reageren, Focus op positief gedrag, Consistentie en Toezicht houden en lieten allemaal
een goede interne betrouwbaarheid zien. De correlaties tussen de factoren waren
redelijk laag, wat betekent dat ze grotendeels onafhankelijke opvoedingsvaardigheden
weergeven. Verder vonden we een verschil in scores op de vijf factoren tussen de twee
groepen ouders. Ouders uit de algemene populatie rapporteerden, zoals verwacht,
betere opvoedingsvaardigheden dan ouders uit de klinische populatie. Bovendien
waren alle factoren negatief gerelateerd aan gedragsproblemen: hoe beter de
opvoedingsvaardigheden, hoe minder de gedragsproblemen. Deze verbanden waren
sterker voor externaliserende dan voor internaliserende (angstig, teruggetrokken,
depressief ) gedragsproblemen. Deze resultaten bieden ondersteuning voor de externe
validiteit van de CWL: het instrument lijkt vijf belangrijke opvoedingsvaardigheden
accuraat te kunnen meten.
Het tweede deel van dit proefschrift beschrijft studies die ingaan op het
emotioneel geheugen van kinderen met psychopathische trekken. Kinderen met deze
trekken kennen weinig tot geen angst, zijn impulsief, bekommeren zich niet om de
gevoelens van anderen en ervaren geen schuldgevoelens. Deze kinderen lopen een
verhoogd risico op het ontwikkelen van serieuze antisociale gedragsproblemen in de
adolescentie en volwassenheid. Eerder onderzoek heeft aangetoond dat effectieve
opvoedingsvaardigheden ongerelateerd zijn aan de gedragsproblemen van kinderen
met psychopathische trekken. Een beperking van het emotioneel geheugen zou een
verklaring kunnen zijn voor waarom deze kinderen minder goed leren van straf. Voor de
studie in Hoofdstuk 5 is een zelfontworpen geheugentest afgenomen bij kinderen tussen
de 8 en 12 jaar met een hoge (n = 24) versus een lage (n = 18) score op psychopathische
trekken. De geheugentest meet de geheugenprestatie voor centrale en perifere (aan
de rand) details van neutrale en emotioneel negatief beladen plaatjes. We vonden
dat herkenning van de details niet verschilde tussen de hoge en lage psychopathie
groepen. Voor de negatieve plaatjes herkenden beide groepen de centrale details beter
in vergelijking met de perifere details. Bij de neutrale plaatjes werden juist de perifere
details beter herkend dan de centrale details. Deze resultaten suggereren dat er geen
verschil is in emotioneel geheugen tussen kinderen met hoge en lage psychopathische
trekken en daarmee dat kinderen met psychopathische trekken geen beperkt
emotioneel geheugen hebben. Dit suggereert dat een beperkt emotioneel geheugen
niet als verklaring kan dienen voor waarom effectieve opvoedingsvaardigheden minder
effect hebben op kinderen met psychopathische trekken.
133
S
| Samenvatting
Hoofdstuk 6 beschrijft een studie waarin emotioneel geheugen werd onderzocht
met behulp van het Deese-Roediger-McDermott paradigma. In dit paradigma krijgen
kinderen een lijst met woorden te horen (bijv. bakker, boter, beleg, etc.) die ze moeten
proberen te onthouden. Deze woorden zijn allemaal geassocieerd aan een nietgepresenteerd themawoord (brood). Wanneer kinderen vervolgens, op de vraag welke
woorden zij gehoord hebben, het thema woord noemen (brood), is er sprake van een
pseudoherinnering. We lieten kinderen met hoge en lage psychopathische trekken vijf
emotioneel neutrale en vijf negatieve woordlijsten horen. Uit de analyses kwam naar
voren dat beide groepen meer neutrale dan negatieve woorden hadden onthouden. We
vonden geen verschillen tussen de groepen voor de neutrale woordlijsten wat betreft het
aantal pseudoherinneringen. Voor de negatieve woordlijsten vonden we echter dat de
kinderen met hoge psychopathische trekken minder pseudoherinneringen opnoemden
dan de kinderen met lage psychopathische trekken. Verder liet de laag psychopathische
groep geen verschil zien in pseudoherinneringen op de negatieve en neutrale
woordlijsten, terwijl de hoog psychopathische groep minder pseudoherinneringen
had voor de negatieve woordlijsten in vergelijking met de neutrale woordlijsten. Deze
resultaten geven aan dat kinderen met hoge psychopathische trekken beter zijn in het
onderscheiden van echte en pseudoherinneringen voor negatief materiaal dan kinderen
met lage psychopathische trekken. Kinderen met hoge psychopathische trekken lijken
daarmee minder vatbaar te zijn voor het ontwikkelen van inaccurate herinneringen
voor negatieve gebeurtenissen. Een verklaring hiervoor kan zijn dat kinderen met
psychopathische trekken een negatieve gebeurtenis als neutraal beschouwen en
daardoor minder waarschijnlijk zijn om de gebeurtenis te herevalueren, waardoor ze
beschermd worden tegen pseudoherinneringen.
Het laatste hoofdstuk (Hoofdstuk 7) bevat een discussie van de belangrijkste
bevindingen van de studies uit de voorgaande hoofdstukken. De kernboodschap
van dit hoofdstuk, en daarmee van dit gehele proefschrift, is dat PMTO effectief is in
het verminderen van externaliserende gedragsproblemen bij kinderen, vergroten
van effectieve opvoedingsvaardigheden, en verminderen van opvoedingsstress en
lichamelijke en psychische klachten bij ouders, maar niet effectiever dan standaard
zorgaanbod. De effectiviteit van PMTO hangt samen met de behandelintegriteit van de
PMTO therapeut. Wanneer de PMTO therapeut zich streng aan de behandelprincipes
houdt, leidt dat tot grotere lange termijn effecten. Verder concluderen we dat een
zwak emotioneel geheugen niet als verklaring kan dienen voor de bevinding uit
eerdere onderzoeken dat kinderen met hoge psychopathische trekken minder goed
reageren op een behandeling die gericht is op het versterken van de effectieve
opvoedingsvaardigheden bij hun ouders.
134
Valorization
addendum
V
Valorization addendum |
The present dissertation describes the results of a large multi-center study examining
the effectiveness of Parent Management Training - Oregon model (PMTO) for parents
of children with externalizing behavior problems. This dissertation also studied factors
that could have an influence on the effectiveness of parent management training in
general, and PMTO in particular. This valorization addendum discusses the findings in a
broader societal and economical context.
Relevance of the research
More than 13% of the children and adolescents in The Netherlands show externalizing
behavior problems (de Looze et al., 2014). These youth are at increased risk of later
antisocial behavior (Coté, Vaillancourt, LeBlanc, Nagin, & Tremblay, 2006; Simonoff et al.,
2004). Moreover, the earlier externalizing behavior problems begin, the higher the risk
of serious problems during adulthood (Moffitt, 1993; Patterson, DeBaryshe, & Ramsey,
1989). When externalizing behavior problems are not properly addressed, about 50
to 75% of the youth will continue to show serious problem behaviors (Nixon, 2002).
Behavior problems are not only a social burden to the community, but also a financial
one. For example, Scott and colleagues (2001) found that by the age of 28, the costs for
individuals who had conduct disorder as a child were 10 times higher than for persons
without psychiatric problems and 3.5 times higher than for individuals who as a child
had displayed only mild behavioral problems. Therefore, it is important to intervene as
soon as a child shows signs of externalizing behavior problems.
In The Netherlands, there was a need for effective treatment programs for children
4-12 years with antisocial conduct problems. Already in 2005, the Ministry of Health,
Welfare and Sport decided to fund the implementation of Parent Management Training
- Oregon model (PMTO), but required that its effectiveness would be evaluated. This
treatment had been found to be effective in other countries, but it could not be
automatically assumed that it would also be effective in The Netherlands. Further, it
is not only important to examine if a certain treatment is effective, but also for whom
and under which circumstances. For example, research has shown that children with
externalizing behavior problems who also exhibit callous-unemotional (CU) traits have
an earlier onset of offending, display higher levels of aggression and delinquency,
and more frequently come into contact with the police as compared to children with
externalizing behaviour problems who do not exhibit these traits (Andershed, Gustafson,
Kerr, & Stattin, 2002; Frick, Cornell, Barry, Bodin, & Dane, 2003; Frick, Stickle, Dandreaux,
Farrell, & Kimonis, 2005; Frick & White, 2008; Lawing, Frick, & Cruise, 2010). Although
not examined directly in this dissertation, there is evidence to suggest that youth with
a combination of externalizing problems and CU traits are less responsive to effective
parenting strategies and to treatments focusing on the improvement of parenting
137
V
| Valorization addendum
skills. Further exploration of these issues would enable us to improve interventions for
children with externalizing problems and if necessary to adjust treatments to the specific
needs of these youngsters and their families. The studies in this thesis empirically tested
a number of factors that might influence the effectiveness of PMTO specifically and
parenting trainings in general. The general discussion in Chapter 7 gives a more detailed
description of the theoretical, clinical, and research implications of our research and
provides directions for future research.
To whom the results of this dissertation may be relevant
First, the results are relevant for parents of children with externalizing behavior
problems and their social environment. Our studies show that PMTO and other
interventions available in The Netherlands are effective in reducing externalizing
behavior problems of children. Furthermore, this dissertation contributes to the
knowledge about factors that could influence the effectiveness of such treatments,
which could be used to optimize evidence-based interventions for children with
externalizing behavior problems. Second, our results are also interesting for PMTO
therapists and PMTO supervisors, especially the findings on treatment fidelity of PMTO
as described in Chapter 3. Although previous research had already demonstrated that
fidelity scores were predictive of positive change in parenting skills and externalizing
behavior problems, our findings show that high treatment fidelity is especially important
for the long-term effectiveness of PMTO. These results may also be relevant for clinicians
conducting other interventions for externalizing behavior problems or even clinicians
carrying out treatments for other psychological problems, such as depression, anxiety
disorders, and substance use disorders. In The Netherlands, clinicians are used to
working according to a treatment protocol. However, only a few of these treatments
have integrated fidelity checks as part of the program. This means that therapists may
drift from the treatment principles and/or may only use elements of the protocol. The
results of our study indicate that it is important to check treatment adherence in order
to optimize the long-term effectiveness of the interventions. Third, since the findings
from this dissertation may contribute to a reduction of the social and financial burden
caused by children’s externalizing behavior problems on the community, the gained
insights are also of interest to policy makers and health insurance companies. Finally, this
dissertation may also be relevant for researchers in child and adolescent mental health.
A randomized controlled trial (RCT) is considered the gold standard for evaluating the
effectiveness of treatment interventions. However, the study presented in Chapter 2
shows how difficult it is to conduct a RCT in regular clinical practice. Our study started
as a RCT, but we encountered several organizational problems which eventually forced
us to continue our study with a quasi-experimental design. These problems are not
138
Valorization addendum |
uncommon when conducting intervention research in a clinical setting and underline
the need to consider alternative research designs (Langendam, Hooft, Heus, Scholten,
& Bossuyt, 2013).
Activities and services
The findings of this dissertation show that PMTO is effective in improving child
externalizing behavior problems and that it seems to be as effective as other interventions
for children with externalizing behavior problems in The Netherlands. This knowledge
can be used in decision making about which treatment to recommend for which
family. For example, the choice of the most suitable intervention could be based on the
resources of the families, such as the availability of time and transportation. However,
this recommendation is still premature, since evidence of the cost-effectiveness and
the long-term (i.e., > 10 years) effectiveness of the treatments is needed before firm
conclusions about relative effectiveness can be drawn. Furthermore, in Chapter 7
of this dissertation, the importance of screening for CU traits is addressed, since this
characteristic could influence the effects of parenting treatment. To achieve optimal
change in externalizing behavior problems in children with CU traits, a focus on positive
reinforcement and parental warmth may be particularly indicated. This dissertation
further demonstrates that the quality of interventions for children with externalizing
behavior problems in The Netherlands is quite good. This implies that the focus should
not necessarily be on implementing new interventions for this target group, but on
improving and optimizing the current treatments and in optimizing their treatment
fidelity. For example, the study presented in Chapter 3 provides evidence for the
influence of treatment integrity on the effectiveness of PMTO. This knowledge could
also be applied to other interventions in order to further increase their effectiveness.
The present dissertation found support for Social Interaction Learning (SIL) theory.
This model states that parents have a direct influence on the behaviors of the child,
implying that by improving the parenting strategies of parents, the behavior problems
of the child will be reduced. This insight underlines the influence that parents have on
their children and this might not only apply to externalizing behavior problems, but
to other problems as well. For example, it has already been shown that the quality of
parenting is related to dental caries in children (de Jong-Lenters, Duijster, Bruist, Thijssen,
& de Ruiter, 2014; Duijster et al., 2015). Therefore, improving effective parenting skills
could also be an effective mechanism in improving other child outcomes.
139
V
| Valorization addendum
Innovation of the research
The study on the effectiveness of PMTO in The Netherlands (Chapter 2) is not
necessarily innovative. The effectiveness of PMTO had already been examined in
other countries and our research design is quite similar to that applied in these other
studies. We used a longitudinal design and compared families receiving PMTO with
other available interventions by using a multimethod and multi-informant approach.
This research was necessary because the effectiveness of interventions should also be
evaluated in The Netherlands, in order to be registered in the Dutch Youth Institute
(Nederlands Jeugdinstituut; NJi) database of effective interventions. Furthermore, the
similarity in research design with other PMTO studies allows for a direct comparison of
the results.
The other studies described in this dissertation are more innovative. First, the
study on treatment integrity of PMTO therapists (Chapter 3) used a different approach
compared to previous studies. For example, earlier studies examined whether treatment
fidelity was predictive of only one outcome measure (i.e., parenting practices or child
behavior problems), while we also assessed the association between fidelity and
parental stress and psychopathological symptoms. In addition, our study used fidelity
scores of certified PMTO therapists, obtained prior to treatment, and examined the
association with effectiveness at different time points. Based on this innovative design,
we were able to show that even for certified PMTO therapists, treatment fidelity varied
systematically with effectiveness on all our outcome measures and that very close
adherence to the PMTO method is especially important for the endurance of the effects
produced by PMTO. Second, the study on the Caregiver Wish List (CWL) described in
Chapter 4 is innovative because no other study can be found in the literature examining
the psychometric properties of this instrument. Therefore, our study was the first to
examine its factor structure, reliability, and validity. Finally, the studies described in
Chapters 5 and 6 also used an innovative approach. So far, only three studies examined
emotional memory in participants with psychopathic (CU) traits, but none of them
focused on children. The study presented in Chapter 5 was the first to theoretically relate
emotional memory in children with CU-traits to unresponsiveness to effective parenting
and parenting training, and used a self-developed memory test which already received
interest from other researchers who investigate emotional memory in children. For
example, our peripheral memory test has been used to examine whether induced stress
affects emotional memory in 5- to 9-year old children (R. Neuenschwander, personal
communication, June 10, 2015). Emotional false memory in children with CU-traits (see
Chapter 6) is a new research area, which could contribute to our knowledge on the role
of memory in learning and maintenance of antisocial cognitions.
140
Valorization addendum |
Knowledge dissemination
The outcomes of the studies described in this dissertation have been, and will be,
communicated in several ways. The studies presented in Chapters 2, 3, and 4 have
recently been finished and are submitted to international peer reviewed journals. The
studies described in Chapters 5 and 6 have already been published in international
journals. Furthermore, the research in Chapter 2, 5, and 6 has been presented at several
national and international conferences, which were attended by both researchers and
professionals from the clinical field. Additionally, results of the studies presented in
this dissertation were communicated through lectures given to university students at
Maastricht University. Communication of the findings in this thesis outside the scientific
field will be done through a press release.
V
141
References
R
References |
Achenbach, T. M. (1991). Manual for the Child
Behavior Checklist/4-18. Burlington: University of
Vermont, Department of Psychiatry.
Alink, L. R. A., Mesman, J., van Zeijl, J., Stolk, M. N.,
Juffer, F., Koot, H. M., . . . van Ijzendoorn, M. H.
(2006). The early childhood aggression curve:
Development of physical aggression in 10- to
50-month-old children. Child Development, 77,
954-966. doi: 10.1111/j.1467-8624.2006.00912.x
American Psychiatric Association. (2000). Diagnostic
and statistical manual of mental disorders (4th ed.,
text rev.). Washington, DC: American Psychiatric
Association.
American Psychiatric Association. (2013). Diagnostic
and statistical manual of mental disorders (5th
ed.). Washington, DC: American Psychiatric
Association.
Amlund-Hagen, K., Ogden, T., & Bjørnebekk, G.
(2011). Treatment outcomes and mediators of
parent management training: A one-year followup of children with conduct problems. Journal of
Clinical Child and Adolescent Psychology, 40, 165178. doi: 10.1080/15374416.2011.546050
Anastassiou-Hadjicharalambous, X., & Warden,
D. (2008). Physiologically-indexed and selfperceived affective empathy in conductdisordered children high and low on callousunemotional traits. Child Psychiatry and Human
Development, 39, 503-517. doi: 10.1007/s10578008-0104-y
Andershed, H. A., Gustafson, S. B., Kerr, M., & Stattin,
H. k. (2002). The usefulness of self-reported
psychopathy-like traits in the study of antisocial
behaviour among non-referred adolescents.
European Journal of Personality, 16, 383-402. doi:
10.1002/per.455
Anderson, C. A., & Bushman, B. J. (2002).
Human
aggression.
Annual Review of
Psychology, 53, 27-51. doi: 10.1146/annurev.
psych.53.100901.135231
Arrindell, W. A., & Ettema, J. M. H. (1986). SCL90: Handleiding bij een Multidimensionele
Psychopathologie-Indicator [Manual for a
Multidimensional Psychopathology Indicator].
Lisse, The Netherlands: Swets & Zeitlinger.
Arrindell, W. A., & Ettema, J. M. H. (2003). Symptom
Checklist. Handleiding bij een multidimensionele
psychopathologie-indicator [Manual for a
Multidimensional Psychopathology Indicator]
(2nd edition). Lisse, The Netherlands: Swets Test
Publishers.
Bank, L., Marlowe, J. H., Reid, J. B., & Patterson,
G. R. (1991). A comparative evaluation of
parent-training interventions for families of
chronic delinquents. Journal of Abnormal Child
Psychology, 19, 15-33. doi: 10.1007/BF00910562
Barzi, F., & Woodward, M. (2004). Imputation
of missing values in practice: Results from
imputation of serum cholestorol in 28 cohort
studies. American Journal of Epidemiology, 160,
34-45. doi: 10.1093/aje/kwh175
Bates, J. E., Pettit, G. S., Dodge, K. A., & Ridge,
B. (1998). Interaction of temperamental
resistance to control and restrictive parenting
in the development of externalizing behavior.
Developmental Psychology, 34, 982-995. doi:
10.1037/0012-1649.34.5.982
Bauer, S., Lambert, M. J., & Nielsen, S. L. (2004). Clinical
significance methods: A comparison of statistical
techniques. Journal of Personality Assessment, 82,
60-70. doi: 10.1207/s15327752jpa8201_11
Baumann, A. A., Rodríguez, M. M. D., Amador, N. G.,
Forgatch, M. S., & Parra-Cardona, J. R. (2014).
Parent Management Training-Oregon model
(PMTOTM) in Mexico City: Integrating cultural
adaptation activities in an implementation
model. Clinical Psychology: Science and Practice,
21, 32-47. doi: 10.1111/cpsp.12059
Bennett, K. J., Lipman, E. L., Brown, S., Racine, Y.,
Boyle, M. H., & Offord, D. R. (1999). Predicting
conduct problems: Can high-risk children be
identified in kindergarten and Grade 1? Journal
of Consulting and Clinical Psychology, 67, 470480. doi: 10.1037/0022-006X.67.4.470
Berger, M., & van Everdingen, J. (2006). De
positionering van PMTO ten opzichte van andere
intensieve ouderinterventies in Nederland.
Utrecht: NIZW.
Bierman, K. L., & Smoot, D. L. (1991). Linking family
characteristics with poor peer relations: The
mediating role of conduct problems. Journal
of Abnormal Child Psychology, 19, 341-356. doi:
10.1007/BF00911236
Bijttebier, P., & Decoene, S. (2009). Assessment
of psychopathic traits in children and
adolescents: Further validation of the Antisocial
Process Screening Device and the Childhood
Psychopathy Scale. European Journal of
Psychological Assessment, 25, 157-163.
Bjørknes, R., & Manger, T. (2013). Can parent training
alter parent practice and reduce conduct
problems in ethnic minority children? A
randomized controlled trial. Prevention Science,
14, 52-63. doi: 10.1007/s11121-012-0299-9
145
R
| References
Blair, R. J. R. (1999). Responsiveness to distress cues
in the child with psychopathic tendencies.
Personality and Individual Differences, 27, 135145.
Blair, R. J. R., Colledge, E., Murray, L., & Mitchell,
D. G. V. (2001). A selective impairment in the
processing of sad and fearful expressions in
children with psychopathic tendencies. Journal
of Abnormal Child Psychology, 29, 491-498. doi:
10.1023/A:1012225108281
Brainerd, C. J., Holliday, R. E., Reyna, V. F., Yang, Y., &
Toglia, M. P. (2010). Developmental reversals in
false memory: Effects of emotional valence and
arousal. Journal of Experimental Child Psychology,
107, 137-154. doi: 10.1016/j.jecp.2010.04.013
Brainerd, C. J., & Reyna, V. F. (2012). Reliability of
children’s testimony in the era of developmental
reversals. Developmental Review. doi: 10.1016/j.
dr.2012.06.008
Brainerd, C. J., Reyna, V. F., & Ceci, S. J. (2008a).
Developmental reversals in false memory: A
review of data and theory. Psychological Bulletin,
134, 343-382. doi: 10.1037/0033-2909.134.3.343
Brainerd, C. J., Reyna, V. F., & Zember, E. (2011).
Theoretical and forensic implications of
developmental studies of the DRM illusion.
Memory & Cognition, 39, 365-380. doi: 10.3758/
s13421-010-0043-2
Brainerd, C. J., Stein, L. M., Silveira, R. A., Rohenkohl,
G., & Reyna, V. F. (2008b). How does negative
emotion cause false memories? Psychological
Science, 19, 919-925. doi: 10.1111/j.14679280.2008.02177.x
Brewin, C. R., & Bradley, C. (1989). Patient preferences
and randomised clinical trials. British Medical
Journal, 299, 313-315.
Broidy, L. M., Nagin, D. S., Tremblay, R. E., Bates, J.
E., Brame, B., Dodge, K. A., . . . Vitaro, F. (2003).
Developmental trajectories of childhood
disruptive
behaviors
and
adolescent
delinquency: A six-site, cross-national study.
Developmental Psychology, 39, 222-245. doi:
10.1037/0012-1649.39.2.222
Buchanan, T. W. (2007). Retrieval of emotional
memories. Psychological Bulletin, 133, 761-779.
doi: 10.1037/0033-2909.133.5.761
Burke, A., Heuer, F., & Reisberg, D. (1992).
Remembering emotional events. Memory &
Cognition, 20, 277-290.
146
Burke, J. D., Loeber, R., & Lahey, B. B. (2007).
Adolescent conduct disorder and interpersonal
callousness as predictors of psychopathy
in young adults. Journal of Clinical Child
and Adolescent Psychology, 36, 334-346. doi:
10.1080/15374410701444223
Cahill, L., & McGaugh, J. L. (1995). A novel
demonstration of enhanced memory associated
with
emotional
arousal.
Consciousness
and Cognition, 4, 410-421. doi: 10.1006/
ccog.1995.1048
Charles, J. M., Bywater, T., & Edwards, R. T. (2011).
Parenting interventions: A systematic review of
the economic evidence. Child: Care, Health and
Development, 37, 462-474. doi: 10.1111/j.13652214.2011.01217.x
Christianson, S.-Ã. (1992). Emotional stress
and eyewitness memory: A critical review.
Psychological Bulletin, 112, 284-309. doi:
10.1037//0033-2909.112.2.284
Christianson, S.-Ã., Forth, A. E., Hare, R. D., Strachan, C.,
Lidberg, L., & Thorell, L.-H. (1996). Remembering
details of emotional events: A comparison
between psychopathic and nonpsychopathic
offenders. Personality and Individual Differences,
20, 437-443. doi: 10.1016/0191-8869(95)00220-0
Christianson, S.-Ã., Loftus, E. F., Hoffman, H., &
Loftus, G. R. (1991). Eye fixations and memory
for emotional events. Journal of Experimental
Psychology: Learning, Memory, and Cognition, 17,
693-701.
Chronis, A. M., Lahey, B. B., Pelham, W. E., Williams, S.
H., Baumann, B. L., Kipp, H., . . . Rathouz, P. J. (2007).
Maternal depression and early positive parenting
predict future conduct problems in young
children with attention-deficit/hyperactivity
disorder. Developmental Psychology, 43, 70-82.
doi: 10.1037/0012-1649.43.1.70
Cirelli, L. K., Dickinson, J., & Poirier, M. (2014). Using
implicit instructional cues to influence false
memory induction. Journal of Psycholinguistic
Research, 44, 485-494. doi: 10.1007/s10936-0149301-y
Conger, R. D., Patterson, G. R., & Ge, X. (1995). It
takes two to replicate: A mediational model
for the impact of parents’ stress on adolescent
adjustment. Child Development, 66, 80-97. doi:
10.2307/1131192
Corson, Y., & Verrier, N. (2007). Emotions and false
memories: Valence or arousal? Psychological
Science, 18, 208-211. doi: 10.1111/j.14679280.2007.01874.x
References |
Coté, S. M., Vaillancourt, T., LeBlanc, J. C., Nagin, D.
S., & Tremblay, R. E. (2006). The development of
physical aggression from toddlerhood to preadolescence: A nation wide longitudinal study
of Canadian children. Journal of Abnormal Child
Psychology, 34, 71-85. doi: 10.1007/s10802-0059001-z
Cunningham, C. E., & Boyle, M. H. (2002). Preschoolers
at risk for attention-deficit hyperactivity disorder
and oppositional defiant disorder: Family,
parenting, and behavioral correlates. Journal
of Abnormal Child Psychology, 30, 555-569. doi:
10.1023/A:1020855429085
Dadds, M. R., Fraser, J., Frost, A., & Hawes, D. J. (2005).
Disentangling the underlying dimensions
of psychopathy and conduct problems in
childhood: A community study. Journal of
Consulting and Clinical Psychology, 73, 400-410.
doi: 10.1037/0022-006X.73.3.400
Dadds, M. R., Maujean, A., & Fraser, J. A. (2003).
Parenting and conduct problems in children:
Australian data and psychometric properties
of the Alabama Parenting Questionnaire.
Australian Psychologist, 38, 238-241. doi:
10.1080/00050060310001707267
Deese, J. (1959). On the prediction of occurrence of
particular verbal intrusions in immediate recall.
Journal of Experimental Psychology, 58, 17-22.
DeGarmo, D. S., & Forgatch, M. S. (2005). Early
development of delinquency within divorced
families: Evaluating a randomized preventive
intervention trial. Developmental Science, 8, 229239. doi: 10.1111/j.1467-7687.2005.00412.x
DeGarmo, D. S., & Forgatch, M. S. (2007). Efficacy of
parent training for stepfathers: From playful
spectator and polite stranger to effective
stepfathering. Parenting: Science and Practice, 7,
331-355. doi: 10.1080/15295190701665631
DeGarmo, D. S., Patterson, G. R., & Forgatch, M.
S. (2004). How do outcomes in a specified
parent training intervention maintain or wane
over time? Prevention Science, 5, 73-89. doi:
10.1023/B:PREV.0000023078.30191.e0
Derogatis, L. R. (1977). SCL-90-R: Administration, scoring
& procedures manual-II for the R(evised) version.
Baltimore: Clinical Psychometric Research.
Dodge, K. A. (1980). Social cognition and children’s
aggressive behavior. Child Development, 51, 162170. doi: 10.2307/1129603
Dadds, M. R., & Salmon, K. (2003). Punishment
insensitivity and parenting: Temperament
and learning as interacting risks for antisocial
behavior. Clinical Child and Family Psychology
Review, 6, 69-86. doi: 10.1023/A:1023762009877
Dolan, M. C., & Fullam, R. (2005). Memory for
emotional events in violent offenders with
antisocial personality disorder. Personality
and Individual Differences, 38, 1657-1667. doi:
10.1016/j.paid.2004.09.028
de Brock, A. J. L. L., Vermulst, A. A., Gerris, J. R. M., &
Abidin, R. R. (1992). Nijmeegse Ouderlijke Stress
Index. Opvoeding in gezin en andere primaire
leefvormen [Nijmegen Parenting Stress Index:
Caregiving in the family and other primary
care systems]. Lisse, The Netherlands: Swets &
Zeitlinger.
Dolan, M. C., & Fullam, R. (2010). Emotional
memory and psychopathic traits in conduct
disordered
adolescents.
Personality and
Individual Differences, 48, 327-331. doi: 10.1016/j.
paid.2009.10.029
de Jong-Lenters, M., Duijster, D., Bruist, M. A., Thijssen,
J., & de Ruiter, C. (2014). The relationship
between parenting, family interaction and
childhood dental caries: A case-control study.
Social Science & Medicine, 116, 49-55. doi: doi:
10.1016/j.socscimed.2014.06.031
and without release contingencies. Journal of Applied
Behavior Analysis, 44, 693-705. doi: 10.1901/
jaba.2011.44-693
de Looze, M., Van Dorsselaer, S., de Roos, S.,
Verdurmen, J., Stevens, G., Gommans, R., . . .
Vollebergh, W. (2014). HBSC 2013. Gezondheid,
welzijn en opvoeding van jongeren in Nederland.
Utrecht: Universiteit Utrecht.
Parental and family-related influences on dental
caries in children of Dutch, Moroccan and
Turkish origin. Community Dentistry and Oral
Epidemiology, 43, 152-162. doi: 10.1111/
cdoe.12134
DeBaryshe, B. D., Patterson, G. R., & Capaldi, D. M.
(1993). A performance model for academic
achievement in early adolescent boys.
Developmental Psychology, 29, 795-804. doi:
10.1037/0012-1649.29.5.795
Eccles, M., Grimshaw, J., Campbell, M., & Ramsay, C.
(2003). Research designs for studies evaluating
the effectiveness of change and improvement
trategies. Quality & Safety in Health Care, 12, 4752. doi: 10.1136/qhc.12.1.47
Donaldson, J. M., & Vollmer, T. R. (2011). An evaluation
and comparison of time-out procedures with
Duijster, D., de Jong-Lenters, M., de Ruiter, C.,
Thijssen, J., van Loveren, C., & Verrips, G. (2015).
147
R
| References
Ehrensaft, M. K., Wasserman, G. A., Verdelli, L.,
Greenwald, S., Miller, L. S., & Davies, M. (2003).
Maternal antisocial behavior, parenting
practices, and behavior problems in boys at risk
for antisocial behavior. Journal of Child and Family
Studies, 12, 27-40. doi: 10.1023/A:1021302024583
Eisenberger, N. I., Lieberman, M. D., & Williams, K.
D. (2003). Does rejection hurt? An fMRI study
of social exclusion. Science, 302, 290-292. doi:
10.1126/science.1089134
Escribano, S., Aniorte, J., & Orgilés, M. (2013). Factor
structure and psychometric properties of the
Spanish version of the Alabama Parenting
Questionnaire (APQ) for children. Psicothema,
25, 324-329. doi: 10.7334/psicothema2012.315
Essau, C. A., Sasagawa, S., & Frick, P. J. (2006).
Psychometric properties of the Alabama
Parenting Questionnaire. Journal of Child and
Family Studies, 15, 597-616. doi: 10.1007/s10826006-9036-y
Everett, G. E., Olmi, D. J., Edwards, R. P., Tingstrom,
D. H., Sterling-Turner, H. E., & Christ, T. J.
(2007). An empirical investigation of timeout with and without escape extinction to
treat
escape-maintained
noncompliance.
Behavior Modification, 31, 412-434. doi:
10.1177/0145445506297725
Fairhurst, K., & Dowrick, C. (1996). Problems with
recruitment in a randomized controlled
trial of counselling in general practice:
Causes and implications. Journal of Health
Services Research and Policy, 1, 77-80. doi:
10.1177/135581969600100205
Fisher, P. A., Burraston, B., & Pears, K. (2005). The Early
Intervention Foster Care Program: Permanent
placement outcomes from a randomized
trial. Child Maltreatment, 10, 61-71. doi:
10.1177/1077559504271561
Fite, P. J., Greening, L., Stoppelbein, L., & Fabiano,
G. A. (2009). Confirmatory factor analysis of
the Antisocial Process Screening Device with
a clinical inpatient population. Assessment, 16,
103-114. doi: 10.1177/1073191108319893
Fontaine, N. M. G., McCrory, E. J. P., Boivin, M., Moffitt,
T. E., & Viding, E. (2011). Predictors and outcomes
of joint trajectories of callous-unemotional traits
and conduct problems in childhood. Journal of
Abnormal Psychology, 120, 730-742. doi: 10.1037/
a0022620
148
Forgatch, M. S., Bullock, B. M., & Patterson, G. R.
(2004). From theory to practice: Increasing
effective parenting through role-play. In
H. Steiner (Ed.), Handbook of mental health
interventions in children and adolescents: An
integrated developmental approach. (pp. 782813). San Francisco: Jossey-Bass.
Forgatch, M. S., & DeGarmo, D. S. (1999). Parenting
through change: An effective prevention
program for single mothers. Journal of
Consulting and Clinical Psychology, 67, 711-724.
doi: 10.1037/0022-006X.67.5.711
Forgatch, M. S., & DeGarmo, D. S. (2011). Sustaining
fidelity following the nationwide PMTO™
implementation in Norway. Prevention Science,
12, 235-246. doi: 10.1007/s11121-011-0225-6
Forgatch, M. S., DeGarmo, D. S., & Beldavs, Z. G. (2005a).
An efficacious theory-based intervention
for stepfamilies. Behavior Therapy, 36, 357365.
doi:
10.1016/S0005-7894(05)80117-0
Forgatch, M. S., Knutson, N. M., & Mayne, T. (1992).
Coder impressions of ODS lab tasks. Eugene:
Oregon Social Learning Center.
Forgatch, M. S., & Patterson, G. R. (2010). Parent
Management Training - Oregon model: An
intervention for antisocial behavior in children
and adolescents. In J. R. Weisz & A. E. Kazdin
(Eds.), Evidence-based psychotherpies for children
and adolescents (pp. 159-178). New York, NY:
Guilford.
Forgatch, M. S., Patterson, G. R., & DeGarmo, D. S.
(2005b). Evaluating fidelity: Predictive validity
for a measure of competent adherence to the
Oregon Model of Parent Management Training.
Behavior Therapy, 36, 3-13. doi: 10.1016/S00057894(05)80049-8
Forgatch, M. S., Patterson, G. R., DeGarmo, D. S.,
& Beldavs, Z. G. (2009). Testing the Orgeon
delinquency model with 9-year follow-up of
the Oregon Divorce study. Development and
Psychopathology, 21, 637-660. doi: 10.1017/
S0954579409000340
Frick, P. J. (2006). Developmental pathways to
conduct disorder. Child and Adolescent
Psychiatric Clinics of North America, 15, 311-331.
doi: 10.1016/j.chc.2005.11.003
Frick, P. J., Bodin, S. D., & Barry, C. T. (2000). Psychopathic
traits and conduct problems in community and
clinic-referred samples of children: Further
development of the Psychopathy Screening
Device. Psychological Assessment, 12, 382-393.
doi: 10.1037/1040-3590.12.4.382
References |
Frick, P. J., Cornell, A. H., Barry, C. T., Bodin, S. D.,
& Dane, H. E. (2003a). Callous-unemotional
traits and conduct problems in the prediction
of conduct problem severity, aggression,
and self-report of delinquency. Journal of
Abnormal Child Psychology, 31, 457-470. doi:
10.1023/A:1023899703866
Greene, R. W., Ablon, J. S., Goring, J. C., Fazio, V., &
Morse, L. R. (2004). Treatment of oppositional
defiant disorder in children and adolescents. In
P. M. Barrett & T. H. Ollendick (Eds.), Handbook
of interventions that work with children and
adolescents: Prevention and treatment. (pp. 369393). New York, NY, US: John Wiley & Sons Ltd.
Frick, P. J., & Hare, R. D. (2001). Antisocial Process
Screening Device. Toronto: Multi-Health Systems.
Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R.
E. (2009). Multivariate data analysis. A global
perspective. New Jersey: Pearson Education Inc.
Frick, P. J., Kimonis, E. R., Dandreaux, D. M., & Farell, J.
M. (2003b). The 4 year stability of psychopathic
traits in non-referred youth. Behavioral Sciences
& the Law, 21, 713-736. doi: 10.1002/bsl.568
Frick, P. J., Lahey, B. B., Loeber, R., Stouthamer-Loeber,
M., Christ, M. A. G., & Hanson, K. (1992). Familial
risk factors to oppositional defiant disorder and
conduct disorder: Parental psychopathology and
maternal parenting. Journal of Consulting and
Clinical Psychology, 60, 49-55. doi: 10.1037/0022006X.60.1.49
Frick, P. J., Ray, J. V., Thornton, L. C., & Kahn, R. E.
(2014). Can callous-unemotional traits enhance
the understanding, diagnosis, and treatment
of serious conduct problems in children
and adolescents? A comprehensive review.
Psychological Bulletin, 140, 1-57. doi: 10.1037/
a0033076
Frick, P. J., Stickle, T. R., Dandreaux, D. M., Farrell, J.
M., & Kimonis, E. R. (2005). Callous-unemotional
traits in predicting the severity and stability of
conduct problems and delinquency. Journal
of Abnormal Child Psychology, 33, 471-487. doi:
10.1007/s10648-005-5728-9
Frick, P. J., & White, S. F. (2008). Research review:
The importance of callous-unemotional traits
for developmental models of aggressive and
antisocial behavior. Journal of Child Psychology
and Psychiatry, 49, 359-375. doi: 10.1111/j.14697610.2007.01862.x
Gallo, D. A. (2010). False memories and fantastic
beliefs: 15 years of the DRM illusion. Memory &
Cognition, 38, 833-848. doi: 10.3758/MC.38.7.833
Glass, S. J., & Newman, J. P. (2009). Emotion
processing in the criminal psychopath: The role
of attention in emotion-facilitated memory.
Journal of Abnormal Psychology, 118, 229-234.
doi: 10.1037/a0014866
Hamilton, W., & Burns, T. G. (2003). Review of WPPSIIII: Wechsler Preschool and Primary Scale of
Intelligence (3rd Ed.). Applied Neuropsychology,
10, 188-190.
Hare, R. D. (1991). Manual for the Hare Psychopathy
Checklist-revised. Toronto: Multi-Health Systems.
Harvey, E., Danforth, J. S., Ulaszek, W. R., & Eberhardt,
T. L. (2001). Validity of the parenting scale for
parents of children with attention-deficit/
hyperactivity disorder. Behaviour Research
and Therapy, 39, 731-743. doi: 10.1016/S00057967(00)00052-8
Hawes, D. J., & Dadds, M. R. (2005). The treatment
of conduct problems in children with callousunemotional traits. Journal of Consulting
and Clinical Psychology, 73, 737-741. doi:
10.1037/0022-006x.73.4.737
Hawes, D. J., & Dadds, M. R. (2006). Assessing
parenting practices through parent-report
and direct observation during parent-training.
Journal of Child and Family Studies, 15, 555-568.
doi: 10.1007/s10826-006-9029-x
Hawes, D. J., Dadds, M. R., Brennan, J., Rhodes, T., &
Cauchi, A. (2013). Revisiting the treatment of
conduct problems in children with callousunemotional traits. Australian and New
Zealand Journal of Psychiatry, 47, 646-653. doi:
10.1177/0004867413484092
Hawes, D. J., Price, M. J., & Dadds, M. R. (2014). Callousunemotional traits and the treatment of conduct
problems in childhood and adolescence: A
comprehensive review. Clinical Child and Family
Psychology Review, 17, 248-267. doi: 10.1007/
s10567-014-0167-1
Hetherington, E. M., Bridges, M., & Insabella, G. M.
(1998). What matters? What does not? Five
perspectives on the association between marital
transitions and children’s adjustment. American
Psychologist, 53, 167-184. doi: 10.1037/0003066X.53.2.167
149
R
| References
Heuer, F., & Reisberg, D. (1990). Vivid memories of
emotional events: The accuracy of remembered
minutiae. Memory & Cognition, 18, 496-506. doi:
10.3758/BF03198482
Hipwell, A. E., Pardini, D. A., Loeber, R., Sembower,
M., Keenan, K., & Stouthamer-Loeber, M. (2007).
Callous-unemotional behaviors in young
girls: Shared and unique effects relative to
conduct problems. Journal of Clinical Child
and Adolescent Psychology, 36, 293-304. doi:
10.1080/15374410701444165
Hodges, K. (2002). Caregiver Wish List. Ypsilanti:
Eastern Michigan University.
Hodges, K. (2005). Caregiver Wish List Handbook.
Ypsilanti: Eastern Michigan University.
Hodges, K., de Ruiter, C., & Thijssen, J. (2009).
Authorized Dutch translation of the Caregiver Wish
List. Maastricht, The Netherlands: Department
of Clinical Psychological Science, Maastricht
University.
Högström, J., Enebrink, P., & Ghaderi, A. (2013). The
moderating role of child callous-unemotional
traits in an Internet-based parent-management
training program. Journal of Family Psychology,
27, 314-323. doi: 10.1037/a0031883
Hogue, A., Henderson, C. E., Dauber, S., Barajas, P.
C., Fried, A., & Liddle, H. A. (2008). Treatment
adherence, competence, and outcome in
individual and family therapy for adolescent
behavior problems. Journal of Consulting
and Clinical Psychology, 76, 544-555. doi:
10.1037/0022-006x.76.4.544
Howe, M. L., Candel, I., Otgaar, H., Malone, C.,
& Wimmer, M. C. (2010). Valence and the
development of immediate and long-term
false memory illusions. Memory, 18, 58-75. doi:
10.1080/09658210903476514
Howe, M. L., Toth, S. L., & Cicchetti, D. (2011). Can
maltreated children inhibit true and false
memories for emotional information? Child
Development, 82, 967-981. doi: 10.1111/j.14678624.2011.01585.x
Howe, M. L., Wimmer, M. C., Gagnon, N., & Plumpton,
S. (2009). An associative-activation theory of
children’s and adults’ memory illusions. Journal
of Memory and Language, 60, 229-251. doi:
10.1016/j.jml.2008.10.002
Hu, L., & Bentler, P. M. (1999). Cutoff criteria fo fit indexes
in covariance structure analysis: Conventional
criteria versus new alternatives. Structural
Equation Modeling: A Multidisciplinary Journal, 6,
1-55. doi: 10.1080/10705519909540118
150
Huey, S. J., Jr., Henggeler, S. W., Brondino, M. J., &
Pickrel, S. G. (2000). Mechanisms of change in
multisystemic therapy: Reducing delinquent
behavior through therapist adherence and
improved family and peer functioning. Journal
of Consulting and Clinical Psychology, 68, 451467. doi: 10.1037/0022-006x.68.3.451
Hukkelberg, S. S., & Ogden, T. (2013). Working
alliance and treatment fidelity as predictors
of externalizing problem behaviors in parent
management training. Journal of Consulting and
Clinical Psychology, 81, 1010-1020. doi: 10.1037/
a0033825
Hunt, R. R., & Einstein, G. O. (1981). Relational and
item-specific information in memory. Journal of
Verbal Learning & Verbal Behavior, 20, 497-514.
Jacobson, N. S., Roberts, L. J., Berns, S. B., &
McGlinchey, J. B. (1999). Methods for defining
and determining the clinical significance of
treatment effects: Description, application, and
alternatives. Journal of Consulting and Clinical
Psychology, 67, 300-307. doi: 10.1037/0022006x.67.3.300
Johnston, C., & Jassy, J. S. (2007). Attention-Deficit/
Hyperactivity Disorder and Oppositional/
Conduct problems: Links to parent-child
interactions. Journal of the Canadian Academy of
Child and Adolescent Psychiatry, 16, 74-79.
Kaufman, A. S., Kaufman, J. C., Balgopal, R., &
McLean, J. E. (1996). Comparison of three
WISC-III short forms: Weighing psychometric,
clinical, and practical factors. Journal of Clinical
Child Psychology, 25, 97-105. doi: 10.1207/
s15374424jccp2501_11
Kazdin, A. E. (2007). Progress in treating children
referred for severe aggressive and antisocial
behavior. NYS Psychologist, 19, 7-12.
Kimonis, E. R., Bagner, D. M., Linares, D., Blake, C.
A., & Rodriguez, G. (2014). Parent training
outcomes among young children with callous–
unemotional conduct problems with or at risk
for developmental delay. Journal of Child and
Family Studies, 23, 437-448. doi: 10.1007/s10826013-9756-8
Kimonis, E. R., Frick, P. J., Fazekas, H., & Loney, B.
R. (2006). Psychopathy, aggression, and the
processing of emotional stimuli in non-referred
girls and boys. Behavioral Sciences & the Law, 24,
21-37. doi: 10.1002/bsl.668
References |
Kjøbli, J., Hukkelberg, S., & Ogden, T. (2013). A
randomized trial of group parent training:
Reducing child conduct problems in real-world
settings. Behaviour Research and Therapy, 51,
113-121. doi: 10.1016/j.brat.2012.11.006
Ledingham, J. E., & Schwartzman, A. E. (1984). A
3-year follow-up of aggressive and withdrawn
behavior in childhood: Preliminary findings.
Journal of Abnormal Child Psychology, 12, 157168. doi: 10.1007/BF00913467
Kjøbli, J., & Ogden, T. (2012). A randomized
effectiveness trial of brief parent training in
primary care settings. Prevention Science, 13,
616-626. doi: 10.1007/s11121-012-0289-y
Lee, I. A., & Preacher, K. J. (2013, September).
Calculation for the test of the difference between
two dependent correlations with one variable in
common [Computer software]. Retrieved from
http://quantpsy.org.
Kleefman, M., Jansen, D. E., Stewart, R. E., & Reijneveld,
S. A. (2014). The effectiveness of Stepping Stones
Triple P parenting support in parents of children
with borderline to mild intellectual disability and
psychosocial problems: a randomized controlled
trial. BMC Medicine, 12, 191-198. doi: 10.1186/
s12916-014-0191-5
Knutson, N. M., Forgatch, M. S., Rains, L. A.,
& Sigmarsdóttir, M. (2009). Fidelity of
implementation rating system (FIMP): The manual
for PMTO. Eugene, Oregon: Implementation
Sciences International, Inc.
Kramer, T. H., Buckhout, R., & Eugenio, P. (1990).
Weapon focus, arousal, and eyewitness memory:
Attention must be paid. Law and Human
Behavior, 14, 167-184. doi: 10.1007/bf01062971
LaBar, K. S., & Cabeza, R. (2006). Cognitive
neuroscience of emotional memory. Nature
Reviews Neuroscience, 7, 54-64. doi: 10.1038/
nrn1825
Laney, C., & Takarangi, M. K. T. (2013). False memories
for aggressive acts. Acta Psychologica, 143, 227234. doi: 10.1016/j.actpsy.2013.04.001
Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1998).
International affective picture system (IAPS):
Technical manual and affective ratings. Centre
for research in psychophysiology, University of
Florida: Gainesville.
Langendam, M. W., Hooft, L., Heus, P., Scholten, R.
J. P. M., & Bossuyt, P. M. M. (2013). Alternatieven
voor Randomized Controlled Trials in onderzoek
naar de effectiviteit van interventies. Amsterdam:
Dutch Cochrane Centre.
Lau, A. S. (2006). Making the case for selective and
directed cultural adaptations of evidence-based
treatments: Examples from parent training.
Clinical Psychology: Science and Practice, 13, 295310. doi: 10.1111/j.1468-2850.2006.00042.x
Lawing, K., Frick, P. J., & Cruise, K. R. (2010). Differences
in offending patterns between adolescent sex
offenders high or low in callous-unemotional
traits. Psychological Assessment, 22, 298-305. doi:
10.1037/a0018707
Levine, L. J., & Edelstein, R. S. (2009). Emotion and
memory narrowing: A review and goal-relevance
approach. Cognition and Emotion, 23, 833-875.
doi: 10.1080/02699930902738863
Little, R. (1988). Missing-data adjustments in large
surveys. Journal of Business and Economic
Statistics, 6, 287-296.
Loeber, R., & Burke, J. D. (2011). Developmental
pathways in juvenile externalizing and
internalizing problems. Journal of Research on
Adolescence, 21, 34-46. doi: 10.1111/j.15327795.2010.00713.x
Loftus, E. F. (2005). Planting misinformation in the
human mind: A 30-year investigation of the
malleability of memory. Learning & Memory, 12,
361-366. doi: 10.1101/lm.94705
Loney, B. R., Frick, P. J., Clements, C. B., Ellis, M. L., &
Kerlin, K. (2003). Callous-unemotional traits,
impulsivity, and emotional processing in
adolescents with antisocial behavior problems.
Journal of Clinical Child and Adolescent Psychology,
32, 66-80. doi: 10.1207/S15374424JCCP3201_07
Martinez, C. R., & Forgatch, M. S. (2001). Preventing
problems with boys’ noncompliance: Effects
of a parent training intervention for divorcing
mothers. Journal of Consulting and Clinical
Psychology, 69, 416-428. doi: 10.1037/0022006X.69.3.416
McCart, M. R., Priester, P. E., Davies, W. H., & Azen, R.
(2006). Differential effectiveness of Behavioral
Parent-Training
and
Cognitive-Behavioral
Therapy for antisocial youth: A meta-analysis.
Journal of Abnormal Child Psychology, 34, 527543. doi: 10.1007/s10802-006-9031-1
McCoy, M. G., Frick, P. J., Loney, B. R., & Ellis, M. L.
(1999). The potential mediating role of parenting
practices in the development of conduct
problems in a clinic-referred sample. Journal
of Child and Family Studies, 8, 477-494. doi:
10.1023/A:1021907905277
151
R
| References
McGraw, K. O., & Wong, S. P. (1996). Forming
inferences about some intraclass correlation
coefficients. Psychological Methods, 1, 30-46. doi:
10.1037/1082-989x.1.1.30
Nixon, R. D. V. (2002). Treatment of behavior
problems in preschoolers: A review of parent
training programs. Clinical Psychology Review, 22,
525-546. doi: 10.1016/S0272-7358(01)00119-2
McMahon, R. J., Witkiewitz, K., & Kotler, J. S. (2010).
Predictive validity of callous-unemotional traits
measured in early adolescence with respect
to multiple antisocial outcomes. Journal of
Abnormal Psychology, 119, 752-763. doi: 10.1037/
a0020796
Obradovic, J., Pardini, D. A., Long, J. D., & Loeber, R.
(2007). Measuring interpersonal callousness
in boys from childhood to adolescence: An
examination of longitudinal invariance and
temporal stability. Journal of Clinical Child
and Adolescent Psychology, 36, 276-292. doi:
10.1080/15374410701441633
Meyer, G. J., Finn, S. E., Eyde, L. D., Kay, G. G., Moreland,
K. L., Dies, R. R., . . . Reed, G. M. (2001). Psychological
testing and psychological assesment: A review
of evidence and issues. American Psychologist,
56, 128-165. doi: 10.1037/0003-066X.56.2.128
Minshew, N. J., Turner, C. A., & Goldstein, G. (2005).
The application of short forms of the Wechsler
Intelligence Scales in adults and children with
high functioning autism. Journal of Autism and
Developmental Disorders, 35, 45-52. doi: 10.1007/
s10803-004-1030-x
Mirandola, C., Toffalini, E., Grassano, M., Cornoldi, C.,
& Melinder, A. (2014). Inferential false memories
of events: Negative consequences protect
from distortions when the events are free from
further elaboration. Memory, 22, 451-461. doi:
10.1080/09658211.2013.795976
Moffitt, T. E. (1993). Adolescence-limited and
life-course-persistent antisocial behavior: A
developmental taxonomy. Psychological Review,
100, 674-701. doi: 10.1037/0033-295X.100.4.674
Morsbach, S. K., & Prinz, R. J. (2006). Understanding
and improving the validity of self-report of
parenting. Clinical Child and Family Psychology
Review, 9, 1-21. doi: 10.1007/s10567-006-0001-5
Nagin, D., & Tremblay, R. E. (1999). Trajectories of
boys’ physical aggression, opposition, and
hyperactivity on the path to physically violent
and nonviolent juvenile delinquency. Child
Development, 70, 1181-1196. doi: 10.1111/14678624.00086
Nicholson, B. C., Fox, R. A., & Johnson, S. D. (2005).
Parenting young children with challenging
behaviour. Infant and Child Development, 14,
425-428. doi: 10.1002/icd.403
Nix, R. L., Pinderhughes, E. E., Dodge, K. A., Bates, J. E.,
Pettit, G. S., & McFadyen-Ketchum, S. A. (1999).
The relation between mothers’ hostile attribution
tendencies and children’s externalizing behavior
problems: The mediating role of mothers’ harsh
discipline practices. Child Development, 70, 896909. doi: 10.1111/1467-8624.00065
152
Ogden, T., & Amlund-Hagen, K. (2008). Treatment
effectiveness of parent management training
in Norway: A randomized controlled trial of
children with conduct problems. Journal of
Consulting and Clinical Psychology, 76, 607-621.
doi: 10.1037/0022-006X.76.4.607
Oliver, P. H., Guerin, D. W., & Coffman, J. K. (2009).
Big five parental personality traits, parenting
behaviors, and adolescent behavior problems:
A mediation model. Personality and Individual
Differences, 47, 631-636. doi: 10.1016/j.
paid.2009.05.026
Otani, H., Libkuman, T. M., Widner, R. L., Jr., & Graves,
E. I. (2007). Memory for emotionally arousing
stimuli: A comparison of younger and older
adults. Journal of General Psychology, 134, 23-42.
doi: 10.3200/GENP.134.1
Otgaar, H., Peters, M., & Howe, M. L. (2012). Dividing
attention lowers children’s but increases
adults’ false memories. Journal of Experimental
Psychology: Learning, Memory, and Cognition, 38,
204-210. doi: 10.1037/a0025160
Oxford, M., Cavell, T. A., & Hughes, J. N. (2003).
Callous/unemotional traits moderate the
relation between ineffective parenting and child
externalizing problems: A partial replication
and extension. Journal of Clinical Child and
Adolescent Psychology, 32, 577-585. doi: 10.1207/
S15374424JCCP3204_10
Patterson, G. R. (1986). Performance models for
antisocial boys. American Psychologist, 41, 432444. doi: 10.1037/0003-066X.41.4.432
Patterson, G. R. (2005). The next generation of PMTO
models. The Behavior Therapist, 28, 25-32.
Patterson, G. R., Chamberlain, P., & Reid, J. B. (1982).
A comparative evaluation of a parent-training
program. Behavior Therapy, 13, 638-650. doi:
10.1016/S0005-7894(82)80021-X
References |
Patterson, G. R., DeBaryshe, B. D., & Ramsey, E. (1989).
A developmental perspective on antisocial
behavior. American Psychologist, 44, 329-335.
doi: 10.1037/0003-066X.44.2.329
Patterson, G. R., DeGarmo, D., & Forgatch, M.
S. (2004). Systematic changes in families
following prevention trials. Journal of
Abnormal Child Psychology, 32, 621-633. doi:
10.1023/B:JACP.0000047211.11826.54
Patterson, G. R., Dishion, T. J., & Yoerger, K.
(2000). Adolescent growth in new forms of
problem behavior: Macro- and micro-peer
dynamics. Prevention Science, 1, 3-13. doi:
10.1023/A:1010019915400
Patterson, G. R., Forgatch, M. S., & DeGarmo, D. S.
(2010). Cascading effects following intervention.
Development and Psychopathology, 22, 949-970.
doi: 10.1017/S0954579410000568
Patterson, G. R., Forgatch, M. S., Yoerger, K. L., &
Stoolmiller, M. (1998). Variables that initiate and
maintain an early-onset trajectory for juvenile
offending. Development and Psychopathology,
10, 531-547. doi: 10.1017/S0954579498001734
Patterson, G. R., Reid, J. B., & Eddy, J. M. (2002). A brief
history of the Oregon model. In J. B. Reid, G. R.
Patterson & J. Snyder (Eds.), Antisocial behavior
in children and adolescents: A developmental
analysis and model for intervention. (pp. 3-20).
Washington, DC US: American Psychological
Association.
Peterson, L., Tremblay, G., Ewigman, B., &
Popkey, C. (2002). The parental daily diary: A
sensitive measure of the process of change
in a child maltreatment prevention program.
Behavior Modification, 26, 627-647. doi:
10.1177/014544502236654
Reef, J., Diamantopoulou, S., van Meurs, I., Verhulst,
F. C., & van der Ende, J. (2011). Developmental
trajectories of child to adolescent externalizing
behavior and adult DSM-IV disorder: Results of
a 24-year longitudinal study. Social Psychiatry
and Psychiatric Epidemiology, 46, 1233-1241. doi:
10.1007/s00127-010-0297-9
Riggs, L., McQuiggan, D. A., Farb, N., Anderson, A. K.,
& Ryan, J. D. (2011). The role of overt attention in
emotion-modulated memory. Emotion, 11, 776785. doi: 10.1037/a0022591
Roediger, H. L., & McDermott, K. B. (1995). Creating
false memories: Remembering words not
presented in lists. Journal of Experimental
Psychology: Learning, Memory, and Cognition, 21,
803-814. doi: 10.1037/0278-7393.21.4.803
Roediger, H. L., Watson, J. M., McDermott, K. B.,
& Gallo, D. A. (2001). Factors that determine
false recall: A multiple regression analysis.
Psychonomic Bulletin & Review, 8, 385-407. doi:
10.3758/BF03196177
Romeo, R., Knapp, M., & Scott, S. (2006). Economic
cost of severe antisocial behaviour in children-And who pays it. British Journal of Psychiatry, 188,
547-553. doi: 10.1192/bjp.bp.104.007625
Rubin, D. B. (1986). Statistical matching using file
concatenation with adjusted weights and
multiple imputations. Journal of Business and
Economic Statistics, 4, 87-94.
Rubin, D. B. (1987). Multiple imputation for
nonresponse in surveys. New York: John Wiley.
Salekin, R. T., Debus, S. A., & Jackson, R. (2008).
Assessing child and adolescent psychopathy
Learning forensic assessment. (pp. 347-383). New
York, NY US: Routledge/Taylor & Francis Group.
Pickel, K. L., Narter, D. B., Jameson, M. M., & Lenhardt,
T. T. (2008). The weapon focus effect in child
eyewitnesses. Psychology, Crime & Law, 14, 61-72.
doi: 10.1080/10683160701391307
Sanders, M. R. (2008). Triple P-Positive Parenting
Program as a public health approach to
strengthening parenting. Journal of Family
Psychology, 22, 506-517. doi: 10.1037/08933200.22.3.506
Prior, M., Smart, D., Sanson, A., & Oberklaid, F. (1999).
Relationships between learning difficulties
and psychological problems in preadolescent
children from a longitudinal sample. Journal
of the American Academy of Child & Adolescent
Psychiatry, 38, 429-436. doi: 10.1097/00004583199904000-00016
Scarboro, M. E., & Forehand, R. (1975). Effects of
two types of response-contingent time-out
on compliance and oppositional behavior
of children. Journal of Experimental Child
Psychology, 19, 252-264. doi: 10.1016/00220965(75)90089-2
R Core Team. (2014). R: A language and environment
for statistical computing. R Foundation for
Statistical Computing, Vienna, Austria.
Schwarz, N. (1999). Self-reports: How the questions
shape the answers. American Psychologist, 54,
93-105. doi: 10.1037/0003-066x.54.2.93
153
R
| References
Scott, S., Knapp, M., Henderson, J., & Maughan, B.
(2001). Financial cost of social exclusion: Follow
up study of antisocial children into adulthood.
British Medical Journal, 323, 1-5. doi: 10.1136/
bmj.323.7306.191
Shaffer, D., Fisher, P., Lucas, C. P., Dulcan, M. K., &
Schwab-Stone, M. E. (2000). NIMH Diagnostic
Interview Schedule for Children Version IV (NIMH
DISC-IV): Description, differences from previous
versions, and reliability of some common
diagnoses. Journal of the American Academy
of Child & Adolescent Psychiatry, 39, 28-38. doi:
10.1097/00004583-200001000-00014
Sharma, V., & Sandhu, G. K. (2006). A community
study of association between parenting
dimensions and externalizing behaviors. Journal
of Indian Association for Child and Adolescent
Mental Health, 2, 48-58.
Shelleby, E. C., & Shaw, D. S. (2014). Outcomes of
parenting interventions for child conduct
problems: A review of differential effectiveness.
Child Psychiatry and Human Development, 45,
628-645. doi: 10.1007/s10578-013-0431-5
Shelton, K. K., Frick, P. J., & Wootton, J. (1996).
Assessment of parenting practices in families
of elementary school-age children. Journal
of Clinical Child Psychology, 25, 317-329. doi:
10.1207/s15374424jccp2503_8
Sibbald, B., & Roland, M. (1998). Understanding
controlled trials. Why are randomised controlled
trials important? British Medical Journal, 316, 201.
doi: 10.1136/bmj.316.7126.201
Siegel, D. J., & Bryson, T. P. (2014, October 21). You
said WHAT about time-outs? The Huffington
Post, retrieved from http://www.huffingtonpost.
com/daniel-j-siegel-md/time-outsoverused_b_6006332.html.
Siegel, D. J., & Bryson, T. P. (2014, September 23).
Time-outs are hurting your child. TIME.
Sigmarsdóttir, M., DeGarmo, D. S., Forgatch, M.
S., & Guðmundsdóttir, E. V. (2013). Treatment
effectiveness of PMTO for children’s behavior
problems in Iceland: Assessing parenting
practices in a randomized controlled trial.
Scandinavian Journal of Psychology, 54, 468-476.
doi: 10.1111/sjop.12078
Sigmarsdóttir, M., Thorlacius, Ö., Guðmundsdóttir,
E. V., & DeGarmo, D. S. (2014). Treatment
effectiveness of PMTO for children’s behavior
problems in iceland: Child outcomes in a
nationwide randomized controlled trial. Family
Process, 1-20. doi: 10.1111/famp.12109
154
Simonoff, E., Elander, J., Holmshaw, J., Pickles,
A., Murray, R., & Rutter, M. (2004). Predictors
of antisocial personality: Continuities from
childhood to adult life. British Journal of
Psychiatry, 184, 118-127. doi: 10.1192/
bjp.184.2.118
Snyder, J. (2002). Reinforcement and coercion
mechanisms in the development of antisocial
behavior: Peer relationships. In J. B. Reid, G. R.
Patterson & J. Snyder (Eds.), Antisocial behavior
in children and adolescents: A developmental
analysis and model for intervention. (pp. 101-122).
Washington, DC US: American Psychological
Association.
Snyder, J., Edwards, P., McGraw, K., Kilgore, K., &
Holton, A. (1994). Escalation and reinforcement
in mother-child conflict: Social processes
associated with the development of physical
aggression. Development and Psychopathology,
6, 305-321. doi: 10.1017/S0954579400004600
Snyder, J., & Patterson, G. R. (1995). Individual
differences in social aggression: A test of a
reinforcement model of socialization in the
natural environment. Behavior Therapy, 26, 371391. doi: 10.1016/S0005-7894(05)80111-X
Spijkers, W., Jansen, D. E., & Reijneveld, S. A. (2013).
Effectiveness of Primary Care Triple P on child
psychosocial problems in preventive child
healthcare: a randomized controlled trial. BMC
Medicine, 11, 240-247. doi: 10.1186/1741-701511-240
Steiger, J. H. (1980). Tests for comparing elements of
a correlation matrix. Psychological Bulletin, 87,
245-251. doi: 10.1037/0033-2909.87.2.245
Steinmetz, K. R. M., Addis, D. R., & Kensinger, E. A.
(2010). The effect of arousal on the emotional
memory network depends on valence.
NeuroImage, 53, 318-324. doi: 10.1016/j.
neuroimage.2010.06.015
Stevens, D., Charman, T., & Blair, R. J. R. (2001).
Recognition of emotion in facial expressions
and vocal tones in children with psychopathic
tendencies. The Journal of Genetic Psychology,
162, 201-211. doi: 10.1080/00221320109597961
Takarangi, M. K. T., Polaschek, D. L. L., Hignett, A.,
& Garry, M. (2008). Chronic and temporary
aggression causes hostile false memories for
ambiguous information. Applied Cognitive
Psychology, 22, 39-49. doi: 10.1002/acp.1327
Talmi, D., & Moscovitch, M. (2004). Can semantic
relatedness explain the enhancement of
memory for emotional words? Memory &
Cognition, 32, 742-751. doi: 10.3758/BF03195864
References |
Thijssen, J., Albrecht, G., Muris, P., & de Ruiter, C.
(2015a). Treatment fidelity as a determinant
of the effectiveness of Parent Management
Training - Oregon model in The Netherlands.
Manuscript submitted for publication.
Thijssen, J., Muris, P., & de Ruiter, C. (2015b). Initial
validation of the Dutch translation of the
Caregiver Wish List, an interview-based scale for
measuring parenting practices. Manuscript
submitted for publication.
Thijssen, J., Otgaar, H., Howe, M. L., & de Ruiter, C.
(2013). Emotional true and false memories
in children with callous-unemotional traits.
Cognition and Emotion, 27, 761-768. doi:
10.1080/02699931.2012.744300
Thijssen, J., Otgaar, H., Meijer, E. H., Smeets, T., & de
Ruiter, C. (2012). Emotional memory for central
and peripheral details in children with callousunemotional traits. Behavioral Sciences & the Law,
30, 506-515. doi: 10.1002/bsl.2021
Thijssen, J., Vink, G., Muris, P., & de Ruiter, C. (2015c).
The effectiveness of Parent Management
Training - Oregon model for children with
externalizing behavior problems in The
Netherlands. Manuscript accepted with revisions.
Tracey, T. J., & Kokotovic, A. M. (1989). Factor structure
of the Working Alliance Inventory. Psychological
Assessment, 1, 207-210. doi: 10.1037/10403590.1.3.207
Tremblay, R. E. (2000). The development of aggressive
behaviour during childhood: What have we
learned in the past century? International
Journal of Behavioral Development, 24, 129-141.
doi: 10.1080/016502500383232
van Buuren, S., Boshuizen, H. C., & Knook, D. L.
(1999). Multiple imputation of missing blood
pressure covariates in survival analysis.
Statistics in Medicine, 18, 681-694. doi: 10.1002/
(SICI)1097-0258(19990330)18:6<681::AIDSIM71>3.0.CO;2-R
van Buuren, S., & Groothuis-Oudshoorn, K. (2011).
Mice: Multivariate Imputation by Chained
Equations in R. Journal of Statistical Software, 45,
1-67.
van Leeuwen, E., & Albrecht, G. (2008). Parent
Management
Training
Oregon.
Ernstig
probleemgedrag neemt af met PMTO. Amsterdam:
PI Research.
Vannucci, M., Nocentini, A., Chiorri, C., & Menesini,
E. (2014). Brief report: Violent false memories
and engagement in aggressive and delinquent
behavior: An investigation in adolescents.
Journal of Adolescence, 37, 1329-1333. doi:
10.1016/j.adolescence.2014.09.007
Vannucci, M., Nocentini, A., Mazzoni, G., &
Menesini, E. (2012). Recalling unpresented
hostile words: False memories predictors of
traditional and cyberbullying. European Journal
of Developmental Psychology, 9, 182-194. doi:
10.1080/17405629.2011.646459
Verhulst, F. C., van der Ende, J., & Koot, H. M. (1996).
Handleiding voor de CBCL/4-18 [Maunal for the
CBCL/4-18]. Rotterdam: Department of child and
adolescent psychiatry, Erasmus Medical Centre/
Sophia.
Viding, E., Fontaine, N. M. G., Oliver, B. R., & Plomin,
R. (2009). Negative parental discipline,
conduct problems and callous-unemotional
traits: Monozygotic twin differences study.
British Journal of Psychiatry, 195, 414-419. doi:
10.1192/bjp.bp.108.061192
Viera, A. J., & Bangdiwala, S. I. (2007). Eliminating
bias in randomized controlled trials: Importance
of allocation concealment and masking. Family
Medicine, 39, 132-137.
Vitacco, M. J., Neumann, G. S., Ramos, V., & Roberts, M.
K. (2003a). Ineffective parenting: A precursor to
psychopathic traits and delinquency in Hispanic
females. Annals of the New York Academy of
Sciences, 1008, 300-303.
Vitacco, M. J., Rogers, R., & Neumann, C. S. (2003b).
The Antisocial Process Screening Device: An
examination of its construct and criterionrelated validity. Assessment, 10, 143-150. doi:
10.1177/1073191103010002005
Waller, R., Gardner, F., Hyde, L. W., Shaw, D. S., Dishion,
T. J., & Wilson, M. N. (2012). Do harsh and positive
parenting predict parent reports of deceitfulcallous behavior in early childhood? Journal of
Child Psychology and Psychiatry, 53, 946-953. doi:
10.1111/j.1469-7610.2012.02550.x
Walter, H. I., & Gilmore, S. K. (1973). Placebo versus
social learning effects in parent training
procedures designed to alter the behavior of
aggressive boys. Behavior Therapy, 4, 361-377.
doi: 10.1016/S0005-7894(73)80116-9
155
R
| References
Webster-Stratton, C. H., & Reid, M. J. (2010). The
incredible years program for children from
infancy to pre-adolescence: Prevention and
treatment of behavior problems. In R. C. Murrihy,
A. D. Kidman & T. H. Ollendick (Eds.), Clinical
handbook of assessing and treating conduct
problems in youth. (pp. 117-138). New York, NY,
US: Springer Science + Business Media.
Wechsler, D. (1991). Manual for the Wechsler
Intelligence Scale for Children - Third edition. San
Antonio: The Psychological Corporation.
Wechsler, D. (2002). Wechsler Preschool and Primary
Scale of Intelligence - Third edition. San Antonio:
The Psychological Corporation.
Williams, L. R., Degnan, K. A., Perez-Edgar, K. E.,
Henderson, H. A., Rubin, K. H., Pine, D. S., . . . Fox,
N. A. (2009). Impact of behavioral inhibition
and parenting style on internalizing and
externalizing problems from early childhood
through adolescence. Journal of Abnormal Child
Psychology, 37, 1063-1075. doi: 10.1007/s10802009-9331-3
Wiltz, N. A., & Patterson, G. R. (1974). An evaluation
of parent training procedures designed to alter
inappropriate aggressive behavior of boys.
Behavior Therapy, 5, 215-221. doi: 10.1016/
S0005-7894(74)80137-1
Wimmer, M. C., & Howe, M. L. (2009). The
development of automatic associative processes
and children’s false memories. Journal of
Experimental Child Psychology, 104, 447-465. doi:
10.1016/j.jecp.2009.07.006
156
Woodworth, M., & Waschbusch, D. (2008). Emotional
processing in children with conduct problems
and callous/unemotional traits. Child: Care,
Health and Development, 34, 234-244. doi:
10.1111/j.1365-2214.2007.00792.x
Wootton, J. M., Frick, P. J., Shelton, K. K., & Silverthorn,
P. (1997). Ineffective parenting and childhood
conduct problems: The moderating role of
callous-unemotional traits. Journal of Consulting
and Clinical Psychology, 65, 301-308. doi:
10.1037/0022-006X.65.2.292.b
Youngstrom, E., Loeber, R., & Stouthamer-Loeber,
M. (2000). Patterns and correlates of agreement
between parent, teacher, and male adolescent
ratings of externalizing and internalizing
problems. Journal of Consulting and Clinical
Psychology, 68, 1038-1050. doi: 10.1037/0022006x.68.6.1038
Zaslow, M. J., Weinfield, N. S., Gallagher, M., Hair, E. C.,
Ogawa, J. R., Egeland, B., . . . De Temple, J. M. (2006).
Longitudinal prediction of child outcomes from
differing measures of parenting in a low-income
sample. Developmental Psychology, 42, 27-37.
doi: 10.1037/0012-1649.42.1.27
Zlomke, K. R., Lamport, D., Bauman, S., Garland,
B., & Talbot, B. (2014). Parenting adolescents:
Examining the factor structure of the Alabama
Parenting Questionnaire for adolescents. Journal
of Child and Family Studies, 23, 1484-1490. doi:
10.1007/s10826-013-9803-5
Dankwoord
(acknowledgments)
D
Dankwoord |
Ik kan bijna niet geloven dat dit proefschrift er dan eindelijk is. Toch wel gek om na al
die jaren eindelijk op dit punt te zijn gekomen. Het was een lang en moeizaam project,
maar wat ben ik trots op het eindresultaat! Dit was me absoluut niet in mijn eentje
gelukt. Gelukkig heb ik hier de ruimte om iedereen te bedanken.
Corine, de belangrijkste bijdrage komt natuurlijk van jou. Je was nog maar een week in
dienst bij de UM, toen ik als derdejaarsstudent op jouw deur klopte om te vragen of je me
wilde begeleiden bij mijn bachelorthese. Toen was je al bezig met het implementeren
van PMTO in Nederland. Na mijn master gaf je me de mogelijkheid om de effectiviteit
van PMTO in Nederland als promovenda te onderzoeken. Een fantastische kans die ik
niet kon laten liggen. Dat het onderzoek zo moeizaam zou verlopen hadden we beiden
niet verwacht. Meerdere keren heb ik op het punt gestaan om de handdoek in de ring
te gooien. Steeds wist jij te voorkomen dat ik het opgaf. Je gaf me de gelegenheid om
extra studies uit te voeren en je hebt me de ruimte gegeven om klinische ervaring op te
doen. Uiteindelijk heb ik met hervonden enthousiasme en motivatie het project kunnen
afmaken. Bedankt dat je in mij en in het project bent blijven geloven.
Peter, jij werd in de loop van het promotietraject als tweede promotor bij mijn project
betrokken en daar ben ik altijd erg blij mee geweest. Je was altijd razendsnel met je
feedback en je keek met een zeer kritische blik naar mijn stukken. Hoewel ik dat ook
wel eens heb vervloekt, moet ik toegeven dat de stukken daar wel altijd beter van zijn
geworden. Vooral in de laatste fase van mijn proefschrift ben je van onschatbare waarde
geweest. Bedankt!
Ik ben veel dank verschuldigd aan de instellingen die hun medewerking hebben verleend
aan het onderzoek: De Bascule, Cardea Jeugdzorg, GGZ kinderen en Jeugd Rivierduinen,
Yorneo en Lindenhout. Jullie hebben veel geïnvesteerd om dit onderzoek mogelijk te
maken. Procedures moesten soms worden aangepast en personeel moest beschikbaar
worden gesteld. In tijden van bezuinigingen was het moeilijk om hier tijd en geld voor
vrij te maken, maar het is uiteindelijk gelukt! Bedankt voor jullie doorzettingsvermogen
en dank aan de bestuurders dat jullie naast de zorg voor cliënten, het belang en de
noodzaak van continue onderzoek steeds meer inzien.
Het uitzetten en vervolgens draaiende houden van onderzoek op meerdere locaties
in het land is een behoorlijke uitdaging en bijna onmogelijk om alleen te doen vanuit
Maastricht. Ik ben dan ook enorm blij dat we op iedere locatie een onderzoekscoördinator
hadden. Inès, Carlijn, Romana, Judith en Erna: ontzettend bedankt voor jullie inzet voor
het onderzoek! Het is voor jullie zeker niet gemakkelijk geweest om de belangen van
159
D
| Dankwoord
zowel Maastricht als jullie eigen instelling op één lijn te krijgen en te houden. Ik ben dan
ook enorm dankbaar dat jullie de moed niet hebben opgegeven. Ook jullie voorgangers
verdienen een dikke pluim.
Ik zou ook graag de gezinnen willen bedanken die aan het PMTO-onderzoek hebben
deelgenomen. Er is heel wat moed voor nodig om hulp te zoeken wanneer je kind
gedragsproblemen heeft. Op zo’n moment staat het water vaak al tot aan de lippen. Om
dan naast de behandeling ook nog deel te nemen aan een onderzoek, waarbij er ieder half
jaar behoorlijk wat vragenlijsten ingevuld moesten worden, is bewonderingswaardig.
Mijn dank aan alle ouders is groot.
I would like to thank my colleagues from the Atferdssenteret in Oslo and from the OSLC
in Oregon. Kristin Nordahl, thank you for teaching me the microsocial coding of the
Structured Interaction Tasks. Julie Stubbs, thank you for your help with improving our
Coder Impressions. Finally, thank you Marion Forgatch, for your interest, suggestions,
and support for our research over the years.
Gonnie, bedankt voor de jarenlange samenwerking vanuit PI Research/Kenniscentrum
PMTO en voor je hulp bij de treatment fidelity studie.
Ferko, jouw enorme enthousiasme tovert iedere keer weer een glimlach op mijn
gezicht. Bedankt voor al je steun aan mij en Corine de afgelopen jaren. Jij hebt altijd in
dit project geloofd en dat werkte aanstekelijk.
Gerko, de multipele geïmputeerde dataset van jou heeft de statistische analyses er niet
gemakkelijker op gemaakt voor mij. Gelukkig kon ik altijd bij je terecht met mijn vragen
en nam je de tijd voor me, zelfs als je een papa-dag had. Zo kon ik snel weer verder.
Bedankt hiervoor!
Lieve Linsey en Nele. Al sinds het begin zijn we kamergenootjes en door de tijd echte
vriendinnen geworden. We hebben samen lief en leed gedeeld met betrekking tot
ons promotieonderzoek, maar ook op privégebied. Ontzettend bedankt voor jullie
emotionele steun op beide fronten. Ik ben enorm trots dat jullie op deze belangrijke
dag mijn paranimfen zijn.
Peggy, hoewel je pas sinds een paar maanden ons nieuwe kamergenootje bent, ben je
een leuke aanwinst. Aangezien je recent jouw proefschrift heb afgerond, kon ik bij jou
terecht voor mijn vele vragen over het proefschrift en alle procedures. Bedankt dat je
de tijd voor mij nam.
160
Dankwoord |
Lieve (oud-)collega’s van de sectie Forensische Psychologie. Jullie hebben mijn AIO-tijd
extra aangenaam gemaakt. Bedankt voor jullie wetenschappelijke inzichten, leerzame
retraites en gezellige lunches. Ik ben dankbaar dat ik met zo’n deskundige, talentvolle
en vooral leuke groep heb mogen werken. Saskia, ook al werk je al een tijdje niet meer
aan de UM, het was supergezellig met jou als collega. Bedankt voor de vele, vooral nietwerk gerelateerde gesprekken.
Ook mijn vriendinnen ben ik dank verschuldigd. Lieve Coby’s, hoewel jullie niet direct
betrokken zijn geweest bij mijn promotietraject, hebben jullie wel voor de nodige
gezelligheid en vertier gezorgd. Hierdoor was ik in staat om mijn werk los te laten en
helemaal te ontspannen. Al vanaf het begin van de middelbare school is onze groep
gevormd en ik vind het fantastisch om jullie te hebben zien ontwikkelen van puberende
meiden naar feestende studenten tot uiteindelijk serieus werkende vrouwen en, in
sommige gevallen, moeders. Iedereen heeft zijn eigen pad gekozen, maar het contact
is gebleven. Ik kijk met veel plezier terug op de vele uitstapjes, vakanties, stedentripjes
en jaarlijkse vriendinnenweekendjes. Deze verzetjes zijn echt nodig geweest, vooral op
momenten dat ik het niet meer zag zitten met mijn project. Ik ben dankbaar voor onze
leuke club meiden (of moet ik dames zeggen, nu we de 30 zijn gepasseerd J). Dat we
dit nog lang mogen volhouden!
Pap en mam, bedankt voor jullie steun de afgelopen jaren. Hoewel jullie je regelmatig
afvroegen waarom ik nog doorging met het project, zijn jullie wel altijd achter mij
blijven staan. Jullie hebben me geleerd om af te maken waar je aan begint. Ook mijn
schoonouders wil ik bedanken voor hun interesse en betrokkenheid.
Allerliefste Guyon, bedankt voor al je steun, hulp en geduld! Alle hoogte- en
dieptepunten van het project heb ik met je gedeeld. Jij hielp me te relativeren wanneer
ik er doorheen zat en je vierde enthousiast ieder hoogtepunt mee. Ik mag me gelukkig
prijzen dat ik jou al zo vroeg in mijn leven ben tegengekomen. We zijn een goed team.
Ik hou heel veel van je.
Lieve Roan, wat heb jij mijn leven verrijkt! Iedere ochtend als ik jouw vrolijke koppie weer
zie, kan mijn dag niet meer stuk. Ik ben supertrots op jou! Liefste Alix, jij hebt ervoor
gezorgd dat ik de laatste maanden een eindsprint heb gemaakt. Mijn proefschrift moest
en zou voor jouw komst af zijn, zodat ik in mijn verlof alle rust zou hebben om van jou
te kunnen genieten. Ik ben blij dat het is gelukt. Samen met papa en jouw grote broer
maak je mij de gelukkigste vrouw op aarde.
161
D
Curriculum
Vitae
C
Curriculum vitae |
Jill Thijssen werd geboren op 10 januari 1985 te Deurne. In juli 2004 behaalde zij haar
vwo-diploma aan het Bouwens van der Boijecollege te Panningen. In september van dat
jaar begon zij aan haar studie Psychologie aan de Universiteit Maastricht en in augustus
2007 behaalt ze haar masterdiploma in Psychology & Law cum laude. Van september
2007 tot en met juni 2008 heeft zij als onderzoekmedewerker aan de Faculteit der
Psychologie en Neurowetenschappen van de Universiteit Maastricht gewerkt aan het
opzetten van de effectiviteitsstudie naar PMTO. Vanaf 1 juli 2008 werd zij aangesteld als
promovenda op de effectiviteitsstudie naar PMTO.
Jill Thijssen was born on January 10, 1985 in Deurne. In July 2004, she finished her
secondary school at the Bouwens van der Boijecollege in Panningen. In September of
that year, she started studying Psychology at Maastricht University and in August 2007
she received her master’s diploma in Psychology & Law, cum laude. From September
2007 until June 2008, she worked as a research-assistant at the Faculty of Psychology
and Neuroscience of Maastricht University. From July 2008, she was appointed at the
same Faculty as a PhD student on the effectiveness study of PMTO.
C
165
| Curriculum vitae
International journal articles
Duijster, D., de Jong-Lenters, M., de Ruiter, C., Thijssen, J., van Loveren, C., & Verrips, G.
H. W. (2015). Parental and family-related influences on dental caries in children of
Dutch, Moroccan and Turkish origin. Community Dentistry and Oral Epidemiology, 43,
152-162. doi: 10.1111/cdoe.12134
de Jong-Lenters, M., Duijster, D., Bruist, M. A., Thijssen, J., & de Ruiter, C. (2014).
The relationship between parenting, family interaction and childhood dental
caries: A case-control study. Social Science & Medicine, 116, 49-55. doi: 10.1016/j.
socscimed.2014.06.031
Thijssen, J., Otgaar, H., Howe, M.L., & de Ruiter, C. (2013). Emotional true and false
memories in children with callous-unemotional traits. Cognition & Emotion, 27, 761768. doi: 10.1080/02699931.2012.744300.
Thijssen, J., Otgaar, H., Meijer, E.H., Smeets, T., & de Ruiter, C. (2012). Emotional memory
for central and peripheral details in children with callous-unemotional traits.
Behavioral Sciences and the Law, 30, 506-515. doi: 10.1002/bsl.2021
Thijssen, J., & de Ruiter, C. (2011). Identifying subtypes of spousal assaulters
using the B-SAFER. Journal of Interpersonal Violence, 26, 1307-1321. doi:
10.1177/0886260510369129
Thijssen, J., & de Ruiter, C. (2011). Instrumental and expressive violence in Belgian
homicide perpetrators. Journal of Investigative Psychology and Offender Profiling, 8,
58-73. doi: 10.1002/jip.130
Dutch journal articles
Thijssen, J., & de Ruiter, C. (2010). De relatie tussen subtypen relationeel geweldplegers
en de adviezen van de reclassering. PROCES, 89, 416-428.
Thijssen, J., & de Ruiter, C. (2010). Strenge bewaking van kwaliteit bij PMTO. Kennis, 4,
8-14.
Book chapter
Thijssen, J., de Ruiter, C., & Albrecht, G. (2008). Preventie van antisociaal gedrag bij
kinderen: Parent Management Training Oregon. [Prevention of antisocial behavior
in children: Parent Management Training Oregon]. In J.R.M. Gerris en R.C.M.E. Engels
(Eds.), Vernieuwingen in jeugd en gezin: Beleidsvisies, gezinsrelaties en interventies (p.
125 - 140). Assen: Van Gorcum.
166
Curriculum vitae |
Submitted articles
Thijssen, J., Vink, G., Muris, P., & de Ruiter, C. (2016). The effectiveness of Parent
Management Training - Oregon Model in clinically referred children with externalizing
behavior problems in The Netherlands. Accepted with revisions
Thijssen, J., Albrecht, G., Muris, P., & de Ruiter, C. (2016). Treatment fidelity as a
determinant of the effectiveness of Parent Management Training-Oregon model in
The Netherlands. Submitted for publication
Thijssen, J., Broers, N., Muris, P., & de Ruiter, C. (2016). Initial validation of the Dutch
translation of the Caregiver Wish List, an interview-based scale for measuring
parenting practices. Submitted for publication
Conference presentations
Thijssen, J., Vink, G., Muris, P., & de Ruiter, C. (2015). The effectiveness of PMTO in The
Netherlands. Paper presented at the Association for Psychological Science, New York,
United States.
Thijssen, J., & de Ruiter, C. (2013). PMTO. Een behandeling voor kinderen met
externaliserende gedragsproblemen. Paper presented at the ‘Forensic Psychology
Update 2.0’, Maastricht, The Netherlands.
Thijssen, J., de Ruiter, C., & Muris, P. (2013). Associations between parenting stress and
self-rated and observed parenting behavior in parents of children with externalizing
behavior problems. Poster presented at the Association for Psychological Science,
Washington DC, United States.
Thijssen, J., Otgaar, H., Meijer, E.H., Smeets, T., Howe, M.L., & de Ruiter, C. (2012). Emotional
memory in children with callous-unemotional traits. Paper presented at the European
Conference on Psychology and Law, Nicosia, Cyprus.
Thijssen, J., Otgaar, H., Meijer, E.H., Smeets, T., & de Ruiter, C. (2011). Emotioneel geheugen
in kinderen met psychopathische trekken. Paper presented at ‘Forensic Psychology
Update’, Maastricht, The Netherlands.
Thijssen, J., & de Ruiter, C. (2010). The predictive validity of the B-SAFER for spousal assault.
Paper presented at the American Psychology-Law Society, Vancouver, Canada.
Van Leeuwen, E., & Thijssen, J. (2009). Preventie van antisociaal gedrag bij kinderen:
Parent Management Training Oregon (PMTO). [Prevention of antisocial behavior of
children: Parent Management Training Oregon (PMTO)]. Poster presented at ‘Jeugd
in Onderzoek’, Nieuwegein, The Netherlands.
Thijssen, J., & de Ruiter, C. (2008). Risk assessment of spousal assaulters using the B-SAFER.
Paper presented at the European Conference on Psychology and Law, Maastricht,
The Netherlands.
167
C
In a poor and infertile
environment it is
difficult to develop,
but within the shadow
of the parent’s love and
attention, a child is able
to bloom and grow to
its full potential.